返回Google圖書搜尋

Mining Email Corpora for Personalized Expert Recommenders

出版	Universitat Politècnica de Catalunya. Facultat d'Informàtica de Barcelona, 2010 (Enginyeria Informàtica (Pla 2003)), 2010
URL	http://books.google.com.hk/books?id=iDFbMwEACAAJ&hl=&source=gbs_api

註釋In every organization there is the necessity to collaborate with people and establish interpersonal relationships. The contact with the right person is crucial for the success of the performed daily tasks. Additionally, personal email corpora contain rich information about all the people that the user knows and about their activities. This fact together with the potential of Social Networks, allows the construction of the real image that surrounds the user. This thesis analyses the importance of searching for collaborators and aims to develop ExpertSN, a personalized Expert Recommender tool based on email Data Mining and Social Network Analysis. The design of ExpertSN started with the analysis of two real use cases and the identification of requirements. According to these requirements, our recommender tool was specified and implemented. As a result, the main contributions of this research work include the differentiation between the different search types people make, the development of appropriate measures to qualify relationships between people and the design of the Social Network Query Language, which gives users a way to express their necessities within the framework of task management, collaboration and personal relationships.