CN114529143A - Method and device for recommending outbound clues and electronic equipment - Google Patents
Method and device for recommending outbound clues and electronic equipment Download PDFInfo
- Publication number
- CN114529143A CN114529143A CN202210016824.2A CN202210016824A CN114529143A CN 114529143 A CN114529143 A CN 114529143A CN 202210016824 A CN202210016824 A CN 202210016824A CN 114529143 A CN114529143 A CN 114529143A
- Authority
- CN
- China
- Prior art keywords
- outbound
- data
- thread
- cue
- seat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000003795 chemical substances by application Substances 0.000 claims abstract description 99
- 238000012545 processing Methods 0.000 claims abstract description 99
- 238000006243 chemical reaction Methods 0.000 claims abstract description 86
- 239000013598 vector Substances 0.000 claims abstract description 75
- 238000004891 communication Methods 0.000 claims abstract description 59
- 238000003860 storage Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 10
- 238000010586 diagram Methods 0.000 description 10
- 238000009826 distribution Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 230000006978 adaptation Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Educational Administration (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a method and a device for recommending an outbound cue, electronic equipment and a computer-readable storage medium, and relates to the technical field of data processing. The method carries out cross processing on at least two kinds of data in the characteristic data of the outbound call seat, the characteristic data of the outbound call clue and the communication data of the outbound call seat and the outbound call clue to generate outbound call cross characteristic vector data; inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat; and obtaining an outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound thread corresponding to each outbound agent, and recommending the outbound thread based on the outbound thread recommendation queue corresponding to each outbound agent. The embodiment can improve the efficiency of outbound calls and the conversion rate of orders.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method and an apparatus for recommending outbound hint, an electronic device, and a computer-readable storage medium.
Background
With the development of the mobile internet, various new sales modes are produced, and the sales forms are gradually on-line. The telephone outbound system is an indispensable important component in many complex product services, such as insurance services, financial loan services, financial investment services, real estate services, automobile services, curriculum services, travel route services, event invitations, and the like. In the face of a scene that the number of outbound seats and outbound threads is one to many, how to improve the efficiency of the outbound seat management and the calling of the outbound threads becomes a technical problem to be solved urgently.
Disclosure of Invention
In view of the above problems, the present application is proposed to provide a method and an apparatus for recommending outbound clues, an electronic device and a computer-readable storage medium, which overcome the above problems or at least partially solve the above problems, and can improve the efficiency of outbound calls and the order conversion rate. The technical scheme is as follows:
in a first aspect, a method for recommending outbound clues is provided, which includes:
acquiring feature data of the outbound seat, feature data of the outbound clue and communication data of the outbound seat and the outbound clue;
performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound clue and the communication data of the outbound seat and the outbound clue to generate outbound cross feature vector data;
inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat;
and obtaining an outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound thread corresponding to each outbound agent, and recommending the outbound thread based on the outbound thread recommendation queue corresponding to each outbound agent.
In a possible implementation manner, the recommending an outbound thread based on the outbound thread recommending queue corresponding to each outbound agent includes:
allocating corresponding outbound threads to each outbound agent based on the outbound thread recommendation queue corresponding to each outbound agent; or
And providing the outbound cue recommendation queue corresponding to each outbound cue for the corresponding outbound cue, and sequentially communicating with the outbound cue by the corresponding outbound cue according to the outbound cue recommendation queue.
In a possible implementation manner, obtaining the outbound thread referral queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound agent corresponding to each outbound thread includes:
determining the highest predicted order conversion rate corresponding to each outbound clue according to the predicted order conversion rate of each outbound clue corresponding to each outbound agent;
taking the outbound seat corresponding to the highest predicted order conversion rate corresponding to each outbound thread as a recommended outbound seat corresponding to each outbound thread;
and counting the outbound clues corresponding to the outbound agents according to the recommended outbound clues corresponding to the outbound clues to obtain an outbound clue recommendation queue corresponding to each outbound agent.
In one possible implementation, the multi-modal outbound cue recommendation model is trained by;
constructing an initial multi-mode outbound cue recommendation model;
collecting sample characteristic data of the outbound agents, sample characteristic data of the outbound clues, sample communication data of the outbound agents and the outbound clues and historical order conversion rates of the outbound agents corresponding to the outbound clues;
and training the initial multi-mode outbound cue recommendation model based on the sample characteristic data of the outbound cue, the sample communication data of the outbound cue and the historical order conversion rate of each outbound cue corresponding to each outbound cue to obtain the trained multi-mode outbound cue recommendation model.
In a possible implementation manner, training the initial multi-mode outbound cue recommendation model based on the sample feature data of the outbound seats, the sample feature data of the outbound cues, the sample communication data of the outbound seats and the outbound cues, and the historical order conversion rate of each outbound cue corresponding to each outbound seat to obtain a trained multi-mode outbound cue recommendation model includes:
performing cross processing on at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound clue and the sample communication data of the outbound seat and the outbound clue to generate outbound cross sample characteristic vector data;
and taking the characteristic vector data of the outbound cross sample as input, taking the historical order conversion rate of each outbound cue corresponding to each outbound seat as output, and training the initial multi-mode outbound cue recommendation model to obtain the trained multi-mode outbound cue recommendation model.
In a possible implementation manner, performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound thread, and the communication data between the outbound seat and the outbound thread to generate outbound cross feature vector data includes:
selecting at least two kinds of data in the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue, and constructing a vector for each kind of data;
performing feature level cross processing on the data of at least two vectors to obtain a first processing result;
performing element-level cross processing on the data of at least two vectors to obtain a second processing result;
combining the data of at least two vectors pairwise, and performing low-order cross processing to obtain a third processing result;
and combining the first processing result, the second processing result and the third processing result, and performing transformation processing by using a preset function to generate outbound cross feature vector data.
