KLASS – Entity Resolution

MATCH TAK MATCH

Different databases utilize unique formats for storing data entries, leading to the presence of numerous data entries pertaining to the same individual across these databases. These entries often exhibit discrepancies in the data they contain. Hence, our objective is to devise an efficient algorithm capable of resolving these entities, distinguishing which entries correspond to the same real-world entity.

Introducing KLASS – Entity Resolution

Our project utilizes advanced artificial intelligence (AI) techniques to perform entity resolution. Through sophisticated machine learning models, we automate the identification and consolidation of duplicated entities across diverse datasets. Through a web app, our project resolves all entries of a particular entity across multiple datasets based on certain entity attributes.

Team members

Tan Rui Yang (DAI), Seetoh Jian Qing (DAI), Cheh Kang Ming (CSD), Goo Jun Xi (CSD), Alphonsus Tan Yong Shing (CSD), Ng Jie Lin (ESD)

Instructors:

Sudipta Chattopadhyay

Writing Instructors:

Bernard Tan Chee Seng

What is Entity Resolution

Entity Resolution is a technique that uses blocking and matching algorithms to identify data entries across multiple datasets that refer to the same real-world entity.

The blocking algorithm serves to reduce the search space by grouping similar entries together into blocks.

Matching is then performed within a subset of blocks to match and search the desired entity.

Blocking

In the process, relevant attributes from each entity are selected to create blocking keys, which consist of combinations of informative attribute values. Entities with identical or similar blocking keys are grouped together in the same block for subsequent comparison during the matching phase.

Matching

Matching is employed to assess the likeness or similarity between all pairs of entities identified as potential matches by the Blocking process, which serves as its input. The matching model subsequently evaluates and calculates a similarity score for each candidate pair provided.

How to use our application

3 Simple Steps~

In partnership with :

Acknowledgements

Special thanks to KLASS Engineering and Solutions for their continuous support and giving us the opportunity to work on this project.

Days

Hours

Minutes

Seconds

Vote the project now!

Vote for our project at the exhibition! Your support is vital in recognizing our creativity. Join us in celebrating innovation and contributing to our success. Thank you for being part of our journey!

Team members :

Tan Rui Yang

Design and Artificial Intelligence (DAI)

Seetoh Jian Qing

Design and Artificial Intelligence (DAI)

Cheh Kang Ming

Computer Science and Design (CSD)

Goo Jun Xi

Computer Science and Design (CSD)

Alphonsus Tan Yong Shing

Computer Science and Design (CSD)

Ng Jie Lin

Engineering Systems and Design (ESD)