The present study is aiming to automatically generate news summaries using illustrated diagram that enables users to understand an overview of the news. To generate illustrated diagram, zero anaphora resolution is needed because the elements of the statements are often omitted in Japanese sentences. In this study, we propose a method for zero anaphora resolution focusing on human nouns that appear in diagram based on the centering theory, and a method for hierarchical management of illustrated diagram using the verb thesaurus. We conducted an experiment to compare summaries generated by the system and summaries of existing conventional services. The experimental result indicated that the illustrated diagram is useful to understand the overview.

We propose the new analysis method of Twitter, which uses 3 types parameters in single Tweet. We show relations of Topics-User-Trend visually based on Hashtag. System contains 3 types Self-Organizing Map used for visualizing. We operate maps and analyze interactively.

Document polishing is indispensable to create easy-to-read text. Hence, in this paper, we defined document polishing skills as to find the problems in the document and to consider modified policy. Then we proposed a tutorial system which can acquire document polishing skills by using TETDM and showed the effect of the system by experiment. According to experiments, users of the system was able to polish document not only from the local perspective but also from global perspective.

Elementary, junior high or high school students are required to understand way of thinking against questions that involve calculation in the process of solution. Students who know formula only have to remember all formula to solve various questions and cannot have ability to apply one to similar questions. In this study, a learning system that assists establishing way of thinking against percentage and speed questions by intuitive understanding iterations is proposed. According to experimental results, the proposed system that emphasis on way of thinking was effective to learning and its radication rather than the comparative system that emphasis on solution by formula.

I. It is possible to visualize the function of Japanese particle topic marker 'wa' and subject marker 'ga' by using graphs. These graphs help us understand the role of 'wa' and 'ga'. And they show the difference between Exhaustive listing and Neutral description. II. People or computers can gather information as 2-dimensional arrays by hearing many sentences of the type 'A wa B ga C'. By using the arrays, they can say the sentences of the type 'A wa B ga C', give the information to the others.

In this paper, we aim to develop an exploratory tweet-browsing system to analyze local reputation trends in geographical regions. We implemented functions for mapping the local reputation information, for extracting local feature words, and for exploratory browsing of local tweets. We conducted an experiment to evaluate whether users can extract local reputation trends about the snap election in 2014 by using our system. The experimental result indicated that users can extract local reputation trends about political topics in a particular prefecture by using our system. After this experiment, we also conducted a questionnaire survey about the usability of our system.