Automating price matching on e-commerce websites using natural language processing : a postgraduate project dissertation presented in partial fulfilment of the requirements for the degree of Masters in Information Technology at Massey University, Auckland, New Zealand

Abstract

With the development of internet, shopping online has become an important part in our daily life. Global B2C e-commerce turnover grew by 24.0% to reach 1,943 billion dollars in 2014. Not only customers need to face a great amount of information while shopping online, the companies also need to catch the information from their competitors. There is a case which a company wanted to realize was a simple way for them to monitor the prices of equivalent products on competitor’s website. Base on the development of the E-commerce platform, after analyzing the requirement of companies and customer, we propose a frame of E-commerce website data extraction, data storage and production matching. We build up a customized web crawler to crawl the production on E-commerce website and extract the production detail for matching. Finally we got average 87.18% matching rate after applying enhance TF/IDF algorithm with weight adjustment.