In a second aspect, an apparatus for recommending outbound clues is provided, comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring the characteristic data of an outbound agent, the characteristic data of an outbound thread and the communication data of the outbound agent and the outbound thread;
the generating module is used for performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound clue and the communication data of the outbound seat and the outbound clue to generate outbound cross feature vector data;
the prediction module is used for inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat;
and the recommending module is used for obtaining the outbound cue recommending queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound cue corresponding to each outbound agent, and recommending the outbound cue based on the outbound cue recommending queue corresponding to each outbound agent.
In one possible implementation, the recommendation module is further configured to:
allocating corresponding outbound threads to each outbound agent based on the outbound thread recommendation queue corresponding to each outbound agent; or
And providing the outbound cue recommendation queue corresponding to each outbound cue for the corresponding outbound cue, and sequentially communicating with the outbound cue by the corresponding outbound cue according to the outbound cue recommendation queue.
In one possible implementation, the recommendation module is further configured to:
determining the highest predicted order conversion rate corresponding to each outbound cue according to the predicted order conversion rate of each outbound cue corresponding to each outbound cue;
taking the outbound seat corresponding to the highest predicted order conversion rate corresponding to each outbound thread as a recommended outbound seat corresponding to each outbound thread;
and counting the outbound clues corresponding to the outbound agents according to the recommended outbound clues corresponding to the outbound clues to obtain an outbound clue recommendation queue corresponding to each outbound agent.
In one possible implementation, the apparatus further includes a training module configured to:
constructing an initial multi-mode outbound cue recommendation model;
collecting sample characteristic data of the outbound agents, sample characteristic data of the outbound clues, sample communication data of the outbound agents and the outbound clues and historical order conversion rates of the outbound agents corresponding to the outbound clues;
and training the initial multi-mode outbound cue recommendation model based on the sample characteristic data of the outbound cue, the sample communication data of the outbound cue and the historical order conversion rate of each outbound cue corresponding to each outbound cue to obtain the trained multi-mode outbound cue recommendation model.
In one possible implementation, the training module is further configured to:
performing cross processing on at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound clue and the sample communication data of the outbound seat and the outbound clue to generate outbound cross sample characteristic vector data;
and taking the characteristic vector data of the outbound cross sample as input, taking the historical order conversion rate of each outbound cue corresponding to each outbound seat as output, and training the initial multi-mode outbound cue recommendation model to obtain the trained multi-mode outbound cue recommendation model.
In one possible implementation, the generating module is further configured to:
selecting at least two kinds of data in the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue, and constructing a vector for each kind of data;
performing feature level cross processing on the data of at least two vectors to obtain a first processing result;
performing element-level cross processing on the data of at least two vectors to obtain a second processing result;
combining the data of at least two vectors pairwise, and performing low-order cross processing to obtain a third processing result;
and combining the first processing result, the second processing result and the third processing result, and performing transformation processing by using a preset function to generate outbound cross feature vector data.
In a third aspect, an electronic device is provided, which comprises a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to perform the method for recommending an outbound thread according to any of the above.
In a fourth aspect, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the method for recommending an outbound cue when the computer program runs.
By means of the technical scheme, the method and the device for recommending the outbound cue, the electronic device and the computer readable storage medium provided by the embodiment of the application can acquire the feature data of the outbound seat, the feature data of the outbound cue and the communication data of the outbound seat and the outbound cue; performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound clue and the communication data of the outbound seat and the outbound clue to generate outbound cross feature vector data; inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat; and then obtaining an outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound thread corresponding to each outbound agent, and recommending the outbound threads based on the outbound thread recommendation queue corresponding to each outbound agent. It can be seen that the embodiment of the application can recommend the outbound clues suitable for the outbound agents by combining with some characteristics of the outbound agents, thereby forming personalized recommendations for each outbound agent, improving the outbound efficiency and the order conversion rate, and further improving the efficiency of the whole telephone outbound system.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments of the present application will be briefly described below.
Fig. 1 is a system configuration diagram showing outbound call thread distribution in the related art;
FIG. 2 is a flow chart illustrating a method for recommending outbound hints provided by an embodiment of the present application;
FIG. 3 is a system block diagram illustrating outbound thread distribution in an embodiment of the present application;
FIG. 4 is an architecture diagram illustrating outbound cable distribution in an embodiment of the present application;
FIG. 5 is a block diagram of a referrer for outbound cues provided by an embodiment of the application;
FIG. 6 is a block diagram of a referral device for outbound cues provided by another embodiment of the present application;
FIG. 7 shows a block diagram of an electronic device according to an embodiment of the application.
Detailed Description
Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that such uses are interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the term "include" and its variants are to be read as open-ended terms meaning "including, but not limited to.
Fig. 1 shows a system structure diagram of outbound thread distribution in the related art, in fig. 1, a retained thread, a newly added thread, and a secondary reach thread are accessed to a CRM (Customer Relationship Management) system, and for two different timeliness threads of a real-time thread and a yesterday thread, three different allocation modes are set, that is, number assignment, card assignment, and operation assignment are set, and an outbound seat acquires an outbound thread under the restriction of the three allocation modes, and then processes of thread dialing, thread connection, and thread conversion are performed. The number taking distribution is that an outbound agent sends a number taking request and acquires an outbound clue from a real-time clue or a yesterday clue queue; the card punching distribution is that after the outbound agent works and punches the card, an outbound clue is obtained from a real-time clue or a yesterday clue queue; the operation allocation is that an operator allocates an outbound thread queue to each outbound seat, so that each outbound seat communicates with the outbound thread according to the outbound thread queue. The related art has the following problems in the whole process:
(1) aiming at the channel with large release amount, the situation that the calling agent cannot be completely dialed exists, so that the waste of calling clues is caused;
(2) the calling-out clues suitable for the calling-out seat cannot be distributed by combining with some characteristics of the calling-out seat, and the efficiency of the whole calling-out and the order conversion rate cannot be improved well.
