CN112214576A - Public opinion analysis method, device, terminal equipment and computer readable storage medium - Google Patents
Public opinion analysis method, device, terminal equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN112214576A CN112214576A CN202010947333.0A CN202010947333A CN112214576A CN 112214576 A CN112214576 A CN 112214576A CN 202010947333 A CN202010947333 A CN 202010947333A CN 112214576 A CN112214576 A CN 112214576A
- Authority
- CN
- China
- Prior art keywords
- public opinion
- main body
- text
- analyzed
- public
- 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.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 79
- 239000011159 matrix material Substances 0.000 claims abstract description 63
- 238000000034 method Methods 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 21
- 239000013598 vector Substances 0.000 claims description 55
- 238000012549 training Methods 0.000 claims description 43
- 238000004590 computer program Methods 0.000 claims description 19
- 230000015654 memory Effects 0.000 claims description 7
- 238000012512 characterization method Methods 0.000 claims 1
- 230000008569 process Effects 0.000 description 8
- 238000007621 cluster analysis Methods 0.000 description 6
- 230000004044 response Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 108010001267 Protein Subunits Proteins 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002996 emotional effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Machine Translation (AREA)
Abstract
The application is applicable to the technical field of computers, and provides a public opinion analysis method, a public opinion analysis device, terminal equipment and a computer-readable storage medium. The method comprises the following steps: acquiring a text to be analyzed, and determining each public opinion main body contained in the text to be analyzed; determining a public opinion question-answer matrix corresponding to the text to be analyzed according to each public opinion main body and the text to be analyzed; and processing the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies. By determining each public opinion main body in the text to be analyzed and respectively carrying out public opinion analysis on each public opinion main body and sentences corresponding to the public opinion main bodies through the first preset model, the public opinion analysis result corresponding to each public opinion main body in the public opinion main bodies can be accurately obtained, and the accuracy of the public opinion analysis comprising the public opinion main bodies can be effectively improved.
Description
Technical Field
The application belongs to the technical field of computers, and particularly relates to a public opinion analysis method, a public opinion analysis device, terminal equipment and a computer-readable storage medium.
Background
At present, public opinion analysis is mostly performed on texts, and is used for judging the emotional tendency of given texts. However, when a plurality of public opinion main bodies are included in one text for public opinion analysis, the result of the public opinion analysis may be biased because the plurality of public opinion main bodies are included in one text; and the public opinion analysis result corresponding to each public opinion main body can not be accurately obtained.
Disclosure of Invention
The embodiment of the application provides a public opinion analysis method, a public opinion analysis device, terminal equipment and a computer readable storage medium, which can solve the technical problem of how to accurately obtain a public opinion analysis result containing a plurality of public opinion main bodies in a text.
In a first aspect, an embodiment of the present application provides a public opinion analysis method, including:
acquiring a text to be analyzed, and determining each public opinion main body contained in the text to be analyzed;
determining a public opinion question and answer matrix corresponding to the text to be analyzed according to each public opinion main body and the text to be analyzed, wherein the public opinion question and answer matrix comprises a plurality of public opinion question and answer vectors, and each public opinion question and answer vector is a vector formed by a public opinion main body and sentences related to the public opinion main body in the text to be analyzed;
and processing the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies.
Optionally, the determining, according to each of the public opinion subjects and the text to be analyzed, a public opinion question-answer matrix corresponding to the text to be analyzed includes:
obtaining a sentence containing a public opinion main body in the text to be analyzed;
determining candidate public sentiment sentences according to the sentences and the text to be analyzed;
and obtaining the public opinion question-answer matrix according to each public opinion main body and the candidate public opinion sentences.
Optionally, the determining a candidate public opinion sentence according to the sentence and the text to be analyzed includes:
clustering analysis is carried out on the sentences and the titles in the text to be analyzed, and related sentences corresponding to the titles are obtained;
obtaining each context sentence corresponding to the related sentence in the text to be analyzed;
and obtaining the candidate public opinion statement according to the related statement and the contextual statement.
Optionally, the obtaining the public opinion question-answer matrix according to each of the public opinion subjects and the candidate public opinion sentences includes:
obtaining candidate public sentiment sentences corresponding to the public sentiment main bodies;
respectively carrying out vectorization processing on each public opinion main body and the candidate public opinion sentences corresponding to each public opinion main body to obtain public opinion question-answer vectors corresponding to each public opinion main body;
and obtaining the public opinion question-answer matrix according to the public opinion question-answer vector corresponding to each public opinion main body.
Optionally, the determining each public opinion main body contained in the text to be analyzed includes:
and performing main body recognition on the text to be analyzed to obtain all public opinion main bodies contained in the text to be analyzed.
Optionally, the performing main body recognition on the text to be analyzed to obtain each public opinion main body contained in the text to be analyzed includes:
performing main body recognition on the text to be analyzed to obtain each initial public opinion main body contained in the text to be analyzed;
the method comprises the steps of obtaining an initial public opinion main body containing the same keywords, determining a target public opinion main body from the obtained initial public opinion main body, determining the target public opinion main body as the public opinion main body contained in a text to be analyzed, and enabling the target public opinion main body to be the initial public opinion main body with the maximum length.
Optionally, the first preset model is a characteristic quantity model BERT trained by utilizing a training public opinion text and a preset public opinion analysis result.
