Biblio

Community question answering (CQA) websites have become very popular platforms attracting numerous participants to share and acquire knowledge and information in Internet However, with the rapid growth of crowdsourcing systems, many malicious users organize collusive attacks against the CQA platforms for promoting a target (product or service) via posting suggestive questions and deceptive answers. These manipulate deceptive contents, aggregating into multiple collusive questions and answers (Q&As) spam groups, can fully control the sentiment of a target and distort the decision of users, which pollute the CQA environment and make it less credible. In this paper, we propose a Pattern and Burstiness based Collusive Q&A Spam Detection method (PBCSD) to identify the deceptive questions and answers. Specifically, we intensively study the campaign process of crowdsourcing tasks and summarize the clues in the Q&As’ vocabulary usage level when collusive attacks are launched. Based on the clues, we extract the Q&A groups using frequent pattern mining and further purify them by the burstiness on posting time of Q&As. By designing several discriminative features at the Q&A group level, multiple machine learning based classifiers can be used to judge the groups as deceptive or ordinary, and the Q&As in deceptive groups are finally identified as collusive Q&A spam. We evaluate the proposed PBCSD method in a real-world dataset collected from Baidu Zhidao, a famous CQA platform in China, and the experimental results demonstrate the PBCSD is effective for collusive Q&A spam detection and outperforms a number of state-of-art methods.

We transfer a key idea from the field of sentiment analysis to a new domain: community question answering (cQA). The cQA task we are interested in is the following: given a question and a thread of comments, we want to re-rank the comments, so that the ones that are good answers to the question would be ranked higher than the bad ones. We notice that good vs. bad comments use specific vocabulary and that one can often predict the goodness/badness of a comment even ignoring the question, based on the comment contents only. This leads us to the idea to build a good/bad polarity lexicon as an analogy to the positive/negative sentiment polarity lexicons, commonly used in sentiment analysis. In particular, we use pointwise mutual information in order to build large-scale goodness polarity lexicons in a semi-supervised manner starting with a small number of initial seeds. The evaluation results show an improvement of 0.7 MAP points absolute over a very strong baseline, and state-of-the art performance on SemEval-2016 Task 3.

Community question answering (cQA) has become an important issue due to the popularity of cQA archives on the Web. This paper focuses on addressing the lexical gap problem in question retrieval. Question retrieval in cQA archives aims to find the existing questions that are semantically equivalent or relevant to the queried questions. However, the lexical gap problem brings a new challenge for question retrieval in cQA. In this paper, we propose to model and learn distributed word representations with metadata of category information within cQA pages for question retrieval using two novel category powered models. One is a basic category powered model called MB-NET and the other one is an enhanced category powered model called ME-NET which can better learn the distributed word representations and alleviate the lexical gap problem. To deal with the variable size of word representation vectors, we employ the framework of fisher kernel to transform them into the fixed-length vectors. Experimental results on large-scale English and Chinese cQA data sets show that our proposed approaches can significantly outperform state-of-the-art retrieval models for question retrieval in cQA. Moreover, we further conduct our approaches on large-scale automatic evaluation experiments. The evaluation results show that promising and significant performance improvements can be achieved.