In recent years, the digitization of regional and municipal materials has been progressing, but it is difficult to grasp how much it is specifically digitized and opened publicly. In this presentation, we estimated the progress of digitization based on the survey of metadata and digital materials using search services such as the National Diet Library Digital Collections (NDL Digital), the NDL Web Archiving Project (NDL WARP), the municipal homepages, Yokohama City Library Digital Archive and the OPAC of each city library focused on the designated cities (Yokohama, Kawasaki, Sagamihara) in Kanagawa Prefecture. How to cooperate between NDL and public libraries for the issue of openning up the digitized regional materials in the future is discussed.

　Basic Act on the Advancement of Public and Private Sector Data Utilization, which came into effect in 2016, requires prefectures to formulate a prefectural plan for the advancement of public and private sector data utilization. According to a survey by the National Strategy office of Information and Communication Technology, as of April 2019, 22 prefectures had formulated the plan. In this study, we focus on the effect of the formulation of the prefectural plan on existing information-related systems. Specifically, through interviews, we will clarify the impact of planning on the information disclosure system.

　In order to promote reuse of research data, it is necessary to develop system infrastructure that meet the need of restricted access, not only open access. We surveyed data sharing workflow of institutions that support restricted access, and categorized the data providing flow. In addition, we implemented the function in the general-purpose repository system WEKO and, considered its operation by simulating the restricted access workflow of JGSS Research Center.

　The Volume of Historical Statistical Tables on Hakodate City was published in 1987, which is the compilation of more than 1,200 historical data tables, where the data covers the period between 1870's and 1970's. The topics of data are composed of climate, politics, finance, fishery, agriculture, industry, culture, education, social infrastructure, and military. This paper examines how we could engage in such a large number of historical data tables from various areas, and discusses the technical issues and research challenges that the digitization of the historical resources would raise.

In the recent years, the universities and research organizations in Japan have committed to actively manage research data, creating the open data environment accessible by public. Open data is expected to make research data more useful and stimulate innovation, and to achieve this goal it is important to identify and understand how the relation to the society in open science can be constructed and maintained, however such methodology to implement the effective relation has not been clarified. Considering the fact that implementation of the relation between an organization and the society is a fundamental function of public relations, in this paper we first extract and examine such functions of public relations already established in the academic pubic relation activities. Then, we investigate these functions inherent in open science. We also discuss some prospects of the open data approach to the public relations.

　In this paper, we summarize the current status and issues of IT education in Bangladesh, and describe the effects of introduction of Hanpuku Learning, a unique teaching method, for Information Technology Engineers Examination (ITEE), which is being executed as a national certificate originally from Japan. It also shows the excellence of young people in Bangladesh which wants to become an IT nation and proposes a possibility that is beneficial to both Japan and Bangladesh. The effect of the introduction is excellent, and it can be determined that the Japanese-style "Hanpuku Learning" adapted to the temperament of Bangladeshi people and sustainable learning habits were established.

　In this study, we conducted text mining on the Diet speeches (both administrative policy speech and prime minister's general policy speech) of post-war Japanese prime ministers to clarify the chronological changes in political thoughts. A total of 179 speeches from the Higashikuni-no-miya Cabinet (1945) to the fourth Abe Cabinet (2019) were collected, and morphological analysis was performed after dividing them into four periods: the rst half and latter half of the Showa era, and the rst half and latter half of the Heisei era. Characteristic words were extracted by calculating TF-IDF values for frequent nouns in each era. Then, a topic model using LDA was created for all speeches, and topics were estimated for each prime minister.

　In the story analysis, analysis focusing on the information possessed by scenes has rarely been performed so far. It is essential to analyze the information dependency of scenes in automatic story generation. In this study, scene dependencies were analyzed using mystery novels. Specifically, the scene was manually extracted from the actual mystery novel the information necessary to compose. After that, the obtained data was summarized for each dependency. Finally, the data obtained was performed network analysis and analyzed the dependencies.

　Patterning and random combination of fundamental plot structure had been general methodology for automatic story generation. However, natural story works have complex story structures that include several combined fundamental story structures. In this research, focused on detective story as an example, a methodology for constructing complex story structure such as serial killing, based on fundamental story patterns of detective story, according to motivations of characters.

　While technologies that helps accumulating and utilizing life logs advances, applications that make use of Web browsing history as life logs have not become widespread. This study aims to enhance the value of Web browsing history as life logs by supporting reflection of information gathering behavior. We propose the method of visualization for the information gathering behavior by presentation of Web page contents and browsing process using Web browsing history. In our method, the Web browsing history is presented by combination of the browsing time series and the topics extracted by Web page text with the method of TF-IDF weighting and k-means clustering. In comparison with the existing method, significant effectiveness of the proposed method was not found in reflection of information gathering behavior. In this paper, we discuss and report on approaches for improving the proposed method.

