An algorithm for analyzing personalized online commercial intention

  • Authors:

  • Derek Hao Hu;Qiang Yang;Ying Li

  • Affiliations:

  • Hong Kong University of Science and Technology;Hong Kong University of Science and Technology;Microsoft Corporation, One Microsoft Way, Redmond, WA

  • Venue:
  • Proceedings of the 2nd International Workshop on Data Mining and Audience Intelligence for Advertising
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

With more and more commercial activities moving onto the Internet, people tend to purchase what they need through Internet or conduct some online research before the actual deals happen. For many Web users, their online commercial activities start from submitting a search query to search engines. Just like the common Web search queries, the queries with commercial intention are usually very short. Recognizing the queries with commercial intention against the common queries will help search engines provide proper search results and advertisements; help Web users obtain the right information they desire and help the advertisers benefit from the potential transactions. The only existing research work, as far as we know, has been done to automatically detect online commercial intention purely based on the issued queries, without considering the Web user's information. However, the intentions behind a query vary a lot for users with different background and interest. The intentions can even be different for the same user, when the query is issued in different contexts. In this paper, we present a novel algorithm, which we name as POINT, for the Personalized Online-commercial INTention detection based on a skip-chain conditional random field model. To accurately detect the commercial intentions of a query, our method comprehensively considers the evidences from the target query, the profile of the user issuing the query, which is inferred from his search history, as well as the similarity of different queries in a personal query log. Our proposed method is validated through extensive experiments on a real search engine query log data set. The experimental results show that our algorithm can clearly improve the performance by more than 10% of personalized online-commercial intention detection.