[in Japanese]

View full abstract

　In this paper, we report the examples of programming education in department of computer engineering at two colleges of different types. Because the social background to be required will be different greatly depending on college types, we set up the ideal goal by comparing the several education types. We also consider the evaluation method and the relationship between other curriculums, because the difference occurs even if using the same programming language for engineering education.

View full abstract

　Today it is becoming more demanding to measure and quantify activities and performance at universities. In this study, we focus on universities’ education activities and explore possible indexes mostly based on Daigaku Portraits, a database of basic information for national and public universities in Japan released in 2012. The results showed that the variables such as number of faculties or book circulation per student had positive correlations with their employment rate, while some other variables usually regarded as indexes of higher education negatively affected the employment rate. It is revealed that how to acquire the reliable data is important for measurement and assessment.

View full abstract

　After the Internet is widely used, new concept and information technology haves shown up; semantic web and linked open data (LOD). These technologies enable information on the Internet machine readable. In many science fields, it is pointed out that the semantic web will play an important role for the interdisciplinary research works. However, there have been few ideas to be ever proposed as a methodology or roadmap to the interdisciplinary science using semantic web. Herein we present a concept of professional knowledge and academic knowledge following collective knowledge proposed as a Web 2.0. Based on the concept, we design an application for interdisciplinary science.

View full abstract

　The authors have been working within "Japan SPI Consortium (JASPIC)" towards establishment of a knowledge structure for process improvement. Our objective is to store useful knowledge, to provide mechanisms for its usage, and to encourage creation of new knowledge. During our efforts to extract, categorize and consolidate knowledge from the presentation materials at the “SPI Japan” conferences organized by JASPIC, we saw problems with respect to the formalization of knowledge, especially its coverage. To solve this issue, we devised a knowledge model that consists of seven information elements as a representation of domain specific knowledge, and then introduced a standard template to be used for presentation abstract. As a result, statistically significant change in the distribution of knowledge coverage was observed, and the amount of descriptions increased for items such as causal analysis, verification and validation of process improvement. In terms of improvement for the quality of knowledge, positive feedbacks were obtained from proposal reviewers and conference participants for improved understandability.

View full abstract

　During idea creation, storing information and knowledge into a computer for visualization makes clear the relationship among pieces of information and enables us to get the whole story of the design. In this research, we focused on M Method which supports a multispace design process, and developed the prototypes of generating Scalable Vector Graphics images that represent M Method charts. We defined the data structure in consideration of the fact that such a chart has a tree structure under the containment relationships, and the prototypes compute the coordinates of rectangular regions in a recursive way.

View full abstract

　”Whiteboard wall(WBW)” is prevalent in the learning commons and group study rooms in Japan and the world. For example, WBW may be used to ask questions to generate an imagemap. However, WBW’s good evidence is not clear. So, first we investigated the intentions and use cases of WBW, we tryed to categorize how to use such lerning tool.

View full abstract

　The essence of Lattice Structure Model proposed by Fukunaga is considered to be the equation having golden number as solution. If we define the concept energy as unobservable enegy ruling the human’s motive force with a meaning, we can understand the upper equation as the complementarity which complete the transition from the old state in the material world to the new state in the material world using the linear storage of the new concept energy which will construct the new state making up for the diffusion of the old concept energy which will maintain the old state. Concept energy has complete-rational meaning attracting between same meanings and repulsing between different meanings. And attracting relation is marked in more shorter time-domain than golden number and repulsing relation in more longer time-domain. Many kinds of confusion in some social revolutions in the human-history can be understood as the deviation from the complementarity.

View full abstract

　We have been exposed to such a variety of seriously changing circumstances as many-sided environmental problems of the earth，exhaustion of resources and energy， disasters and accidents that are growing much bigger， troubles and conflict between nations, regions, others， and soon.
　These plural, many-sided, and complex problems of global scale will continue to give us severer selective pressure for survival throughout 21century.
　In order to cope with those problems plurally many-sidedly and comprehensively， we have to create a new paradigm， which could get us continue to contrive both horizontal-domain knowledge against vertical-domain knowledge and high-dimensional general knowledge， what’s more， which could develop the relationship between the local part and the global whole of our thinking and acting into much more self-conclusively harmonious one through putting such two kinds of knowledge contrived by the paradigm into practice.

View full abstract

　A disaster, like a big earthquake, can be regarded as a kind of “Catastrophe”. In this study, we concentrated on the catastrophe, which is predicted, but not known when to happen. The aim of this study is to find the cooperative behavior when the catastrophe occurs using the “Catastrophe Game” modified from Public Goods Game. As a result, we could not observe (a) the effect of advance notice of a catastrophe. On the other hand, we found (b) victims’ cooperative behavior after catastrophe in Total Catastrophe Game in which occurred to all players. Similarly, we did (c) victims’ cooperative behavior and (d) non-victims’ cooperative behavior after catastrophe in Partial Catastrophe Game in which to a part of the players.

View full abstract

　If terms in Japanese technical documents are linked to entries of English Wikipedia, much information about these terms are obtained. This paper conducts cross-lingual entity linking which links terms in Japanese document to entries in English Wikipedia. This paper proposes the ways of word sense disambiguation for that entity linking and proves the effectiveness of those ways using the data manually constructed.