In a second aspect, an embodiment of the present application provides a public opinion analysis device, including:
the main body determining module is used for acquiring texts to be analyzed and determining entity name public opinion main bodies contained in the texts to be analyzed;
a matrix determining module, configured to determine a public sentiment question-answer matrix corresponding to the text to be analyzed according to each of the public sentiment bodies and the text to be analyzed, where the public sentiment question-answer matrix includes a plurality of public sentiment question-answer vectors, and each public sentiment question-answer vector is a vector composed of a public sentiment body and a sentence in the text to be analyzed, the sentence being associated with the public sentiment body;
and the result analysis module is used for processing the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies.
Optionally, the matrix determining module includes:
the sentence determining unit is used for acquiring a sentence containing a public opinion main body in the text to be analyzed;
the candidate public opinion sentence determining unit is used for determining candidate public opinion sentences according to the sentences and the texts to be analyzed;
and the matrix determining unit is used for obtaining the public opinion question and answer matrix according to each public opinion main body and the candidate public opinion sentences.
Optionally, the candidate public opinion statement determining unit includes:
the cluster analysis subunit is used for carrying out cluster analysis on the sentences and the titles in the texts to be analyzed to obtain related sentences corresponding to the titles;
the upper and lower sentence acquisition sub-unit is used for acquiring each context sentence corresponding to the related sentence in the text to be analyzed;
and the candidate public sentiment sentence sub-unit is used for obtaining the candidate public sentiment sentences according to the related sentences and the contextual sentences.
Optionally, the matrix determining unit includes:
a main body corresponding sentence determining sub-unit, configured to obtain candidate public sentiment sentences corresponding to each public sentiment main body;
the vector determination sub-unit is used for respectively carrying out vectorization processing on each public opinion main body and the candidate public opinion sentences corresponding to each public opinion main body to obtain the public opinion question-answer vector corresponding to each public opinion main body;
and the matrix determination sub-unit is used for obtaining the public opinion question-answer matrix according to the public opinion question-answer vector corresponding to each public opinion main body.
Optionally, the subject determination module includes:
and the main body determining unit is used for carrying out main body identification on the text to be analyzed to obtain each public opinion main body contained in the text to be analyzed.
Optionally, the subject determination unit includes:
the initial public opinion main body determining and dividing unit is used for carrying out main body recognition on the text to be analyzed to obtain each initial public opinion main body contained in the text to be analyzed;
the public opinion main body determining and dividing unit is used for acquiring initial public opinion main bodies containing the same keywords, determining a target public opinion main body from the acquired initial public opinion main bodies, and determining the target public opinion main body as the public opinion main body contained in the text to be analyzed, wherein the target public opinion main body is the initial public opinion main body with the maximum length.
In a third aspect, an embodiment of the present application provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the public opinion analysis method according to any one of the first aspect is implemented.
In a fourth aspect, the present application provides a computer-readable storage medium, where a computer program is stored, and when executed by a processor, the computer program implements the public opinion analysis method according to any one of the first aspect.
In a fifth aspect, the present application provides a computer program product, which when run on a terminal device, causes the terminal device to execute the public opinion analysis method according to any one of the above first aspects.
Compared with the prior art, the embodiment of the application has the advantages that:
in the embodiment of the application, each public opinion main body contained in the text to be analyzed is determined, and sentences related to each public opinion main body in the text to be analyzed are determined; and then carrying out vectorization processing on the public opinion main bodies and sentences related to the public opinion main bodies to obtain public opinion question and answer vectors corresponding to the public opinion main bodies, forming a public opinion question and answer matrix by the public opinion question and answer vectors and inputting the public opinion question and answer matrix into a first preset model to be processed, so that public opinion analysis results corresponding to the public opinion main bodies are respectively obtained, namely, the public opinion analysis results corresponding to the public opinion main bodies in the public opinion main bodies can be accurately obtained by determining the public opinion main bodies in the text to be analyzed and respectively carrying out public opinion analysis on the public opinion main bodies and the sentences corresponding to the public opinion main bodies through the first preset model, and the accuracy of the public opinion analysis comprising the public opinion main bodies can be effectively improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic flow chart illustrating a public opinion analysis method according to an embodiment of the present application;
fig. 2 is a schematic flow chart illustrating a public opinion analysis method according to a second embodiment of the present application;
fig. 3 is a schematic structural diagram of a public opinion analysis device according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to" determining "or" in response to detecting ". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
Furthermore, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used for distinguishing between descriptions and not necessarily for describing or implying relative importance.
Reference throughout this specification to "one embodiment" or "some embodiments," or the like, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," or the like, in various places throughout this specification are not necessarily all referring to the same embodiment, but rather "one or more but not all embodiments" unless specifically stated otherwise. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless expressly specified otherwise.
The public opinion analysis method provided by the embodiment of the application can be applied to mobile phones, tablet computers, wearable devices, vehicle-mounted devices, Augmented Reality (AR)/Virtual Reality (VR) devices, notebook computers, ultra-mobile personal computers (UMPCs), netbooks, Personal Digital Assistants (PDAs) and other terminal devices, and the embodiment of the application does not limit the specific types of the terminal devices at all.
The execution main body of the public opinion analysis method provided by the embodiment of the application is terminal equipment, and the terminal equipment acquires texts to be analyzed and determines all public opinion main bodies contained in the texts to be analyzed; the terminal equipment determines a public opinion question-answer matrix corresponding to the text to be analyzed according to each public opinion main body and the text to be analyzed, wherein the public opinion question-answer matrix comprises a plurality of public opinion question-answer vectors, and each public opinion question-answer vector is a public opinion main body and a vector formed by sentences related to the public opinion main body in the text to be analyzed; and the terminal equipment processes the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies.
