Editing Messaging spam (section)

==In opinion-based recommender systems==
{{See also|Recommender system}}In an opinion based [[recommender system]], an important concern is how to evaluate the user-generated reviews on the items. One of the purpose of this evaluation is to identify malicious or spam reviews. Poorly written reviews are considered helpless to the recommender system. However, even if a review is well generated, they can still be harmful to the recommender system by their biased prejudice to form an actual advertisement or slander towards a target item.

Current approach of spam detection methods includes analyzing the spam text and identifying the spam reviewers by their reviews and activities. For the first kind, a [[machine learning]] application on review text has been developed.<ref>Li, Fangtao, et al. "Learning to identify review spam." ''IJCAI Proceedings-International Joint Conference on Artificial Intelligence''. Vol. 22. No. 3. 2011.</ref> For the second kind, researchers use network motif analysis technique to identify spam reviewers by their recurring reviewing activity.<ref>O'Callaghan, Derek, et al. "Network analysis of recurring YouTube spam campaigns." ''arXiv preprint arXiv:1201.3783'' (2012).</ref>