BOOTSTRAP: Mining of Comparable Things
Kanchan Kundgir, Poonam Dere, Surabhi Mohite, Priti Mithari · 2015
Abstract — Making Comparisons between things is a typical part of human decision making process. But however, it is difficult to know what are to be compared and what can be the alternatives. For e.g., if someone is interested in certain products such as digital cameras, then he /she would want to know what the alternatives are and compare different cameras before making any purchase. This type of comparison activity is very common in our daily life but requires high knowledge skill in order to make much better choice. Therefore, to address this difficulty, we are presenting a system to automatically mine comparable entities from comparative questions that users posted online. In this paper, we focus on finding a set of comparable entities provided a user’s input entity. For example, provided an entity like Nokia N95 (a mobile phone), we want to find comparable entities such as Nokia N82, Blackberry and so on. To ensure high precision and high recall, we are developing a system that uses weakly-supervised bootstrapping method for comparative question identification and comparable entity extraction by leveraging a large online question archive. Our system calculates the precision and recall for a particular session depending on the correct comparators suggested. The result varies depending on the sessions. The results will prove to be very useful in helping users ’ exploration of alternative choices by suggesting comparable entities based on other users ’ prior requests.