View full abstract

　Characteristic bipolar waveforms were frequently observed by the electric waveform receiver onboard the lunar orbiter named KAGUYA. An algorithm to extract these bipolar waveforms is now under development, but the total amount of the waveform data is about 190GB and it is too huge to apply the algorithm on a general PC workstation. In the present study, we attempt to accelerate computation time by a parallel and distributed workflow named “Pwrake” implemented on NICT science cloud. We report the evaluation results of the efficiency of the data processing.

View full abstract

　Many people think about disaster prevention and reduction for real with increasing natural disasters such as earthquake and heavy rain. Web articles appear promising when we obtain information on disasters. However since there exist a number of articles about disasters, one cannot always get intended information or decide immediately that the articles are beneficial. In this research, we attempt an automatic classification of the articles for the purpose of a disaster article database that we will construct. The classifier takes out words from each article and uses machine learning for categorization where the article is labeled as one of “disaster prevention”, “support”, “heavy rain”, and “earthquake”. This paper describes the classifier together with a classification experiment.

View full abstract

　A class “Knowledge Exploring Skills” is established to learn information literacy in the University of Tsukuba. A bibliographic search game started since 2012. The game allows students to improve their understanding of purpose and skills of bibliographic search. This paper describes an evaluation of the bibliographic search game in 2013. The evaluation use a concept map to compare knowledge structure of students before and after the game. As a result, the bibliographic game lead to the structural changes which the students acquire knowledge about the bibliographic search.

View full abstract

　Many local governments and related organizations create their own characters, called, "Local Characters" or "Yurukyara" to promote their cities and towns. Authors conducted a survey to learn what are the aims of local characters, how well their aims were fulfilled, how are such characters used, what the status of intellectual properties and their licensing, etc. Many local governments felt the aims were fulfilled to some extent, but the characters still not know widely. They faced difficulties in keeping necessary staff people, and maintaining doll suit physically. Most governments held intellectual properties themselves, but licensed such characters often free of charge.

View full abstract

　The increasingly widespread use of electronic health records systems mean that large amounts of medical information are accumulated in text format. In order to support computer readability and analysis of text by natural language processing (NLP) of medical information, we have released ComeJisyoV5-1 with 77,760 entries of medical terms for morphological analysis of text.
　Furthermore, in order to aid NLP of medical text, knowing the kinds of typographical errors present is significant so that they can be reduced and that the terminology can still be interpreted by the computer. Overall, 53 kinds of typographical errors were found and analyzed following an ethically approved investigation of medical information in two facilities.
　As a result, in the two-step conversion process, whereby typed Roman alphabetic characters are converted into kana and then exchanged for selected kanji, most errors occurred in the process of exchanging kana for kanji resulting in 46 terms being converted into homophones or another same-sounding kanji.

View full abstract

　The challenge concerning opening data begins to be seen while the promotion of open data becomes the global tide, and various actions are accomplished in all parts of the world. Therefore, in this study, it pays its attention to a local government public information paper and clarifies a challenge when a public information paper is converted to open data. Specifically, it refers to the results of the local government public information paper open data proof experiment that an open data society for the study of ASPIC nonprofit corporation went for and clarifies a challenge to occur in data reduction and registration.

View full abstract

We analyze the frequency at which press releases from universities form the basis of articles in newspapers. The number of universities that have been publishing press releases has been growing. Moreover, the number of newspaper articles on the university sector has also been growing. In this study, based on data of press releases and two major national newspapers, we investigate what the relations between press releases and newspapers are and how different the scientific research are reported in the two newspapers.

View full abstract

　In this paper, we propose a method resolving polysemy of terminology to select appropriate translation. Terminology polysemy in academic texts must be resolved for retrieving in or translating to other languages, since the polysemy may depend on the research domains, though it is not an easy task for non-experts of the domain. We focused on that the Japanese and English abstracts included in a paper is a cross-lingual corpus, and built a model of LDA latent topic space from a large set of the corpora, which can be used to resolve terminology polysemy.

View full abstract

　This study intends to extract central conception by performing quantitative analysis on "Taiyo Corpus", which is a text corpus of highly respected magazines of modern written Japanese, in order to grasp shift in public opinions of the modern language establishment within Japan. The data consists of 10 genres for "Taiyo Corpus" (1,487 articles written in literary style and 1,741 articles written in colloquial style). Several words distinctive to each genre were extracted according to an index based on IDF measures. These distinctive words of each genre were classified into conceptual categories according to thesaurus definition of category levels. Hierarchical cluster analysis was applied to visualize the difference between literary and colloquial style. The analysis results indicate that the public concern of Japanese at the time has affected the conversion of opinion.

View full abstract

　This study extracted an overview of the cultural heritage rescue that took place after the Great East Japan Earthquake, by the approach of analyzing quantitatively text of activity records. Time-series changes of rescue activities for cultural heritage was clarified, by counting the appeared words those corresponding to three categories which is activities, cultural asset, damage situation. Since the required action and damage situation will be constantly changing after the disaster, it is necessary to perform future measures and disaster reduction based on the characteristics of those.

View full abstract

[in Japanese]

View full abstract