In recent years, a large amount of information becomes available on computer networks such as World Wide Web. In World Wide Web, a lot of information is provided by a lot of information resources. The number of information resources is overwhelming, but the variety of information resources is also astonishing. Information resources varies in a wide range, i.e., from public database services to personal web pages. As a result, we bother to find important information from such various and massive information resources. In this research, we model an information resource on the web "Information collector/publisher" as not only gathering information on the web not just as an information publishing entity but an entity working both information publishing and information gathering. Such information entities are connected to each other in various levels, e.g. connected explicitly by links and human network of authors. Information gathering should be understood as an activity in such a community of information entities. We aim in this research to propose models and methods to enhance information sharing among people. Human is the most rich and exible information entity that can oer and gather information simultaneously. In order to achieve the goal, we set two topics. One is to identify relationship between information entities. In this topic, we first investigate relationship among people by analyzing WWW bookmarks as topic sharing, and showed that hierarchical structure is impor- tant to share topics among people. We also discuss how to identify relationship between large hierarchical information such as Yahoo! Internet Directory. The other is to handle network composed with information entities and their relations. We propose a method to re-congigure network to improve network structure. We also build a practical system in which human network is gathered and used in services for participants of conferences. In Chapter 1, we discuss about information sharing and show two important points of information sharing especially among heterogeneous information entities and show four researches about these two points in later chapters. In Chapter 2, we discuss and propose a method to help information sharing among heterogeneous information entities. We propose a system called kMedia that can assist users to form knowledge for community by showing shared topics networks (STN) among them. kMedia uses WWW bookmarks as information entities. We conducted an experiment to know how kMedia can support users. One result is that folder recommendation is more eective than page recommen- dation. The other is that recommendation is more eective for people belonging to the same real communities than those to dierent communities. According to these result, we propose a new measurement called â€category resemblance †that is recommendation measurement based on resemblance of folder structures. This measurement shows higher than all other system generated parameters and human evaluation to detect human relationship. In Chapter 3, we propose a method called WebHical. This method is for align- ing information from one information entities with hierarchical structure to an- other. It is based on kMedia method and Hical method. We adopted the statistic method and the SMART algorithm to measure the similarity among hi- erarchical structures. We construct a system to evaluate the performance of our method. The results of this experiment reveal that the proposed method can be used to help sharing information among information entities with dierent hierarchical structure. In Chapter 4, we propose an algorithm called Neighborhood Matchmaker Method to optimize networks of information entities. Interpersonal network is one of a network of information entities and it is useful in various utilization of informa- tion like information gathering. However it is usually formed locally and often independently. In order to adapt various needs for information utilization, it is necessary to extend and optimize it. Using the neighborhood matchmaker method, we can increase a new friend who is expected to share interests via all own neighborhoods on the interpersonal network. Iteration of matchmaking is used to optimize interpersonal networks. We simulate the Neighborhood Match- maker Method with the practical data and the random data and compare the results by our method with those by the central server model. The neighborhood matchmaker method can reach almost the same results obtained by the sever model with each type of data. In Chapter 5, we discuss importance and utilization of interpersonal network in a community system through the result of management and analysis of the scheduling support system for academic conferences. The important feature of the system is generation and utilization of interpersonal network to support in- formation exchanging and information discovery among participants. We applied this system to the academic conference called JSAI2003. We obtained 276 users and their interpersonal networks. We found not only that a lot of participants enjoyed to form interpersonal networks but also that the formed network was useful for them in information browsing and recommendation. Finally, we conclude this paper in Chapter 6. |