Back to Search Start Over

Item Matching using Text Description and Similarity Search

Authors :
Appel, Ana Paula
Silva, Anderson Luis de Paula
Silva, Adriana Reigota
Santos, Caique Dutra
da Silva, Thiago Logo
de Araujo, Rafael Poggi
de Aquino, Luiz Carlos Faray
Publication Year :
2022

Abstract

In this paper, we focus on the problem of item matching using only the description. Those specific items not only lack a unique code but also contain short text descriptions, making the item matching process difficult. Our goal is to compare products using only the description provided by the purchase process. Therefore, evaluating other characteristics and differences can uncover possible flaws during the acquiring phase. However, the text of the items that we were working on was very small, with numbers due to the nature of the products and we have a limited amount of time to develop the solution which was 8 weeks. As result, we showed that working using a well-oriented methodology we were able to deliver a successful MVP and achieve the results expected with up to 55% match.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2206.14097
Document Type :
Working Paper