In order to solve the above technical problem, an embodiment of the present application provides a method for recommending an outbound thread, which models an outbound seat and an outbound thread and can form a personalized recommendation for each outbound seat. As shown in fig. 2, the method for recommending an outbound call cue may be applied to an electronic device such as a server, a personal computer, a smart phone, a tablet computer, and a smart watch, and specifically may include the following steps S201 to S204:
step S201, acquiring feature data of the outbound seat, feature data of the outbound clue and communication data of the outbound seat and the outbound clue;
step S202, at least two kinds of data of the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue are processed in a cross mode to generate outbound cross characteristic vector data;
step S203, inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat;
and step S204, obtaining an outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound agent corresponding to each outbound thread, and recommending the outbound thread based on the outbound thread recommendation queue corresponding to each outbound agent.
The method and the device can acquire the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue; performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound clue and the communication data of the outbound seat and the outbound clue to generate outbound cross feature vector data; then, inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat; and then obtaining an outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound thread corresponding to each outbound agent, and recommending the outbound threads based on the outbound thread recommendation queue corresponding to each outbound agent. It can be seen that the embodiment of the application can recommend the outbound clues suitable for the outbound seats by combining with some characteristics of the outbound seats, so that personalized recommendation is formed for each outbound seat, the outbound efficiency and the order conversion rate can be improved, and the efficiency of the whole telephone outbound system is further improved.
The feature data of the outbound seat mentioned in step S201 above may be some feature data of the outbound seat itself, such as age, work experience, education experience, and the like, and may also be, for example, the number of outbound calls, the number of outbound threads, outbound completion order data, and the like, which is not limited in this embodiment of the application.
The outbound call thread may be a user who may have a need for insurance services, real estate services, automobile services, courseware services, and the like. The characteristic data of the outbound cue may be the age, work experience, education experience, social income, family condition, etc. of the outbound cue, and may also be the configuration condition of the outbound corresponding service, etc., which is not limited in the embodiment of the present application.
The communication data of the outbound seat and the outbound thread may be call voice data, call text data, instant messaging text data, and the like, which is not limited in the embodiment of the present application.
In step S202, at least two kinds of data of the feature data of the outbound seat, the feature data of the outbound thread, and the communication data between the outbound seat and the outbound thread are processed in a cross manner, which may be cross processing the feature data of the outbound seat and the feature data of the outbound thread, cross processing the feature data of the outbound seat and the communication data between the outbound seat and the outbound thread, cross processing the feature data of the outbound thread and the communication data between the outbound seat and the outbound thread, or cross processing the feature data of the outbound seat, the feature data of the outbound thread, and the communication data between the outbound seat and the outbound thread, so as to generate outbound cross feature vector data.
One possible implementation is provided in the embodiments of the present application, and the multi-modal outbound cue recommendation model may be trained through the following steps a1 to A3:
step A1, constructing an initial multi-mode outbound thread recommendation model;
step A2, collecting sample characteristic data of the outbound seats, sample characteristic data of the outbound clues, sample communication data of the outbound seats and the outbound clues and historical order conversion rates of the outbound seats corresponding to the outbound clues;
and step A3, training the initial multi-mode outbound thread recommendation model based on the sample characteristic data of the outbound seats, the sample characteristic data of the outbound threads, the sample communication data of the outbound seats and the outbound threads and the historical order conversion rate of each outbound seat corresponding to each outbound thread to obtain the trained multi-mode outbound thread recommendation model.
In step a1, the initial multi-modal outbound cue recommendation model may be constructed based on xdeepfm (extremely Deep factorization model), where xdeepfm integrates two modules, namely CIN (Compressed Interaction Network) and DNN (Deep Neural Networks), to help the model learn high-order feature interactions in an explicit and implicit manner, and the integrated linear module and Deep Neural module also make the model have learning capabilities of both memory and generalization. To improve the versatility of the model, different modules in xdepfm share the same input data. In a specific application scenario, different modules may also access different respective input data, for example, a linear module may still access many cross features extracted according to prior knowledge to improve memory capability, and in CIN or DNN, in order to reduce the computational complexity of the model, only a part of sparse feature subsets may be imported.
Corresponding to the feature data of the outbound seat, the feature data of the outbound cue and the communication data of the outbound seat and the outbound cue, the sample feature data of the outbound seat may be some feature data of the outbound seat itself, such as age, work experience, education experience, and the like, and may also be outbound times, outbound cue number, outbound completion order data, and the like, which is not limited in the embodiment of the present application.
The outbound call thread may be a user who may have a need for insurance services, real estate services, automobile services, courseware services, and the like. The sample characteristic data of the outbound cue may be an age, a work experience, an educational experience, a social income, a family condition, etc. of the outbound cue, and may also be a configuration condition of the outbound corresponding service, etc., which is not limited in the embodiment of the present application.
The sample communication data of the outbound seat and the outbound thread may be call voice data, call text data, instant messaging text data, and the like, which is not limited in the embodiment of the present application.
