Detecting collusive spamming activities in community question answering

Community Question Answering (CQA) portals provide rich sources of information on a variety of topics. However, the authenticity and quality of questions and answers (Q&As) has proven hard to control. In a troubling direction, the widespread growth of crowdsourcing websites has created a large-s...

Full description

Bibliographic Details
Main Authors: Liu, Yuli, Liu, Yiqun, Zhou, Ke, Zhang, Min, Ma, Shaoping
Format: Conference or Workshop Item
Published: 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/45045/
_version_ 1848797056087359488
author Liu, Yuli
Liu, Yiqun
Zhou, Ke
Zhang, Min
Ma, Shaoping
author_facet Liu, Yuli
Liu, Yiqun
Zhou, Ke
Zhang, Min
Ma, Shaoping
author_sort Liu, Yuli
building Nottingham Research Data Repository
collection Online Access
description Community Question Answering (CQA) portals provide rich sources of information on a variety of topics. However, the authenticity and quality of questions and answers (Q&As) has proven hard to control. In a troubling direction, the widespread growth of crowdsourcing websites has created a large-scale, potentially difficult-to-detect workforce to manipulate malicious contents in CQA. The crowd workers who join the same crowdsourcing task about promotion campaigns in CQA collusively manipulate deceptive Q&As for promoting a target (product or service). The collusive spamming group can fully control the sentiment of the target. How to utilize the structure and the attributes for detecting manipulated Q&As? How to detect the collusive group and leverage the group information for the detection task? To shed light on these research questions, we propose a unified framework to tackle the challenge of detecting collusive spamming activities of CQA. First, we interpret the questions and answers in CQA as two independent networks. Second, we detect collusive question groups and answer groups from these two networks respectively by measuring the similarity of the contents posted within a short duration. Third, using attributes (individual-level and group-level) and correlations (user-based and content-based), we proposed a combined factor graph model to detect deceptive Q&As simultaneously by combining two independent factor graphs. With a large-scale practical data set, we find that the proposed framework can detect deceptive contents at early stage, and outperforms a number of competitive baselines.
first_indexed 2025-11-14T19:57:48Z
format Conference or Workshop Item
id nottingham-45045
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:57:48Z
publishDate 2017
recordtype eprints
repository_type Digital Repository
spelling nottingham-450452020-05-04T18:40:55Z https://eprints.nottingham.ac.uk/45045/ Detecting collusive spamming activities in community question answering Liu, Yuli Liu, Yiqun Zhou, Ke Zhang, Min Ma, Shaoping Community Question Answering (CQA) portals provide rich sources of information on a variety of topics. However, the authenticity and quality of questions and answers (Q&As) has proven hard to control. In a troubling direction, the widespread growth of crowdsourcing websites has created a large-scale, potentially difficult-to-detect workforce to manipulate malicious contents in CQA. The crowd workers who join the same crowdsourcing task about promotion campaigns in CQA collusively manipulate deceptive Q&As for promoting a target (product or service). The collusive spamming group can fully control the sentiment of the target. How to utilize the structure and the attributes for detecting manipulated Q&As? How to detect the collusive group and leverage the group information for the detection task? To shed light on these research questions, we propose a unified framework to tackle the challenge of detecting collusive spamming activities of CQA. First, we interpret the questions and answers in CQA as two independent networks. Second, we detect collusive question groups and answer groups from these two networks respectively by measuring the similarity of the contents posted within a short duration. Third, using attributes (individual-level and group-level) and correlations (user-based and content-based), we proposed a combined factor graph model to detect deceptive Q&As simultaneously by combining two independent factor graphs. With a large-scale practical data set, we find that the proposed framework can detect deceptive contents at early stage, and outperforms a number of competitive baselines. 2017-04-03 Conference or Workshop Item PeerReviewed Liu, Yuli, Liu, Yiqun, Zhou, Ke, Zhang, Min and Ma, Shaoping (2017) Detecting collusive spamming activities in community question answering. In: 26th International Conference on World Wide Web, 3-7 April 2017, Perth, Australia. Community Question Answering; Crowdsourcing Manipulation; Spam Detection; Factor Graph https://doi.org/10.1145/3038912.3052594 10.1145/3038912.3052594 10.1145/3038912.3052594 10.1145/3038912.3052594
spellingShingle Community Question Answering; Crowdsourcing Manipulation; Spam Detection; Factor Graph
Liu, Yuli
Liu, Yiqun
Zhou, Ke
Zhang, Min
Ma, Shaoping
Detecting collusive spamming activities in community question answering
title Detecting collusive spamming activities in community question answering
title_full Detecting collusive spamming activities in community question answering
title_fullStr Detecting collusive spamming activities in community question answering
title_full_unstemmed Detecting collusive spamming activities in community question answering
title_short Detecting collusive spamming activities in community question answering
title_sort detecting collusive spamming activities in community question answering
topic Community Question Answering; Crowdsourcing Manipulation; Spam Detection; Factor Graph
url https://eprints.nottingham.ac.uk/45045/
https://eprints.nottingham.ac.uk/45045/
https://eprints.nottingham.ac.uk/45045/