Fig. 1 shows a schematic flowchart of a public opinion analysis method provided in an embodiment of the present application, and the method may be applied to the terminal device as an example and not a limitation. As shown in fig. 1, the method may include:
s101, obtaining a text to be analyzed, and determining each public opinion main body contained in the text to be analyzed;
the text to be analyzed is a text which needs to be analyzed for public sentiment, and the text to be analyzed may include one public sentiment main body or a plurality of public sentiment main bodies. After the terminal device obtains the text to be analyzed, the main body recognition can be performed on the text to be analyzed, and all public opinion main bodies contained in the text to be analyzed are obtained. The public opinion main body can be a company name contained in the text to be analyzed, and can also be a character name contained in the text to be analyzed. For example, when the text to be analyzed includes public sentiment subjects such as "AA company", "management meeting", and "BB", after the terminal device acquires the text to be analyzed, the terminal device may recognize the "AA company", "management meeting", and "BB" in the text to be analyzed through the subject recognition model.
The subject recognition model may be a model composed of a Long Short-Term Memory network (LSTM) and a Conditional Random Field (CRF), or a model composed of an iterative conditional random field (ID-CNN) and a Conditional Random Field (CRF).
In a possible implementation manner, the main body recognition is performed on the text to be analyzed, and obtaining each public opinion main body included in the text to be analyzed may be: performing main body recognition on the text to be analyzed to obtain each initial public opinion main body contained in the text to be analyzed; the method comprises the steps of obtaining an initial public opinion main body containing the same keywords, determining a target public opinion main body from the obtained initial public opinion main body, determining the target public opinion main body as the public opinion main body contained in a text to be analyzed, and enabling the target public opinion main body to be the initial public opinion main body with the maximum length.
In the implementation mode, after the terminal device obtains the text to be analyzed, the main body recognition of the text to be analyzed can be performed through the main body recognition model, so that a plurality of initial public opinion main bodies contained in the text to be analyzed are obtained. Then, the terminal equipment screens a plurality of initial public opinion main bodies to obtain the initial public opinion main bodies containing the same keywords, and determines a target public opinion main body from the obtained initial public opinion main bodies; then, the terminal device may determine a target public opinion main body as the public opinion main body included in the text to be analyzed, where the target public opinion main body is an initial public opinion main body having a maximum length.
For example, when the text to be analyzed includes public sentiment subjects such as "AA company, AA, american AA company, management society, american management society, BB company, and BB", after the terminal device acquires the text to be analyzed, the terminal device may identify initial public sentiment subjects such as "AA company, AA, american AA company, management society, american management society, BB company, and BB" in the text to be analyzed through the subject identification model. Then, the terminal device further screens each initial public opinion main body, can obtain initial public opinion main bodies such as "AA company", "AA" and "american AA company" having the same keyword "AA", and can determine that the target public opinion main body is "american AA company" from the initial public opinion main bodies such as "AA company", "AA" and "american AA company". Meanwhile, initial public opinion subjects such as the management meeting and the American management meeting with the same keyword can be obtained, and the target public opinion subject can be determined to be the American management meeting from the initial public opinion subjects such as the management meeting and the American management meeting. Meanwhile, initial public opinion main bodies such as 'BB company' and 'BB' with the same keyword 'BB' can be obtained, and the target public opinion main body can be determined to be 'BB company' from the initial public opinion main bodies such as 'BB company' and 'BB'. Then, the terminal device determines "united states AA company", "united states management society", and "BB company" as the public opinion subjects contained in the text to be analyzed.
S102, determining a public opinion question and answer matrix corresponding to the text to be analyzed according to each public opinion main body and the text to be analyzed, wherein the public opinion question and answer matrix comprises a plurality of public opinion question and answer vectors, and each public opinion question and answer vector is a public opinion main body and a vector formed by sentences related to the public opinion main body in the text to be analyzed;
in this embodiment, after obtaining each public opinion main body included in the text to be analyzed, the terminal device may determine sentences associated with each public opinion main body in the text to be analyzed, and combine a public opinion main body and the sentences associated with the public opinion main body into a public opinion question-answer vector; and then forming a public opinion question-answer matrix by the public opinion question-answer vectors corresponding to the plurality of public opinion main bodies. Wherein, the sentence is a natural sentence in the text to be analyzed. The sentence related to the public opinion main body may be a sentence including the public opinion main body, or may be a sentence including the public opinion main body, one or more previous sentences of the sentence, and one or more next sentences of the sentence. The sentence related to the public sentiment main body in the public sentiment question-answer vector can be one sentence or a plurality of sentences.
For example, when a public opinion main body is "american AA company", the terminal device determines that a sentence "american AA company" related to "american AA company" is troubled and infringed by CC in a text to be analyzed, and the terminal device composes the "american AA company" and the "american AA company troubled and infringed by CC into a public opinion answer vector. Then, the terminal equipment forms a public opinion question-answer matrix by the plurality of public opinion question-answer vectors.
In one possible implementation, as shown in fig. 2, S102 may include S201, S202, and S203.
S201, obtaining a sentence containing a public opinion main body in the text to be analyzed;
specifically, after determining each public opinion main body contained in the text to be analyzed, the terminal device may screen out all sentences containing the public opinion main bodies from the text to be analyzed. Each sentence may be a natural sentence in the text to be analyzed.
