TY - GEN
T1 - Finding hard questions by knowledge gap analysis in question answer communities
AU - Chen, Ying Liang
AU - Kao, Hung Yu
PY - 2010
Y1 - 2010
N2 - The Community Question Answer (CQA) service is a typical forum of Web2.0 in sharing knowledge among people. There are thousands of questions have been posted and solved every day. Because of the above reasons and the variant users in CQA service, the question search and ranking are the most important researches in the CQA portal. In this paper, we address the problem of detecting the question being easy or hard by means of a probability model. In addition, we observed the phenomenon called knowledge gap that is related to the habit of users and use knowledge gap diagram to illustrate how much knowledge gap in different categories. In this task, we propose an approach called knowledge-gap-based difficulty rank (KG-DRank) algorithm that combines the user-user network and the architecture of the CQA service to solve this problem. The experimental results show our approach leads to a better performance than other baseline approaches and increases the F-measure by a factor ranging from 15% to 20%.
AB - The Community Question Answer (CQA) service is a typical forum of Web2.0 in sharing knowledge among people. There are thousands of questions have been posted and solved every day. Because of the above reasons and the variant users in CQA service, the question search and ranking are the most important researches in the CQA portal. In this paper, we address the problem of detecting the question being easy or hard by means of a probability model. In addition, we observed the phenomenon called knowledge gap that is related to the habit of users and use knowledge gap diagram to illustrate how much knowledge gap in different categories. In this task, we propose an approach called knowledge-gap-based difficulty rank (KG-DRank) algorithm that combines the user-user network and the architecture of the CQA service to solve this problem. The experimental results show our approach leads to a better performance than other baseline approaches and increases the F-measure by a factor ranging from 15% to 20%.
UR - http://www.scopus.com/inward/record.url?scp=78650891552&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78650891552&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-17187-1_36
DO - 10.1007/978-3-642-17187-1_36
M3 - Conference contribution
AN - SCOPUS:78650891552
SN - 3642171869
SN - 9783642171864
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 370
EP - 378
BT - Information Retrieval Technology - 6th Asia Information Retrieval Societies Conference, AIRS 2010, Proceedings
T2 - 6th Asia Information Retrieval Societies Conference, AIRS 2010
Y2 - 1 December 2010 through 3 December 2010
ER -