Date: Aug. 8, 2020
Source: College of Information Science and Technology / College of Cyber Security

A paper titled An Extensible Block Scheme-Based Method for Entity Matching, led by undergraduates from JNU's College of Cyber Security under the supervision of Prof. Zhang Jilian, has been accepted for publication in the Data Integration to Knowledge Graphs (DI2KG) workshop of VLDB 2020. The three first authors, Wang Jiawei, Ye Haizhou and Huang Jianhui, are undergraduates in the 2017 class. Their team, JNU_Cyber, also won the Entity Matching Challenge.
 

Established in 1975, VLDB – for Very Large Data Bases -- is a premier annual international industrial forum for data management, database researchers, vendors, practitioners, application developers and users. The forum published important research on topics including expert systems, query optimization, SQL query language, rapid association rule mining, NoSQL column storage and location-based services.
The paper, focusing on entity-matching that applies to fields including  e-commerce, natural language processing and multi-source data matching, comes up with a more extendable and effective method for identifying identical data entities based on descriptions in a database. The research results can be widely used in e-commerce, recommendation systems, natural language processing and other practical situations.