Learning to rank with a Weight Matrix

Learning to rank, a task applying machine learning techniques to rank the expected information of the users, such as movie items users might be interested in. It is useful for collaborative filtering, which is regarded as a hot subfield of computer supported collaborative work(CSCW). In this paper, we propose an algorithm based on RankBoost to rank expected information of the users more accurately. The main advantage of the algorithm against RankBoost is to add a Weight Matrix regularizer to rank the relevance levels of the information smoothly and locally based on graph methods. The experimental results on the public LETOR datasets show that the proposed algorithm performs better than the baseline algorithm, indicating that the method is promising.

I. Introduction

Recently, people would like to obtain vast quantities of information from world wide web with little cost, which depends on the ranking results of the ranking models. Hence, ranking is crutial for information retrieval. Combined with machine learning techniques, ranking becomes a valuable research direction of information retrieval and is called learning to rank. Learning to rank applies machine learning techniques to train and obtain ranking functions to properly rank a set of documents for a given query. It has received increasing attention in Information Retrieval (IR) and machine learning research. T. Joachims, Y. Freund and C. Burges regarded ranking problems as pairwise classification problems, and solved the corresponding optimization problems based on different techniques, such as SVM, Boosting and Neural Networks[2], [3], [4].