In the embodiment of the present application, a possible implementation manner is provided, in the step A3, based on sample feature data of the outbound seats, sample feature data of the outbound threads, sample communication data of the outbound seats and the outbound threads, and historical order conversion rates of each outbound seat corresponding to each outbound thread, the initial multi-modal outbound thread recommendation model is trained to obtain a trained multi-modal outbound thread recommendation model, which specifically includes the following steps A3-1 and A3-2:
a3-1, performing cross processing on at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound clue and the sample communication data of the outbound seat and the outbound clue to generate outbound cross sample characteristic vector data;
it can be understood that, at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound thread and the sample communication data of the outbound seat and the outbound thread are processed in a cross manner, the sample characteristic data of the outbound seat and the sample characteristic data of the outbound thread may be processed in a cross manner, the sample characteristic data of the outbound seat and the sample communication data of the outbound seat and the outbound thread may be processed in a cross manner, the sample characteristic data of the outbound thread and the sample communication data of the outbound seat and the outbound thread may be processed in a cross manner, and the sample characteristic data of the outbound seat, the sample characteristic data of the outbound thread and the sample communication data of the outbound seat and the outbound thread may be processed in a cross manner, so as to generate the outbound cross sample characteristic vector data.
And step A3-2, taking the characteristic vector data of the outbound cross sample as input, taking the historical order conversion rate of each outbound cue corresponding to each outbound seat as output, and training the initial multi-mode outbound cue recommendation model to obtain the trained multi-mode outbound cue recommendation model.
According to the method and the device, the adaptation degree of the outbound seats and the outbound clues is continuously optimized through adjustment of the model parameters, so that personalized recommendations are formed for each outbound seat, and the outbound efficiency and the order conversion rate can be improved.
In the embodiment of the present application, a possible implementation manner is provided, where the step S204 obtains the outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound agent corresponding to each outbound thread, and specifically may include the following steps B1 to B3:
step B1, determining the highest predicted order conversion rate corresponding to each outbound cue according to the predicted order conversion rate corresponding to each outbound cue of each outbound cue;
step B2, taking the outbound seat corresponding to the highest predicted order conversion rate corresponding to each outbound thread as the recommended outbound seat corresponding to each outbound thread;
and step B3, counting the outbound clues corresponding to the outbound clues according to the recommended outbound clues corresponding to the outbound clues to obtain an outbound clue recommendation queue corresponding to each outbound clue.
For example, there are A, B, C, D and the like as outbound agents, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and the like as outbound threads, the predicted order conversion rate of each outbound agent A, B, C, D corresponding to outbound thread 1 is 0.5, 0.6, 0.4, 0.7, the predicted order conversion rate of each outbound agent A, B, C, D corresponding to outbound thread 2 is 0.4, 0.2, 0.5, 0.3, the predicted order conversion rate of each outbound agent A, B, C, D corresponding to outbound thread 3 is 0.2, 0.4, 0.6, 0.3 and the like, the highest predicted order conversion rate corresponding to each outbound thread may be determined, for example, the highest predicted order conversion rate corresponding to outbound thread 1 is 0.7, the highest predicted order conversion rate corresponding to outbound thread 2 is 0.5, the highest predicted order conversion rate corresponding to outbound thread 3 is 0.6, and the like, and the highest predicted order conversion rate corresponding to each outbound thread may be 0.6, serving as a recommended outbound seat corresponding to each outbound thread; and then, according to the recommended outbound cue corresponding to each outbound cue, counting the outbound cues corresponding to each outbound cue to obtain an outbound cue recommendation queue corresponding to each outbound cue, that is, outbound cue D calls outbound cue 1, outbound cue C calls outbound cue 2 and call cue 3, it should be noted that this is only an example and does not limit the embodiments of the present application.
In the embodiment of the present application, a possible implementation manner is provided, in the step S204, the external call thread is recommended based on the external call thread recommendation queue corresponding to each external call seat, specifically, the corresponding external call thread is allocated to each external call seat based on the external call thread recommendation queue corresponding to each external call seat; or the outbound cue recommendation queue corresponding to each outbound cue is provided for the corresponding outbound cue, and the corresponding outbound cue is communicated with the outbound cue in sequence according to the outbound cue recommendation queue. The method and the device can recommend the outbound clues suitable for the outbound agents by combining with some characteristics of the outbound agents, so that personalized recommendation is formed for each outbound agent, the outbound efficiency and the order conversion rate can be improved, and the efficiency of the whole telephone outbound system is improved.
In the embodiment of the present application, a possible implementation manner is provided, in the above step S202, cross-processing is performed on at least two kinds of data in feature data of an outbound agent, feature data of an outbound thread, and communication data of the outbound agent and the outbound thread to generate outbound cross-feature vector data, specifically, cross-processing may also be performed by using xdeepfm, and the method may include the following steps C1 to C5:
step C1, selecting at least two kinds of data of the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue, and constructing a vector for each kind of data;
step C2, performing feature level cross processing on the data of at least two vectors to obtain a first processing result;
step C3, performing element-level cross processing on the data of at least two vectors to obtain a second processing result;
step C4, combining the data of at least two vectors pairwise, and performing low-order cross processing to obtain a third processing result;
and step C5, combining the first processing result, the second processing result and the third processing result, and performing transformation processing by using a preset function to generate the outbound cross feature vector data.
The embodiment of the application can make full use of the relation between data, extracts more recessive features, gives consideration to high-order and low-order processing, makes the data more fully utilized, obtains more accurate prediction results later, and meets the requirements of practical application scenes.
In practical application, the feature data of the outbound agent, the feature data of the outbound thread and the communication data of the outbound agent and the outbound thread are different fields, and the processing method is to embed the three fields into vectors with the same latitude. For example, the feature vectors of two fields are (a1, a2, a3) and (b1, b2, b3), and the specific intersection processing includes:
(1) performing feature level intersection between data of different filtered, namely performing Hadamard product on all elements between vectors, and then performing convolution transformation under a certain weight, f (w (a1 b1, a2 b2, a3 b 3));
(2) all filtered vector data are crossed in element level, namely after Hadamard products are made on each element between vectors, different weight values are given to the result after each product, and then linear transformation is carried out, wherein f (w1 a1 b1, w2 a2 b2 and w3 a3 b 3);
(3) combining all the data in pairs, and performing low-order cross processing f (w (a1, a2, a3, b1, b2, b 3));
combining the three processing results together, and performing transformation processing by using a preset function to obtain the outbound cross feature vector data. The preset function can be set according to actual situations, and the embodiment does not limit this. It should be noted that the above examples are merely illustrative and do not limit the embodiments of the present application.