S202, determining candidate public opinion sentences according to the sentences and the text to be analyzed;
specifically, after acquiring all sentences including the main body of public opinion, the terminal device acquires candidate public opinion sentences associated with each sentence in the text to be analyzed. The candidate public opinion sentences may be sentences including a public opinion main body, or sentences including a public opinion main body, one or more previous sentences of the sentences, and one or more next sentences of the sentences. The candidate public opinion sentence can be one sentence or a plurality of sentences.
In a possible implementation manner, determining a candidate public opinion sentence according to the sentence and the text to be analyzed may be: clustering analysis is carried out on the sentences and the titles in the text to be analyzed, and related sentences corresponding to the titles are obtained; obtaining each context sentence corresponding to the related sentence in the text to be analyzed; and obtaining the candidate public opinion statement according to the related statement and the contextual statement.
In this implementation, after acquiring all sentences including a public opinion main body, the terminal device may perform cluster analysis on all sentences including the public opinion main body and titles in a text to be analyzed by using a k-means clustering algorithm (k-means clustering algorithm). Specifically, all sentences including public opinion subjects and titles in texts to be analyzed are divided into 2 groups, 2 objects are randomly selected as initial clustering centers, then the distance between each object and each clustering center is calculated, and each object is assigned to the clustering center closest to the object. The cluster centers and the objects assigned to them represent a class cluster. After completing one-time allocation of all objects, the terminal device may recalculate the clustering center according to the existing objects in the class cluster, and then perform the clustering operation again. This process will be repeated until some termination condition is met. When the termination condition is satisfied, the clustering is terminated, and the terminal device may determine all sentences in the cluster where the title is located as related sentences corresponding to the title.
The termination condition may be that the number of changes of the objects in each type of cluster is less than or equal to a preset number, or that cluster centers which are less than the preset number change again, or that the sum of squared errors is locally minimum. The object may be a sentence or a title after vectorization processing.
In the implementation manner, after obtaining all the relevant sentences, the terminal device obtains context sentences corresponding to the relevant sentences in the text to be analyzed. And the terminal equipment combines each related statement and the context statement corresponding to the related statement in sequence respectively to obtain the candidate public opinion statement corresponding to each related statement. And then all the candidate public sentiment sentences are combined together to form the candidate public sentiment sentences. The context sentences are sentences in which the text to be analyzed is arranged in the front sentence or the front few sentences of the related sentences, and sentences in which the text to be analyzed is arranged in the back sentence or the back few sentences of the related sentences.
S203, obtaining the public opinion question-answer matrix according to each public opinion main body and the candidate public opinion sentences.
Specifically, after determining candidate public sentiment sentences, the terminal equipment combines a public sentiment main body and the candidate public sentiment sentences corresponding to the public sentiment main body to generate a public sentiment question-answer vector; and then forming a public opinion question-answer matrix by the public opinion question-answer vectors corresponding to the plurality of public opinion main bodies.
In one possible implementation manner, the public opinion question and answer matrix obtained according to each public opinion main body and the candidate public opinion sentences may be: obtaining candidate public sentiment sentences corresponding to the public sentiment main bodies; respectively carrying out vectorization processing on each public opinion main body and the candidate public opinion sentences corresponding to each public opinion main body to obtain public opinion question-answer vectors corresponding to each public opinion main body; and obtaining the public opinion question-answer matrix according to the public opinion question-answer vector corresponding to each public opinion main body.
In this implementation, after acquiring all candidate public sentiment sentences, the terminal device may acquire the candidate public sentiment sentences corresponding to each public sentiment main body respectively. Then, each candidate public opinion statement can be segmented through a segmentation tool, and each candidate public opinion statement is vectorized through a term frequency-inverse document frequency (TF-IDF) method aiming at the keywords. The vectorized candidate public opinion statement can be expressed as a vector
Fi＝TFIDFi. Wherein,
D＝{d1,d2,...,dn}，
tfithe word frequency of the ith word in the candidate public sentiment sentence is shown, D is the number of texts in the training set, WnFor the vocabulary size, ε is a smoothing factor, 1{ i ∈ djMeans if djThe word i is included, which is 1. Then, the nearness quantization processing can be carried out on each public opinion main body through TF-IDF. Then, each public opinion main body which is vectorized and the candidate public opinion sentences which correspond to the public opinion main bodies and are vectorized can be combined to obtain the public opinion question and answer vector corresponding to each public opinion main body. Then, the terminal equipment combines the public sentiment question and answer vectors together in sequence according to the sequence of the public sentiment main body to form a public sentiment question and answer matrix. The candidate public sentiment sentences corresponding to one public sentiment main body may include one or more sentences. The format of the public opinion question-answer vector can be [ candidate public opinion sentence corresponding to the public opinion main body, public opinion main body]. The public opinion question-answer matrix is a matrix formed by a plurality of public opinion question-answer vectors.
And S103, processing the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies.
In the embodiment of the application, after the terminal device obtains the public opinion question and answer matrix, the public opinion main bodies in each public opinion question and answer vector in the public opinion question and answer matrix and the candidate public opinion sentences related to the public opinion main bodies are respectively input into the first preset model by sentences 2 and 1 for processing, and the public opinion analysis results corresponding to the public opinion main bodies are respectively obtained. The first preset model may be a token quantity model (BERT) trained by using a training public sentiment text and a preset public sentiment analysis result. The training public opinion text is a text for training the BERT model, and the preset public opinion analysis result is a public opinion analysis result corresponding to each training public opinion main body in the training public opinion text. And presetting the public opinion analysis result as a given value. The preset public opinion analysis result can be positive, negative and neutral.
