Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Makenna Pradeep PDF A APPROVED.pdf (385.75 KB)

ETD Abstract Container

Abstract Header

Predicting Closed Versus Open Questions Using Machine Learning for Improving Community Question Answering Websites

Makkena, Pradeep Kumar
http://rave.ohiolink.edu/etdc/view?acc_num=ysu1516375999820403

Abstract Details

2017, Master of Computing and Information Systems, Youngstown State University, Department of Computer Science and Information Systems.
Community question answer (CQA) websites add great value to the information available on the Web and they have been gaining popularity in the past few years. Popular CQA websites have millions of users asking thousands of questions every day. To maintain the quality of content, site moderators monitor and close the questions which do not follow the community guidelines. Stack Overflow is a very popular CQA website for programmers with more than 8 million users. Every day thousands of questions are posted on Stack Overflow and some of these questions do not follow the community guidelines and they will be closed by the moderators. Manual moderation of questions is a tedious task because of the sheer volume of the questions. The main purpose of this thesis is to build a machine learning classifier to predict whether or not a question will be closed. Using various post features, comment features and user features from Stack Overflow data, machine learning models were created using various classification algorithms. Apache Spark machine learning library was used to train and test these models and we found that model trained with random forest algorithm gave the best results with an accuracy of 77.60%. We found that features derived from the body of the post contribute more to the accuracy of the model whereas features derived from the user's table contribute less.
Alina Lazar, PhD (Advisor)
Yong Zhang, PhD (Committee Member)
Feng Yu, PhD (Committee Member)
35 p.
Computer Science; Information Systems
Predicting closed questions; Stack overflow closed questions; Classification using Apache Spark

Recommended Citations

Citations

  • Makkena, P. K. (2017). Predicting Closed Versus Open Questions Using Machine Learning for Improving Community Question Answering Websites [Master's thesis, Youngstown State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ysu1516375999820403

    APA Style (7th edition)

  • Makkena, Pradeep Kumar. Predicting Closed Versus Open Questions Using Machine Learning for Improving Community Question Answering Websites . 2017. Youngstown State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ysu1516375999820403.

    MLA Style (8th edition)

  • Makkena, Pradeep Kumar. "Predicting Closed Versus Open Questions Using Machine Learning for Improving Community Question Answering Websites ." Master's thesis, Youngstown State University, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=ysu1516375999820403

    Chicago Manual of Style (17th edition)

ysu1516375999820403
831
© 2017, some rights reserved.Creative Commons License Predicting Closed Versus Open Questions Using Machine Learning for Improving Community Question Answering Websites by Pradeep Kumar Makkena is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by Youngstown State University and OhioLINK.