University of Twente Student Theses

Login

Augmenting the process of schema matching with machine learning-based Intelligence Amplification

Boerrigter, T.H. (2021) Augmenting the process of schema matching with machine learning-based Intelligence Amplification.

[img] PDF
5MB
Abstract:In this thesis an approach is explored, set up and validated for creating a human-computer relationship in the process of schema matching. This relationship is known as Intelligence Amplification, and serves to establish a 1 + 1 > 2 situation. Schema matching in this research is used in system integration, where matches have to be created between source and destination elements. Our model tries to predict as accurately as possible matches between different entities and attributes. Prediction is based on a Deep Neural Network, which is set up suited to the context of the company of eMagiz. When predictions are accurate and fast enough, it should be beneficial for the human schema matcher time wise. With this, repetitive work can be prevented, or exploration of schemas can be set in motion. The outcomes of this thesis represent a tuned semi automatic schema matcher of which its performance is known in the different matching situations. Therefore, it is also known when the schema matcher is beneficial.
Item Type:Essay (Master)
Clients:
eMagiz Services b.v., Enschede, The Netherlands
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:50 technical science in general, 54 computer science
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:https://purl.utwente.nl/essays/88064
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page