The advent of Web 2.0 enabled users to actively contribute content online through user reviews, blogs, and social media posts. This user-generated content contains valuable information that captures the experiences and opinions of individual users. Online reviews in particular have become an indispensable part in the decision process of online users. There are several challenges in analyzing review content and extracting useful knowledge. In [1,2] we consider the problem of selecting useful sets of reviews. In [3] we mine jointly reviews and micro-reviews (tips), and we consider the problem of synthesizing a review that summarizes a collection of micro-reviews. In [4] we consider the problem of summarizing micro-reviews for a collection of entities, such as, the restaurants in a specific neighborhood. In [5] we perform prediction of the topic of a future review using sentiment information. Reviews can also be used for comparative evaluation, extraction of aspects and facets, and answering user questions.
User-generated content includes also content produced on social media platforms such as Twitter or Reddit. This content is more topical and time-sensitive and can be mined to extract information for a variety of tasks. In [6] we studied the information in the bio’s of Twitter users, and how they can be used for friendship recommendations. In [7] we considered the problem of assessing the vulnerability of Reddit users to become targets of trolling, or of aggression.
[1] Panayiotis Tsaparas, Alexandros Ntoulas, Evimaria Terzi. Selecting a comprehensive set of reviews. KDD 2011
[2] Thanh-Son Nguyen, Hady Wirawan Lauw, Panayiotis Tsaparas. Review Selection Using Micro-Reviews. IEEE Trans. Knowl. Data Eng. 27(4): 1098-1111 (2015)
[3] Thanh-Son Nguyen, Hady Wirawan Lauw, Panayiotis Tsaparas. Review Synthesis for Micro-Review Summarization. WSDM 2015
[4] Thanh-Son Nguyen, Hady Wirawan Lauw, Panayiotis Tsaparas. Micro-review synthesis for multi-entity summarization. Data Min. Knowl. Discov. 31(5): 1189-1217 (2017)
[5] Ziyu Lu, Nikos Mamoulis, Evaggelia Pitoura, Panayiotis Tsaparas. Sentiment-Based Topic Suggestion for Micro-Reviews. ICWSM 2016
[6] Konstantinos Semertzidis, Evaggelia Pitoura, Panayiotis Tsaparas. How people describe themselves on Twitter. DBSocial 2013
[7] Paraskevas Tsantarliotis, Evaggelia Pitoura, Panayiotis Tsaparas. Defining and predicting troll vulnerability in online social media. Social Netw. Analys. Mining 7(1): 26:1-26:15 (2017)