Research Seminar 9/5: “A combinatorial approach to entity matching for products”

Research Seminar 9/5: “A combinatorial approach to entity matching for products”

School of Science and Technology
International Hellenic University

Thursday 9 May 2019

International Hellenic University, Lecture Room B1

Seminar Title

A combinatorial approach to entity matching for products”

Dr Leonidas Akritidis

Speaker information:
Leonidas Akritidis is a post-doctoral researcher at the Data Structuring and Engineering (DaSE) Lab of the Department of Electrical & Computer Engineering, University of Thessaly, Greece. He is also an MSc studies instructor in the same Department, teaching Data Structures, Algorithms and World Wide Web technologies. He obtained his PhD degree in 2013, and his BSc from the Department of Electrical & Computer Engineering of the Aristotle University of Thessaloniki, Greece, in 2003. His research interests include Data Mining, Machine Learning, Large-scale Data Processing, Big Data Engineering, and Information Retrieval.

Presentation at a glance:
The continuous growth of the e-commerce industry has rendered the problem of product retrieval particularly important. As more enterprises move their activities on the Web, the volume and the diversity of the product-related information increase quickly. These factors make it difficult for the users to identify and compare the features of their desired products. Recent studies proved that the standard similarity metrics cannot effectively identify identical products, since similar titles often refer to different products and vice-versa. Other studies employed external data sources (search engines) to enrich the titles; these solutions are rather impractical since the process of fetching external data is inefficient. In this presentation we will review the state-of-the-art approaches to entity matching and we will introduce UPM, an unsupervised algorithm for matching products by their titles. UPM is independent of any external sources and consists of three stages: during the first stage, the algorithm analyzes the titles and extracts combinations of words out of them. These combinations are evaluated in stage 2 according to several criteria, and the most appropriate of them are selected to form the initial clusters. The third phase is a post-processing verification stage which performs a refinement of the initial clusters by correcting the erroneous matches. This stage is designed to operate in combination with all clustering approaches, especially when the data possess properties which prevent the co-existence of two data points within the same cluster. We shall also present experimental results which demonstrate the superiority of the algorithm against multiple string similarity metrics and clustering methods.