Rule-Based Entity Resolution using Distinct Tree
Mitty Abraham1, Safiya K.M2
1Mitty Abraham, PG Scholar, Department of Computer Science and Engineering, Ilahia College of Engineering and Technology, MVPA, (Kerala), India.
2Safiya K.M, Assistant Professor, Department of Computer Science and Engineering, Ilahia College of Engineering and Technology, MVPA, (Kerala), India.
Manuscript received on 13 October 2015 | Revised Manuscript received on 22 October 2015 | Manuscript Published on 30 October 2015 | PP: 44-47 | Volume-5 Issue-5, October 2015 | Retrieval Number: E2208105515/15©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Entity resolution identifies object referring to the same entity .Entity resolution is performed by generating rules from training set and applies them on records. Traditional ER considered each attribute value as the rule in a random fashion and performs conjunction with other rules according to length threshold .This method is very complex and tedious. Our proposed method generated rules from a distinct tree using RL method, which consider the length criteria and RNN methods which does not. Distinct tree is formed by arranging attribute and its value of records in the training set in a particular fashion .These generated rules are applied to the dataset for entity identification .Our experimental results show that the proposed method is more accurate.
Keywords: Entity Resolution, Length Criteria.
Scope of the Article: Search-Based Software Engineering