　Recent years, an attempt for the purpose of publishing, sharing, reusing data called "Open Science" has become active for easier access to papers and research data. Along with this, in Japan, academic repositories are built at many academic research institutions using a repository system named WEKO. When we consider using WEKO as a data repository, it is not easy for the users to search the data which they wish because metadata are not well standardized in many academic elds. In the present study, we developed a new user interface for the data repository using WEKO3, which is expected to be a mainstream of repository system in the future. This interface is a system that manages items by assigning tags to them, and aims to form collective knowledge by sharing tags among multiple repositories.

　In recent years, there has been increasing interest in Open Data. The utilization of Open Data has been promoted, and many organizations including the national government, local governments and other organizations are working on publishing and utilizing Open Data. Open data of Kagoshima City has been published in July 2016 as CSV format. Under Open Data of local governments, even if they are disclosed due to the difference in data format and the lack of linkage of data, there is no active utilization yet. In this research, we focused on the vocabulary that corresponds to the predicate of RDF form for Open Data, and proposed a method to suggest predicates by the clustering of item names for RDF of Open Data as teacher signals of Deep Learning. Then, we presented results of the predicate suggestion.

　Television has become an application on the Internet. This study has, therefore, investigated whether it is necessary for Botswana Television (Btv) to establish its website. The feasibility and usability of the website were verified by a web prototype. A pilot survey was conducted by the use of questionnaire for both qualitative and quantitative approach. The participants revealed 67% satisfaction with the proposed prototype and indicated that the web application would enhance how they obtain valuable information like news, sports, and previous television programs conveniently. The observed positive response may be a lead on how Btv can improve its online service delivery, accelerate government's efforts of promoting an informed and knowledgeable society as well as safeguarding the citizens' constitutional rights to know.

　We developed a multiple-choice question generation supporting system for information processing subjects. The system operates using a browser, and the user first specifies a correct word. By using DBpedia for the system to obtain a summary of the correct word and related words and by editing and selecting the query results, the user finally acquires the question sentence and wrong choices. Evaluation experiments showed that the creation time was shorter by using the system.

　In this study, we analyzed the story structure of a genre called Role Playing Game (RPG) in computer games. Two works, the "Dragon Quest" series and the "Final Fantasy" series, were selected as targets. The analysis was done by the tags based on the elements of stories that are expected to appear frequently. As a result, it is clear that the basic RPG scenario consists of a continuous and nested structure of narrative units called quests, and that the quest itself can be expressed by three elements: "occurrence", "elapse", and "ending".

　The relationship between the emotional state of a character and the background music in a scene of a story is not yet clear. However, if this relationship is clarified, it can be used to build an automatic BGM selection system and automatic BGM generation system suitable for a story scene. This research focused on computer role playing games because of popularity and nature of story, and created data set for an automatic BGM selection system. In particular, evaluated emotional state of characters and acoustic features in scenes were digitized as 477 data set, and correspondences of those data set were analyzed.

　In this paper, features are extracted from story analysis and quantitative analysis with the aim of structuring the story of a heroic mythology. The usefulness of the method was demonstrated by comparing the obtained features with mythological findings. The story analysis was performed on the story of heroic mythology, the story was expressed by the continuous relation of the function, and the result of the story analysis was subjected to N-gram analysis and chi-square test to extract the features of the stories. The features of the stories extracted as a result of the analysis correspond to the features of each mythology system in mythology, and it is considered that the method of this paper quantitatively extracted the features of the story of heroic mythology.

　”Happy-Ending”and ”Bad-Ending”are the most major categorizations in the narrative, but little work has been done on quantitative analysis about Happy-Ending works and Bad-Ending works. This study was made to extract Comedy and Tragedy in Otsuichi’s short stories. Firstly, the target datas were classified into three genres, and a new Japanese Sentiment Polarity Dictionary was created based on the thesaurus. Secondly, polar words in the text were extracted with Polarity Dictionary. In addition,we extract features of Happy-Ending works and Bad-Ending works by hierarchical cluster analysis and multiple regression analysis. As a result, this study showed that polar words categorized as ”happy”,”love”, and ”respect”in the last quarter of the story contributed to the Happy-Ending works. This study also showed that polar words categorized as ”sorrow”and ”sad”in the last quarter of the story contributed to the Bad-Ending works.

　In this study, we constructed a dictionary of emotion words classified into 12 types of emotions in order to extract the emotional state of the characters from the story. We selected a novel called ”Kino’s Journey -The Beautiful World-”as an analysis target, and extracted emotion words using a dictionary. The number of emotional words extracted was treated as multidimensional data and factor analysis was performed. As a result, it was classified into three factors: ironic story factors, surprising story factors, and tragic story factors. In addition, we extracted the differences in emotions for each role of the characters in each factor.

　In this study, we focused on the importance of the diary and analyzed the differences in stylistic features depending on the attributes of the writer. Collected text from blog. Words indicating stylistic features were extracted. Word vectors were created using tf-idf. Factor analysis was performed based on word vectors. Four categories of stylistic features were analyzed: old men, old women, young men, and young women. As a result, it was found that the style was different depending on the attributes. Stylistic features common to all categories were also found.