In the above, various implementation manners of each link in the embodiment shown in fig. 2 are introduced, and the method for recommending an outbound hint provided by the embodiment of the present application is further described through a specific embodiment.
For a company, a high-quality customer is valuable, time for sales staff is also valuable, and it is particularly important to improve the order conversion rate of each sales staff and reduce unnecessary time waste. In order to solve the existing problems, a recommendation system taking salesmen as a core is built, and optimal customers of the salesmen are distributed. The salesperson can call out the seat, and the customer can be a call-out thread.
The method is improved on the basis of the original system, under the condition of adapting to the original CRM system, the data transfer mode is adjusted, a recommended link is added, and the problems in the system are solved in the link. Fig. 3 shows a system structure diagram of outbound thread distribution in the embodiment of the present application, a recommendation system is added in fig. 3, and the recommendation system performs the above steps S201 to S204 to recommend outbound threads based on an outbound thread recommendation queue corresponding to each outbound seat, and further can be divided into an outbound seat corresponding to a real-time outbound thread and an outbound seat corresponding to a non-real-time outbound thread, and then accesses a CRM system to perform real-time recommendation. Here, the recommendation system interfaces to the CRM system:
(1) classifying different channel users to enter a recommendation model based on big data calculation processing;
(2) and (4) combining sales data of the outbound agents, accessing data flowing out of the recommendation system into the CRM system, and performing real-time recommendation.
Optimizing sale adaptation user recommendation algorithm:
(1) periodically acquiring statistical data and text data of sales personnel, analyzing the part of speech, making a feature vector and putting the feature vector into a recommendation model;
it can be understood that, based on the text data and the statistical data, algorithms such as NLP (Natural Language Processing) and bayes can be used to efficiently implement the portrayal of the sales portrait.
(2) Processing more user data, and performing cross processing with salesman data;
(3) and recommending the parameters and the structure of the model for optimization.
Fig. 4 shows an architecture diagram of outbound call thread distribution in the embodiment of the present application, in fig. 4, the application may include a CRM system, the application layer may include call text recall, sales portrait vector stitching, user portrait vector stitching, xdeepfm model prediction, thread rearrangement distribution, and the data layer may include sales data, user data, singleton data, call text data, and the like.
On the basis of the existing system framework, the embodiment of the application improves the data processing and using efficiency by asynchronous processing of mass text data, sales portrait data and user data and transformation of a data flow process, introduces multidimensional data such as sales behaviors and habits, models clues in a sales angle, rearranges and distributes the clues, and greatly improves the utilization rate of the clues; through the adjustment of model parameters, the degree of adaptation of sales and users is continuously optimized, and the net sale yield and the APL (Average Product of Labor, Average Labor yield) are improved on the service result.
It should be noted that, in practical applications, all the possible embodiments described above may be combined in a combined manner at will to form possible embodiments of the present application, and details are not described here again.
Based on the method for recommending the outbound clues provided by the embodiments, the embodiment of the application also provides a device for recommending the outbound clues based on the same inventive concept.
Fig. 5 is a block diagram of a device for recommending an outbound call thread according to an embodiment of the present application. As shown in fig. 5, the apparatus for recommending an outbound thread may include an obtaining module 510, a generating module 520, a predicting module 530, and a recommending module 540.
An obtaining module 510, configured to obtain feature data of the outbound seat, feature data of the outbound thread, and communication data between the outbound seat and the outbound thread;
a generating module 520, configured to perform cross processing on at least two types of data in the feature data of the outbound seat, the feature data of the outbound thread, and the communication data between the outbound seat and the outbound thread, so as to generate outbound cross feature vector data;
the prediction module 530 is used for inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat;
and the recommending module 540 is configured to obtain the outbound thread recommending queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound agent corresponding to each outbound thread, and recommend the outbound thread based on the outbound thread recommending queue corresponding to each outbound agent.
In the embodiment of the present application, a possible implementation manner is provided, and the recommending module 540 shown in fig. 5 is further configured to:
allocating corresponding outbound threads to each outbound agent based on the outbound thread recommendation queue corresponding to each outbound agent; or
And providing the outbound thread recommendation queue corresponding to each outbound seat for the corresponding outbound seat, and sequentially communicating with the outbound thread by the corresponding outbound seat according to the outbound thread recommendation queue.
In the embodiment of the present application, a possible implementation manner is provided, and the recommending module 540 shown in fig. 5 is further configured to:
determining the highest predicted order conversion rate corresponding to each outbound clue according to the predicted order conversion rate of each outbound clue corresponding to each outbound agent;
taking the outbound seat corresponding to the highest predicted order conversion rate corresponding to each outbound thread as a recommended outbound seat corresponding to each outbound thread;
and counting the outbound clues corresponding to the outbound agents according to the recommended outbound agents corresponding to the outbound clues to obtain the outbound clue recommendation queue corresponding to the outbound agents.
In an embodiment of the present application, a possible implementation manner is provided, as shown in fig. 6, the apparatus shown in fig. 5 may further include a training module 610, where the training module 610 is configured to:
constructing an initial multi-mode outbound cue recommendation model;
collecting sample characteristic data of the outbound agents, sample characteristic data of the outbound clues, sample communication data of the outbound agents and the outbound clues and historical order conversion rates of the outbound agents corresponding to the outbound clues;
training the initial multi-mode outbound cue recommendation model based on the sample characteristic data of the outbound cue, the sample communication data of the outbound cue and the historical order conversion rate of each outbound cue corresponding to each outbound cue to obtain the trained multi-mode outbound cue recommendation model.
