Abstract | ||
---|---|---|
User reviews, containing a wealth of user opinion information, play an important role for product’s online word of mouth, which have great reference value for potential customers and service/product providers. But the problem of information overload caused by the massive reviews makes users difficult to find high-quality reviews effectively. Most current methods of evaluating review quality focus on review’s content. However, the reviewer’s expertise also has a positive effect on evaluation of review’s quality. In this paper, we propose a new method to rank the reviews according to their quality. Firstly, reviewer’s quality of special topic is measured based on his/her historical review data with a topic model. Then, the coverage of attributes described in review content are integrated to measure the review’s quality based on a learning to rank model. A series of experiments are implemented on a real world dataset to verify the proposed method’s effectiveness. |
Year | Venue | Field |
---|---|---|
2018 | DASFAA Workshops | Learning to rank,Information overload,Information retrieval,Computer science,Word of mouth,Topic model,Database |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
12 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ju Zhang | 1 | 6 | 7.56 |
Yuming Lin | 2 | 37 | 4.76 |
Taoyi Huang | 3 | 0 | 1.01 |
You Li | 4 | 5 | 4.64 |