DocumentCode :
3538532
Title :
Using Graph Analysis Approach to Support Question & Answer on Enterprise Social Network
Author :
Ke Ning ; Ning Li ; Liang-Jie Zhang
Author_Institution :
Kingdee Res. Kingdee Int. Software Group, Co. Ltd., Shenzhen, China
fYear :
2012
fDate :
6-8 Dec. 2012
Firstpage :
146
Lastpage :
153
Abstract :
Enterprise Social Network (ESN) service is getting more popular recently. It can help employees to communicate and collaborate efficiently with colleagues, with customers and with suppliers. One significant phenomenon happening on ESN is question & answer: people posting questions to the network to get answers from friends or friends-of-friends. However, existing ESN platforms do not have good support to this process. In this paper, we propose a method to better support question & answer on ESN, purely by using a graph analysis approach. Based on the questioner´s initial input list of potential answerers, it can extract a shared-interest group of people, whose interest is close to the initial list of potential answerers, and sort the group of people according to a score of interest distance, and then recommend them to the questioner. To evaluate its applicability, the method is implemented in KDWeibo the most popular ESN platform in China, and the results are promising.
Keywords :
business data processing; graph theory; question answering (information retrieval); recommender systems; social networking (online); China; ESN service; KDWeibo; enterprise social network; graph analysis approach; question-and-answer; score-of-interest distance; shared-interest group; Algorithm design and analysis; Blogs; Collaboration; Companies; Engines; Social network services; User interfaces; Enterprise Social Network; Graph Analysis; Question & Answer; Recommender System; Shared Interest;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Services Computing Conference (APSCC), 2012 IEEE Asia-Pacific
Conference_Location :
Guilin
Print_ISBN :
978-1-4673-4825-6
Type :
conf
DOI :
10.1109/APSCC.2012.36
Filename :
6478210
Link To Document :
بازگشت