When the first preset model is trained, the training public opinion texts are processed in the same manner as in S101 and S102 to obtain a training public opinion question-answer matrix. And inputting a training public sentiment main body in each training public sentiment question-answer vector in the training public sentiment question-answer matrix and a training sentence associated with the training public sentiment main body into the BERT model in a sentence 2 mode and a sentence 1 mode respectively. And then, inputting a preset public opinion analysis result into the BERT model. The method comprises the steps of training a BERT model through a training public opinion main body, training sentences relevant to the training public opinion main body and a preset public opinion analysis result to obtain a first preset model. The training public opinion main body is a public opinion main body contained in the training public opinion text, and the training public opinion question-answer matrix is obtained by processing the training public opinion text in the S101 and S102 modes.
Specifically, when the first preset model is trained, the terminal device processes the training public opinion text in the same manner as in S101 and S102 to obtain each training public opinion main body and a training public opinion sentence corresponding to each training public opinion main body, and marks a preset public opinion analysis result corresponding to each training public opinion main body. Then, the terminal device may perform vectorization processing on each training public opinion main body and the training public opinion sentences corresponding to the training public opinion main body, so as to obtain the public opinion question-answer vector [ seg ] corresponding to each training public opinion main bodyi,enti]. Then, the terminal device may transmit the public sentiment question-answer vector [ segi,enti]Preset public opinion analysis result label corresponding to the public opinion question-answer vectoriAnd inputting the BERT model for processing to obtain the training public sentiment analysis result corresponding to each training public sentiment main body. Wherein, entiFor the ith training public opinion principal, segiFor the training public sentiment sentence corresponding to the ith training public sentiment subject, labeliAnd presetting a public opinion analysis result corresponding to the ith training public opinion main body. Subsequently, the terminal device can calculate a training error according to the training public opinion analysis result and the preset public opinion analysis result. When the training error is larger than the preset error, the terminal equipment can adjust the model parameters of the BERT model, and continues to train the adjusted BERT model by using the training public sentiment text and the corresponding preset public sentiment analysis result until the training error is smaller than or equal to the preset error, so that the trained BERT model is obtained.
Specifically, when the terminal device identifies the public opinion main body in the text to be analyzed through the main body identification model, the obtained public opinion main body can be represented as
Wherein, entiIs the main body of the ith public opinion,
and
respectively are the coordinates of the public opinion main body in the text to be analyzed. Then, the terminal equipment determines a sentence containing the public opinion main body according to the coordinates of the public opinion main body, wherein the sentence containing the public opinion main body can be expressed as
Wherein a public sentiment subject may appear in a plurality of sentences of a text, NiThe number of sentences appearing in the text to be analyzed for the public opinion main body i. Then, the terminal device checks all the sentences SiAnd the title s in the text to be analyzedTPerforming cluster analysis to obtain a title sTCorresponding related statement Sc＝[s1,s2,...,sj]Wherein s isjIs one sentence in the related sentence. Then, the terminal device obtains the sentence s related to the text to be analyzedjCorresponding context statement sj-n,sj+n. Wherein n is a natural number. The terminal device then follows a correlation statement sjAnd the context statement s of the related statementj-n,sj+nObtaining a candidate public sentiment sentence cj＝[sj-n,sj,sj+n]. Then, the terminal equipment acquires candidate public sentiment sentences corresponding to all public sentiment main bodies i
CiThe candidate public sentiment sentences are corresponding to the public sentiment main bodies. Then, the terminal device sends each public opinion main agent entiCandidate public sentiment sentences C corresponding to each public sentiment main bodyiRespectively carrying out vectorization treatment to obtain public sentiment question-answer vectors [ C ] corresponding to the public sentiment main bodiesi,enti]. Then, the terminal device sends the candidate public sentiment sentence C corresponding to the public sentiment main bodyiDetermined as sentence 2 and public opinion main bodyentiInputting the mode determined as sentence 1 into a BERT model for processing to obtain a public opinion main body entiAnd (5) corresponding public opinion analysis results. BERT model on reception CiAnd entiThereafter, C may be addediAnd entiConverts each word in (1) into the sum of word Embedding (Token Embedding), Segment Embedding (Segment Embedding) and Position Embedding (Position Embedding), and takes CLS as a start mark and SEP as a separator of two sentences. Then, the BERT model pairs the converted CiAnd entiAnd carrying out public sentiment analysis to obtain a public sentiment analysis result.
In summary, by determining each public opinion main body contained in the text to be analyzed, and determining sentences associated with each public opinion main body in the text to be analyzed; and then carrying out vectorization processing on the public opinion main bodies and sentences related to the public opinion main bodies to obtain public opinion question and answer vectors corresponding to the public opinion main bodies, forming a public opinion question and answer matrix by the public opinion question and answer vectors and inputting the public opinion question and answer matrix into a first preset model to be processed, so that public opinion analysis results corresponding to the public opinion main bodies are respectively obtained, namely, the public opinion analysis results corresponding to the public opinion main bodies in the public opinion main bodies can be accurately obtained by determining the public opinion main bodies in the text to be analyzed and respectively carrying out public opinion analysis on the public opinion main bodies and the sentences corresponding to the public opinion main bodies through the first preset model, and the accuracy of the public opinion analysis comprising the public opinion main bodies can be effectively improved.