In an embodiment of the present application, a possible implementation manner is provided, and the training module 610 is further configured to:
carrying out cross processing on at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound clue and the sample communication data of the outbound seat and the outbound clue to generate outbound cross sample characteristic vector data;
and training the initial multi-mode outbound cue recommendation model by taking the outbound cross sample feature vector data as input and the historical order conversion rate of each outbound cue corresponding to each outbound seat as output to obtain the trained multi-mode outbound cue recommendation model.
In an embodiment of the present application, a possible implementation manner is provided, and the generating module 520 is further configured to:
selecting at least two kinds of data from the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue, and constructing a vector for each kind of data;
performing feature level cross processing on the data of at least two vectors to obtain a first processing result;
performing element-level cross processing on the data of at least two vectors to obtain a second processing result;
combining the data of at least two vectors pairwise, and performing low-order cross processing to obtain a third processing result;
and combining the first processing result, the second processing result and the third processing result, and performing transformation processing by using a preset function to generate outbound cross feature vector data.
Based on the same inventive concept, an embodiment of the present application further provides an electronic device, which includes a processor and a memory, where the memory stores a computer program, and the processor is configured to execute the computer program to perform the method for recommending an outbound call thread according to any of the above embodiments.
In an exemplary embodiment, there is provided an electronic device, as shown in fig. 7, an electronic device 700 shown in fig. 7 including: a processor 701 and a memory 703. The processor 701 is coupled to a memory 703, such as via a bus 702. Optionally, the electronic device 700 may also include a transceiver 704. It should be noted that the transceiver 704 is not limited to one in practical applications, and the structure of the electronic device 700 is not limited to the embodiment of the present application.
The Processor 701 may be a CPU (Central Processing Unit), a general-purpose Processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other Programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or execute the various illustrative logical blocks, modules, and circuits described in connection with the disclosure herein. The processor 701 may also be a combination of computing functions, e.g., comprising one or more microprocessors, DSPs, and microprocessors, among others.
The Memory 703 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical Disc storage, optical Disc storage (including Compact Disc, laser Disc, optical Disc, digital versatile Disc, blu-ray Disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to these.
The memory 703 is used for storing application program codes for executing the present invention, and is controlled by the processor 701. The processor 701 is configured to execute application program code stored in the memory 703 to implement the content shown in the foregoing method embodiments.
Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 7 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
Based on the same inventive concept, the present application further provides a computer-readable storage medium, in which a computer program is stored, where the computer program is configured to execute the method for recommending an outbound cue of any one of the above embodiments when the computer program runs.
It can be clearly understood by those skilled in the art that the specific working processes of the system, the apparatus, and the module described above may refer to the corresponding processes in the foregoing method embodiments, and for the sake of brevity, the detailed description is omitted here.
Those of ordinary skill in the art will understand that: the technical solution of the present application may be essentially or wholly or partially embodied in the form of a software product, where the computer software product is stored in a storage medium and includes program instructions for enabling an electronic device (e.g., a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application when the program instructions are executed. And the aforementioned storage medium includes: a U disk, a removable hard disk, a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, an optical disk, or the like.
Alternatively, all or part of the steps of implementing the foregoing method embodiments may be implemented by hardware (an electronic device such as a personal computer, a server, or a network device) associated with program instructions, which may be stored in a computer-readable storage medium, and when the program instructions are executed by a processor of the electronic device, the electronic device executes all or part of the steps of the method described in the embodiments of the present application.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present application; such modifications or substitutions do not depart from the scope of the present application.
Claims (14)
1. A method for recommending outbound cues, comprising:
acquiring feature data of the outbound seat, feature data of the outbound clue and communication data of the outbound seat and the outbound clue;
performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound clue and the communication data of the outbound seat and the outbound clue to generate outbound cross feature vector data;
inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat;
and obtaining an outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound thread corresponding to each outbound agent, and recommending the outbound thread based on the outbound thread recommendation queue corresponding to each outbound agent.
2. The method for recommending outbound threads according to claim 1, wherein recommending outbound threads based on the outbound thread recommendation queue corresponding to each outbound agent comprises:
allocating corresponding outbound threads to each outbound agent based on the outbound thread recommendation queue corresponding to each outbound agent; or
And providing the outbound cue recommendation queue corresponding to each outbound cue for the corresponding outbound cue, and sequentially communicating with the outbound cue by the corresponding outbound cue according to the outbound cue recommendation queue.
3. The method for recommending outbound threads according to claim 1, wherein obtaining the outbound thread recommendation queue corresponding to each outbound agent according to the predicted order conversion rate of each outbound agent corresponding to each outbound thread comprises:
determining the highest predicted order conversion rate corresponding to each outbound clue according to the predicted order conversion rate of each outbound clue corresponding to each outbound agent;
taking the outbound seat corresponding to the highest predicted order conversion rate corresponding to each outbound thread as a recommended outbound seat corresponding to each outbound thread;
and counting the outbound clues corresponding to the outbound agents according to the recommended outbound clues corresponding to the outbound clues to obtain an outbound clue recommendation queue corresponding to each outbound agent.
4. The method of claim 1, wherein the multi-modal outbound cue recommendation model is trained by the steps of;
constructing an initial multi-mode outbound cue recommendation model;
collecting sample characteristic data of the outbound agents, sample characteristic data of the outbound clues, sample communication data of the outbound agents and the outbound clues and historical order conversion rates of the outbound agents corresponding to the outbound clues;
training the initial multi-mode outbound cue recommendation model based on the sample characteristic data of the outbound agents, the sample characteristic data of the outbound cues, the sample communication data of the outbound agents and the outbound cues and the historical order conversion rate of each outbound cue corresponding to each outbound cue, and obtaining the trained multi-mode outbound cue recommendation model.
