Loading...

Automating Moderators’ Actions in Online Question-Answering Communities

Annamoradnejad, Eisa | 2022

302 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 55208 (19)
  4. University: Sharif University of Technolog
  5. Department: Computer Engineering
  6. Advisor(s): Habibi, Jafar; Fazli, Mohammad Amin
  7. Abstract:
  8. Online question-answering communities, as reliable sources for exchanging experts' opinions, have specific rules to maintain their content quality. Due to their large number of users and posts, manual control and approval by administrators is not plausible, and these systems require solutions that are more scalable. The current dominant solution, i.e., the use of crowdsourcing and relying on user reports, has serious problems, including the slow speed of handling violations, the waste of time of users, and the discouraging feedback from the community towards new users. Although the automation of moderation actions via artificial intelligence methods would solve the existing problems, the previous efforts failed to replace crowdsourcing. By examining the advantages and disadvantages of previous works, a six-component process is proposed to classify questions based on their accordance with user guidelines. The process, which contains three feature extraction components, is only dependent on the textual content and does not use features from user profiles or community feedback. In the design and modeling of the process, several state-of-the-art artificial intelligence approaches were utilized to increase the accuracy and speed of the proposed solution. The evaluation results for classifying questions based on their quality show precision of between 79 and 92 percent. The results for the F1-score metric indicate a 9–19% improvement compared to the baseline models and a 7–16% increase as a direct result of the feature extraction components related to the system context and questions’ subjective features. Q&A platforms can utilize the proposed solution as a violation detection system, a quality assessment system before posting a new question, or a training system for new users. Future research can use the proposed process as well as the feature extraction techniques to solve other content moderation tasks.
  9. Keywords:
  10. Crowdsourcing ; Expert System ; Safware Evolution ; Question-Answer Sites ; Moderation Actions