Corrected some changes that were mistakenly rolled back in Version 3.2.

We added the following methods for clustering, which are specialized to Clean-Clean ER:

1. Connected Components Clustering for CCER
2. Ricochect SR Clustering for CCER
3. Best Assignment Heuristic with Time-out
4. Best Match Clustering
5. Exact Clustering
6. Kiraly Clustering

changelog:

1. added two more workflows
   • Workflow 1: This workflow is suitable for applications that require robust performance and high effectiveness, for applications involving data with a highly heterogeneous schema and for users who want try a pre-configured solution without any manual intervention.
   • Workflow 2: This workflow is suitable for application that emphasize time efficiency at the cost of slightly lower effectiveness, provided that the schema of the data allows for successfully applying similarity joins. It requires manual parameter fine-tuning and good knowledge of the data at hand, which is facilitated by JedAI's data exploration functionality.
   • Workflow 3: This workflow is suitable for applications with restricted temporal or computational resources, as it can maximize the number of detected matches in these settings.
2. deployed to maven central

As described in "The return of JedAI: End-to-End Entity Resolution for Structured and Semi-Structured Data", PVLDB, 11 (12): 1950 - 1953, 2018 (https://www.vldb.org/pvldb/vol11/p1950-papadakis.pdf).

Version 1.3 of JedAI-core features GNU Trove in all workflow steps, thus minimizing the memory footprint and the running time of all supported methods. It also contains documentation for every method and configuration parameter as well as a main method for running JedAI-core through the command line. Several bugs have been fixed, too.

The new version of JedAI Toolkit fixes few bugs of v1.0, but most importantly includes an updated GUI, while the JedAI library has been decoupled from Lucene and has been mavenized.

This release concludes the first round of development of JedAIToolkit. It includes the first version of the desktop application (jedaiDesktopApp-1-0.jar) and the first version of the open-source Java library (jedaiLibrary-1-0.jar). In total, both components support 6 data formats, 8 methods for Block Building, 4 for Block Cleaning, 7 for Comparison Cleaning, 3 for Entity Matching and 7 for Entity Clustering.