5. The method for recommending outbound threads according to claim 4, wherein the training of the initial multi-modal outbound thread recommendation model is performed based on the sample feature data of the outbound agents, the sample feature data of the outbound threads, the sample communication data of the outbound agents and the outbound threads, and the historical order conversion rate of each outbound agent corresponding to each outbound thread, so as to obtain the trained multi-modal outbound thread recommendation model, and the method comprises the following steps:
performing cross processing on at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound clue and the sample communication data of the outbound seat and the outbound clue to generate outbound cross sample characteristic vector data;
and taking the characteristic vector data of the outbound cross sample as input, taking the historical order conversion rate of each outbound cue corresponding to each outbound seat as output, and training the initial multi-mode outbound cue recommendation model to obtain the trained multi-mode outbound cue recommendation model.
6. The method of claim 1, wherein the cross-processing at least two of the feature data of the outbound agent, the feature data of the outbound thread, and the communication data between the outbound agent and the outbound thread to generate outbound cross feature vector data comprises:
selecting at least two kinds of data in the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue, and constructing a vector for each kind of data;
performing feature level cross processing on the data of at least two vectors to obtain a first processing result;
performing element-level cross processing on the data of at least two vectors to obtain a second processing result;
combining the data of at least two vectors pairwise, and performing low-order cross processing to obtain a third processing result;
and combining the first processing result, the second processing result and the third processing result, and performing transformation processing by using a preset function to generate outbound cross feature vector data.
7. An apparatus for recommending outbound clues, comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring the characteristic data of an outbound seat, the characteristic data of an outbound thread and the communication data of the outbound seat and the outbound thread;
the generating module is used for performing cross processing on at least two kinds of data in the feature data of the outbound seat, the feature data of the outbound clue and the communication data of the outbound seat and the outbound clue to generate outbound cross feature vector data;
the prediction module is used for inputting the outbound cross feature vector data into a pre-trained multi-mode outbound thread recommendation model, and predicting the order conversion rate of each outbound thread corresponding to each outbound seat by using the multi-mode outbound thread recommendation model to obtain the predicted order conversion rate of each outbound thread corresponding to each outbound seat;
and the recommending module is used for obtaining the outbound cue recommending queue corresponding to each outbound seat according to the predicted order conversion rate of each outbound seat corresponding to each outbound cue, and recommending the outbound cue based on the outbound cue recommending queue corresponding to each outbound seat.
8. The apparatus for recommending outbound hints of claim 7, wherein said recommending module is further configured to:
allocating corresponding outbound threads to each outbound agent based on the outbound thread recommendation queue corresponding to each outbound agent; or
And providing the outbound cue recommendation queue corresponding to each outbound cue for the corresponding outbound cue, and sequentially communicating with the outbound cue by the corresponding outbound cue according to the outbound cue recommendation queue.
9. The apparatus for recommending outbound hints of claim 7, wherein said recommending module is further configured to:
determining the highest predicted order conversion rate corresponding to each outbound clue according to the predicted order conversion rate of each outbound clue corresponding to each outbound agent;
taking the outbound seat corresponding to the highest predicted order conversion rate corresponding to each outbound thread as a recommended outbound seat corresponding to each outbound thread;
and counting the outbound clues corresponding to the outbound agents according to the recommended outbound clues corresponding to the outbound clues to obtain an outbound clue recommendation queue corresponding to each outbound agent.
10. The apparatus for recommending outbound hints of claim 7, further comprising a training module for:
constructing an initial multi-mode outbound cue recommendation model;
collecting sample characteristic data of the outbound agents, sample characteristic data of the outbound clues, sample communication data of the outbound agents and the outbound clues and historical order conversion rates of the outbound agents corresponding to the outbound clues;
and training the initial multi-mode outbound cue recommendation model based on the sample characteristic data of the outbound cue, the sample communication data of the outbound cue and the historical order conversion rate of each outbound cue corresponding to each outbound cue to obtain the trained multi-mode outbound cue recommendation model.
11. The device for recommending outbound hints of claim 10, wherein said training module is further configured to:
performing cross processing on at least two sample data in the sample characteristic data of the outbound seat, the sample characteristic data of the outbound clue and the sample communication data of the outbound seat and the outbound clue to generate outbound cross sample characteristic vector data;
and taking the characteristic vector data of the outbound cross sample as input, taking the historical order conversion rate of each outbound cue corresponding to each outbound seat as output, and training the initial multi-mode outbound cue recommendation model to obtain the trained multi-mode outbound cue recommendation model.
12. The apparatus for recommending outbound hints of claim 1, wherein said generating module is further configured to:
selecting at least two kinds of data in the characteristic data of the outbound seat, the characteristic data of the outbound clue and the communication data of the outbound seat and the outbound clue, and constructing a vector for each kind of data;
performing feature level cross processing on the data of at least two vectors to obtain a first processing result;
performing element-level cross processing on the data of at least two vectors to obtain a second processing result;
combining the data of at least two vectors pairwise, and performing low-order cross processing to obtain a third processing result;
and combining the first processing result, the second processing result and the third processing result, and performing transformation processing by using a preset function to generate outbound cross feature vector data.
13. An electronic device, comprising a processor and a memory, wherein the memory has stored therein a computer program, the processor being configured to execute the computer program to perform the method of recommending outbound cues as claimed in any one of claims 1 to 6.