Fig. 3 is a block diagram illustrating a configuration of a public opinion analyzing apparatus according to an embodiment of the present application, which corresponds to the public opinion analyzing method according to the above embodiment.
Referring to fig. 3, the apparatus includes:
a main body determining module 301, configured to obtain a text to be analyzed, and determine a main body of each entity name contained in the text to be analyzed;
a matrix determining module 302, configured to determine a public sentiment question-answer matrix corresponding to the text to be analyzed according to each of the public sentiment bodies and the text to be analyzed, where the public sentiment question-answer matrix includes a plurality of public sentiment question-answer vectors, and each public sentiment question-answer vector is a vector composed of a public sentiment body and a sentence in the text to be analyzed that is associated with the public sentiment body;
the result analysis module 303 is configured to process the public opinion question-answer matrix through a first preset model, and obtain public opinion analysis results corresponding to each public opinion main body respectively.
Optionally, the determining module 302 may include:
the sentence determining unit is used for acquiring a sentence containing a public opinion main body in the text to be analyzed;
the candidate public opinion sentence determining unit is used for determining candidate public opinion sentences according to the sentences and the texts to be analyzed;
and the matrix determining unit is used for obtaining the public opinion question and answer matrix according to each public opinion main body and the candidate public opinion sentences.
Optionally, the candidate public opinion statement determination unit may include:
the cluster analysis subunit is used for carrying out cluster analysis on the sentences and the titles in the texts to be analyzed to obtain related sentences corresponding to the titles;
the upper and lower sentence acquisition sub-unit is used for acquiring each context sentence corresponding to the related sentence in the text to be analyzed;
and the candidate public sentiment sentence sub-unit is used for obtaining the candidate public sentiment sentences according to the related sentences and the contextual sentences.
Optionally, the matrix determining unit may include:
a main body corresponding sentence determining sub-unit, configured to obtain candidate public sentiment sentences corresponding to each public sentiment main body;
the vector determination sub-unit is used for respectively carrying out vectorization processing on each public opinion main body and the candidate public opinion sentences corresponding to each public opinion main body to obtain the public opinion question-answer vector corresponding to each public opinion main body;
and the matrix determination sub-unit is used for obtaining the public opinion question-answer matrix according to the public opinion question-answer vector corresponding to each public opinion main body.
Optionally, the subject determining module 301 may include:
and the main body determining unit is used for carrying out main body identification on the text to be analyzed to obtain each public opinion main body contained in the text to be analyzed.
Alternatively, the subject determination unit may include:
the initial public opinion main body determining and dividing unit is used for carrying out main body recognition on the text to be analyzed to obtain each initial public opinion main body contained in the text to be analyzed;
the public opinion main body determining and dividing unit is used for acquiring initial public opinion main bodies containing the same keywords, determining a target public opinion main body from the acquired initial public opinion main bodies, and determining the target public opinion main body as the public opinion main body contained in the text to be analyzed, wherein the target public opinion main body is the initial public opinion main body with the maximum length.
It should be noted that, for the information interaction, execution process, and other contents between the above-mentioned devices/units, the specific functions and technical effects thereof are based on the same concept as those of the embodiment of the method of the present application, and specific reference may be made to the part of the embodiment of the method, which is not described herein again.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
An embodiment of the present application further provides a terminal device, where the terminal device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, the processor implementing the steps of any of the various method embodiments described above when executing the computer program.
The embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments may be implemented.
The embodiments of the present application provide a computer program product, which when running on a terminal device, enables the terminal device to implement the steps in the above method embodiments when executed.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium and can implement the steps of the embodiments of the methods described above when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable storage medium may include at least: any entity or device capable of carrying computer program code to the apparatus/terminal device, recording medium, computer memory, read-only memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, and software distribution medium. Such as a usb-disk, a removable hard disk, a magnetic or optical disk, etc. In certain jurisdictions, computer-readable storage media may not be an electrical carrier signal or a telecommunications signal in accordance with legislative and proprietary practices.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The above-mentioned 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 may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (10)
1. A public opinion analysis method is characterized by comprising the following steps:
acquiring a text to be analyzed, and determining each public opinion main body contained in the text to be analyzed;
determining a public opinion question and answer matrix corresponding to the text to be analyzed according to each public opinion main body and the text to be analyzed, wherein the public opinion question and answer matrix comprises a plurality of public opinion question and answer vectors, and each public opinion question and answer vector is a vector formed by a public opinion main body and sentences related to the public opinion main body in the text to be analyzed;
and processing the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies.
2. The public opinion analysis method according to claim 1, wherein the determining a public opinion question and answer matrix corresponding to the text to be analyzed according to each of the public opinion main bodies and the text to be analyzed comprises:
obtaining a sentence containing a public opinion main body in the text to be analyzed;
determining candidate public sentiment sentences according to the sentences and the text to be analyzed;
and obtaining the public opinion question-answer matrix according to each public opinion main body and the candidate public opinion sentences.
3. The public opinion analysis method according to claim 2, wherein the determining candidate public opinion sentences according to the sentences and the text to be analyzed comprises:
clustering analysis is carried out on the sentences and the titles in the text to be analyzed, and related sentences corresponding to the titles are obtained;
obtaining a context sentence corresponding to the related sentence in the text to be analyzed;
and obtaining the candidate public opinion statement according to the related statement and the contextual statement.
