Recently, exchange of opinions with service on the online is performed widely. However, in the exchange of opinions that there is the unspecified number of participant, there are the problems that opinions to lack sense of cooperation increase. Therefore, we use unselfish contribution degree to let the participant of the exchange of opinions be conscious of another person and group total profit. We suggest the exchange of opinions support system which promotes the input of the cooperative opinion by doing so it.

These days information propagation method that transmits information to many people is proposed. Many previous methods regard the network structure as important. However, these methods decide the quantity of information propagation without sufficiently considering a conversation topic of users. Therefore, we propose quantification method of information propagation considering communication between users in Social Network. We verified experimental results in twitter data.

On user generated contents sites, user's comments include the user's impression to the content. The authors believe that user's comments can be used as the data mining resource to evaluate the contents. In this paper, the authors focus on bilibili.tv, a Chinese video sharing site, and analyze user's emotional comments on the site. They also show co-relation between comments and popularity such as number of replay and bookmark.

This paper proposes a system that intends to facilitate inducement of a user's interest by showing various types of information related to photographs. Our proposed system focuses on meta-data (e.g., date, place) attached to each information and uses them to link with another information. For instance, this system connects a news article and music with a photograph by using meta-data on "date." Similarly, the system connects two news articles in which a common named entity appears are connected each other, or two songs which the common singer has sung are connected each other. With the system, we conducted a user study to observe a user's behavior under unrestrained information exploration task. Based on the results obtained, we found that the user's knowledge and interests influence the difference in the frequency of access to information that has different modality, and that every user often access to information that has common date.

In this paper, we propose and discuss extraction of term explanations depending on context. When we read some documents, we encounter some unknown terms. Although there are various explanations of a term, it is important to give users good explanations suitable for their context for understanding the unknown terms. We pay attention to brace expressions which appears frequently, and study those expressions in explanation of them. We also discuss a system that extracts explanations for terms from the Web.

The goal of our research is to support people who look for a present using E-commerce websites. When choosing presents, we need to consider his or her tastes and interests and explore many candidates with a trial and error manner to pick the appropriate one. This paper proposes a system that shows a connection between a variety of genres to facilitate such exploration. We conducted an experiment to confirm performances of the system. As a result, we found that exploration time and the number of explored items increased. The participants repeated a trial and error with a wider range of goods in comparison to a conventional system.

Knowledge discovery on text mining requires a trial-and-error process so that a user's informational requirements are unclear when they start his or her exploration. Our purpose is to support a user's information seeking behaviour on text mining. In this paper, we observe how a user behave on TETDM: Total Environment for Text Date Mining. According to the experiment, we found that 5 usability problems and 1 problem for TETDM. From obtained results, we sort the system requirements and propose that a design criteria to facilitate a user's information seeking.

In TETDM challenge, we develop and distribute an integrated environment to flexibly combine multiple text mining techniques. Text mining techniques include numerous tasks such as salient sentence extraction, keyword extraction, topic extraction, textual coherence evaluation, multi-document summarization, and text clustering. Although tools that individually perform one or more of the above-mentioned tasks exist, it is difficult to integrate and activate multiple tools for a particular task. Therefore, TETDM is suggested as a system that can integrate numerous tools and can contribute to our creative work. In this paper, framework of knowledge creation support with TETDM is described. Knowledge creation consists of "trials and errors" and "logical interpretations of the facts".