14. A computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the method for recommending outbound cues according to any one of claims 1 to 6 when running.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210016824.2A CN114529143A (en) | 2022-01-07 | 2022-01-07 | Method and device for recommending outbound clues and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210016824.2A CN114529143A (en) | 2022-01-07 | 2022-01-07 | Method and device for recommending outbound clues and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114529143A true CN114529143A (en) | 2022-05-24 |
Family
ID=81621427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210016824.2A Pending CN114529143A (en) | 2022-01-07 | 2022-01-07 | Method and device for recommending outbound clues and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114529143A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829582A (en) * | 2023-02-16 | 2023-03-21 | 北京健康之家科技有限公司 | Intelligent thread allocation method and device, computer equipment and readable storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9426281B1 (en) * | 2013-09-05 | 2016-08-23 | Noble Systems Corporation | Automatic calling list scheduler |
| US9762733B1 (en) * | 2016-09-21 | 2017-09-12 | Genesys Telecommunications Laboratories, Inc. | System and method for recommending communication mediums based on predictive analytics |
| CN108650426A (en) * | 2018-02-08 | 2018-10-12 | 中移在线服务有限公司 | A kind of call center prediction calling-out method and system |
| CN111064849A (en) * | 2019-12-25 | 2020-04-24 | 北京合力亿捷科技股份有限公司 | Call center system based line resource utilization and management and control analysis method |
| CN112163155A (en) * | 2020-09-30 | 2021-01-01 | 深圳前海微众银行股份有限公司 | Information processing method, device, equipment and storage medium |
| KR20210118634A (en) * | 2020-03-23 | 2021-10-01 | 주식회사 베스트브라더스 | Call center service efficiency improvement system based on data and method thereof |
| US11176461B1 (en) * | 2017-08-29 | 2021-11-16 | Massachusetts Mutual Life Insurance Company | System and method for managing routing of customer calls to agents |
| CN113840040A (en) * | 2021-09-22 | 2021-12-24 | 平安普惠企业管理有限公司 | Man-machine cooperation outbound method, device, equipment and storage medium |
-
2022
- 2022-01-07 CN CN202210016824.2A patent/CN114529143A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9426281B1 (en) * | 2013-09-05 | 2016-08-23 | Noble Systems Corporation | Automatic calling list scheduler |
| US9762733B1 (en) * | 2016-09-21 | 2017-09-12 | Genesys Telecommunications Laboratories, Inc. | System and method for recommending communication mediums based on predictive analytics |
| US11176461B1 (en) * | 2017-08-29 | 2021-11-16 | Massachusetts Mutual Life Insurance Company | System and method for managing routing of customer calls to agents |
| CN108650426A (en) * | 2018-02-08 | 2018-10-12 | 中移在线服务有限公司 | A kind of call center prediction calling-out method and system |
| CN111064849A (en) * | 2019-12-25 | 2020-04-24 | 北京合力亿捷科技股份有限公司 | Call center system based line resource utilization and management and control analysis method |
| KR20210118634A (en) * | 2020-03-23 | 2021-10-01 | 주식회사 베스트브라더스 | Call center service efficiency improvement system based on data and method thereof |
| CN112163155A (en) * | 2020-09-30 | 2021-01-01 | 深圳前海微众银行股份有限公司 | Information processing method, device, equipment and storage medium |
| CN113840040A (en) * | 2021-09-22 | 2021-12-24 | 平安普惠企业管理有限公司 | Man-machine cooperation outbound method, device, equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829582A (en) * | 2023-02-16 | 2023-03-21 | 北京健康之家科技有限公司 | Intelligent thread allocation method and device, computer equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115485690A (en) | Batch technique for handling unbalanced training data of chat robots | |
| CN111325353A (en) | Method, device, equipment and storage medium for calculating contribution of training data set | |
| US11748452B2 (en) | Method for data processing by performing different non-linear combination processing | |
| CN110020022B (en) | Data processing method, device, equipment and readable storage medium | |
| US20240005165A1 (en) | Machine learning model training method, prediction method therefor, apparatus, device, computer-readable storage medium, and computer program product | |
| CN107807935B (en) | Using recommended method and device | |
| CN112418929A (en) | Data sharing method and device | |
| CN114529143A (en) | Method and device for recommending outbound clues and electronic equipment | |
| CN109447694B (en) | User characteristic analysis method and system | |
| WO2022001397A1 (en) | Interaction qualification level determination method, apparatus, and device, and storage medium | |
| CN113806634A (en) | Service package recommendation method and device and server | |
| CN112734462B (en) | Information recommendation method, device, equipment and medium | |
| CN111597441B (en) | Information processing method and device and electronic equipment | |
| CN114549048A (en) | Method and device for recommending outbound user and electronic equipment | |
| KR20230000461A (en) | System and method for ai learning on financial counseling data and recommending financial product therethrough | |
| Jorge et al. | Intelligent call routing for telecommunications call-centers | |
| CN113342998A (en) | Multimedia resource recommendation method and device, electronic equipment and storage medium | |
| CN111897910A (en) | Information pushing method and device | |
| CN112926834B (en) | Industrial APP quality evaluation method, device, equipment and medium | |
| CN117217852B (en) | Behavior recognition-based purchase willingness prediction method and device | |
| Hassan et al. | Multicriteria Decision Making for the Choice of the Global System for Mobile Telecommunication | |
| CN117391490A (en) | Evaluation information processing method and device for financial business and computer equipment | |
| CN118446775A (en) | Bank product information recommendation method and device | |
| CN118673202A (en) | Request processing method, device and system and storage medium | |
| CN111260466A (en) | Method and device for processing reach task |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100102 201 / F, block C, 2 lizezhong 2nd Road, Chaoyang District, Beijing Applicant after: Beijing Shuidi Technology Group Co.,Ltd. Address before: 100102 201, 2 / F, block C, No.2 lizezhong 2nd Road, Chaoyang District, Beijing Applicant before: Beijing Health Home Technology Co.,Ltd. |