4. The public opinion analysis method according to claim 2, wherein the obtaining of the public opinion question and answer matrix according to each of the public opinion main bodies and the candidate public opinion sentences comprises:
obtaining candidate public sentiment sentences corresponding to the public sentiment main bodies;
respectively carrying out vectorization processing on each public opinion main body and the candidate public opinion sentences corresponding to each public opinion main body to obtain public opinion question-answer vectors corresponding to each public opinion main body;
and obtaining the public opinion question-answer matrix according to the public opinion question-answer vector corresponding to each public opinion main body.
5. The public opinion analysis method according to any one of claims 1 to 4, wherein the determining of each public opinion main body included in the text to be analyzed includes:
and performing main body recognition on the text to be analyzed to obtain all public opinion main bodies contained in the text to be analyzed.
6. The public opinion analysis method according to claim 5, wherein the main body recognition of the text to be analyzed to obtain each public opinion main body included in the text to be analyzed comprises:
performing main body recognition on the text to be analyzed to obtain each initial public opinion main body contained in the text to be analyzed;
the method comprises the steps of obtaining an initial public opinion main body containing the same keywords, determining a target public opinion main body from the obtained initial public opinion main body, determining the target public opinion main body as the public opinion main body contained in a text to be analyzed, and enabling the target public opinion main body to be the initial public opinion main body with the maximum length.
7. The public opinion analysis method according to any one of claims 1 to 4, wherein the first preset model is a characterization quantity model BERT trained by using a training public opinion text and a preset public opinion analysis result.
8. A public opinion analysis device, characterized by comprising:
the main body determining module is used for acquiring texts to be analyzed and determining entity name public opinion main bodies contained in the texts to be analyzed;
a matrix determining module, configured to determine a public sentiment question-answer matrix corresponding to the text to be analyzed according to each of the public sentiment bodies and the text to be analyzed, where the public sentiment question-answer matrix includes a plurality of public sentiment question-answer vectors, and each public sentiment question-answer vector is a vector composed of a public sentiment body and a sentence in the text to be analyzed, the sentence being associated with the public sentiment body;
and the result analysis module is used for processing the public opinion question-answer matrix through a first preset model to respectively obtain public opinion analysis results corresponding to the public opinion main bodies.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the public opinion analysis method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the public opinion analysis method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010947333.0A CN112214576B (en) | 2020-09-10 | 2020-09-10 | Public opinion analysis method, public opinion analysis device, terminal equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010947333.0A CN112214576B (en) | 2020-09-10 | 2020-09-10 | Public opinion analysis method, public opinion analysis device, terminal equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112214576A true CN112214576A (en) | 2021-01-12 |
| CN112214576B CN112214576B (en) | 2024-02-06 |
Family
ID=74049336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010947333.0A Active CN112214576B (en) | 2020-09-10 | 2020-09-10 | Public opinion analysis method, public opinion analysis device, terminal equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112214576B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114201600A (en) * | 2021-12-10 | 2022-03-18 | 北京金堤科技有限公司 | Public opinion text abstract extraction method, device, equipment and computer storage medium |
| CN114201601A (en) * | 2021-12-10 | 2022-03-18 | 北京金堤科技有限公司 | Public opinion text abstract extraction method, device, equipment and computer storage medium |
| CN114547167A (en) * | 2022-01-27 | 2022-05-27 | 启明信息技术股份有限公司 | Automobile public opinion sentiment analysis method |
| CN118095252A (en) * | 2024-01-04 | 2024-05-28 | 网智天元科技集团股份有限公司 | Training method, recognition method and device for public opinion associated enterprise recognition model |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103544255A (en) * | 2013-10-15 | 2014-01-29 | 常州大学 | Text semantic relativity based network public opinion information analysis method |
| US20140046891A1 (en) * | 2012-01-25 | 2014-02-13 | Sarah Banas | Sapient or Sentient Artificial Intelligence |
| CN108733644A (en) * | 2018-04-09 | 2018-11-02 | 平安科技（深圳）有限公司 | A kind of text emotion analysis method, computer readable storage medium and terminal device |
| CN109977225A (en) * | 2019-03-13 | 2019-07-05 | 咪咕文化科技有限公司 | Public opinion analysis method and device |
| CN110110323A (en) * | 2019-04-10 | 2019-08-09 | 北京明略软件系统有限公司 | A kind of text sentiment classification method and device, computer readable storage medium |
| CN110705255A (en) * | 2019-10-12 | 2020-01-17 | 京东数字科技控股有限公司 | Method and device for detecting association relation between sentences |
| CN110705300A (en) * | 2019-09-27 | 2020-01-17 | 上海烨睿信息科技有限公司 | Emotion analysis method, emotion analysis system, computer terminal and storage medium |
| CN111177374A (en) * | 2019-12-13 | 2020-05-19 | 航天信息股份有限公司 | Active learning-based question and answer corpus emotion classification method and system |
| CN111191438A (en) * | 2019-12-30 | 2020-05-22 | 北京百分点信息科技有限公司 | Emotion analysis method and device and electronic equipment |
| CN111222032A (en) * | 2019-12-17 | 2020-06-02 | 中国平安人寿保险股份有限公司 | Public opinion analysis method and related equipment |
| WO2020147395A1 (en) * | 2019-01-17 | 2020-07-23 | 平安科技（深圳）有限公司 | Emotion-based text classification method and device, and computer apparatus |
-
2020
- 2020-09-10 CN CN202010947333.0A patent/CN112214576B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140046891A1 (en) * | 2012-01-25 | 2014-02-13 | Sarah Banas | Sapient or Sentient Artificial Intelligence |
| CN103544255A (en) * | 2013-10-15 | 2014-01-29 | 常州大学 | Text semantic relativity based network public opinion information analysis method |
| CN108733644A (en) * | 2018-04-09 | 2018-11-02 | 平安科技（深圳）有限公司 | A kind of text emotion analysis method, computer readable storage medium and terminal device |
| WO2020147395A1 (en) * | 2019-01-17 | 2020-07-23 | 平安科技（深圳）有限公司 | Emotion-based text classification method and device, and computer apparatus |
| CN109977225A (en) * | 2019-03-13 | 2019-07-05 | 咪咕文化科技有限公司 | Public opinion analysis method and device |
| CN110110323A (en) * | 2019-04-10 | 2019-08-09 | 北京明略软件系统有限公司 | A kind of text sentiment classification method and device, computer readable storage medium |
| CN110705300A (en) * | 2019-09-27 | 2020-01-17 | 上海烨睿信息科技有限公司 | Emotion analysis method, emotion analysis system, computer terminal and storage medium |
| CN110705255A (en) * | 2019-10-12 | 2020-01-17 | 京东数字科技控股有限公司 | Method and device for detecting association relation between sentences |
| CN111177374A (en) * | 2019-12-13 | 2020-05-19 | 航天信息股份有限公司 | Active learning-based question and answer corpus emotion classification method and system |
| CN111222032A (en) * | 2019-12-17 | 2020-06-02 | 中国平安人寿保险股份有限公司 | Public opinion analysis method and related equipment |
| CN111191438A (en) * | 2019-12-30 | 2020-05-22 | 北京百分点信息科技有限公司 | Emotion analysis method and device and electronic equipment |
Non-Patent Citations (2)
| Title |
|---|
| ZHILIANG ZHU ET AL.: "Hot Topic Detection Based on a Refined TF-IDF Algorithm", 《IEEE ACCESS》, pages 26996 - 27007 * |
| 胡东瑶: "面向司法领域的舆情监测技术研究与系统实现", 《中国优秀硕士学位论文全文数据库 社会科学I辑》, pages 120 - 469 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114201600A (en) * | 2021-12-10 | 2022-03-18 | 北京金堤科技有限公司 | Public opinion text abstract extraction method, device, equipment and computer storage medium |
| CN114201601A (en) * | 2021-12-10 | 2022-03-18 | 北京金堤科技有限公司 | Public opinion text abstract extraction method, device, equipment and computer storage medium |
| CN114201601B (en) * | 2021-12-10 | 2023-03-28 | 北京金堤科技有限公司 | Public opinion text abstract extraction method, device, equipment and computer storage medium |
| CN114547167A (en) * | 2022-01-27 | 2022-05-27 | 启明信息技术股份有限公司 | Automobile public opinion sentiment analysis method |
| CN118095252A (en) * | 2024-01-04 | 2024-05-28 | 网智天元科技集团股份有限公司 | Training method, recognition method and device for public opinion associated enterprise recognition model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112214576B (en) | 2024-02-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840321B (en) | Text recommendation method and device and electronic equipment | |
| CN108334490B (en) | Keyword extraction method and keyword extraction device | |
| CN112214576A (en) | Public opinion analysis method, device, terminal equipment and computer readable storage medium | |
| CN113139387B (en) | Semantic error correction method, electronic device and storage medium | |
| CN109299280B (en) | Short text clustering analysis method and device and terminal equipment | |
| CN112668320A (en) | Model training method and device based on word embedding, electronic equipment and storage medium | |
| CN111178056A (en) | Deep learning based file generation method and device and electronic equipment | |
| CN107330009B (en) | Method and apparatus for creating topic word classification model, and storage medium | |
| CN111401044A (en) | Title generation method and device, terminal equipment and storage medium | |
| CN117828042A (en) | Question and answer processing method, device, equipment and medium for financial service | |
| CN111444321B (en) | Question answering method, device, electronic equipment and storage medium | |
| CN111538830A (en) | French retrieval method, French retrieval device, computer equipment and storage medium | |
| CN111858966B (en) | Knowledge graph updating method and device, terminal equipment and readable storage medium | |
| WO2024179519A1 (en) | Semantic recognition method and apparatus | |
| CN110705308B (en) | Voice information domain identification method and device, storage medium and electronic equipment | |
| CN112632956A (en) | Text matching method, device, terminal and storage medium | |
| CN114077834A (en) | Method, device and storage medium for determining similar texts | |
| CN115687790B (en) | Advertisement pushing method and system based on big data and cloud platform | |
| CN115062783B (en) | Entity alignment method and related device, electronic equipment and storage medium | |
| CN114528851B (en) | Reply sentence determination method, reply sentence determination device, electronic equipment and storage medium | |
| CN113378541B (en) | Text punctuation prediction method, device, system and storage medium | |
| CN114254634A (en) | Multimedia data mining method, device, storage medium and equipment | |
| CN113656538A (en) | Method and device for generating regular expression, computing equipment and storage medium | |
| CN113360742A (en) | Recommendation information determination method and device and electronic equipment | |
| CN116578693B (en) | Text retrieval method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |