Research Outline
When the evolution of the distribution of academic information in the history of mankind is viewed in the current ICT environment, it is seen that researchers desire to access the latest academic research data in a fraction of a second and safely share their own research and procedures among suitable co-workers, quickly commercialize and publish the results of their research, and, in the process, find ways of sharing information by which they can acquire greater competitive funding and better co-researchers.
However, there are several theoretical and technical difficulties here. Firstly, the explosive increase in the amount of academic research data regarding various fields in the web space has made access to data nearly impossible. As the needs of researchers vary widely, the search technology to achieve this requires the technology of a push-type information search and information recommendation that will provide accurate information specific to individual researchers; the second difficulty arises here.
Therefore, we aim to overcome this challenge by integrating the three methods, namely, data mining and ontology, obtaining feedback from the users themselves through social media methods, and associative search utilizing reliable external data and the information attached with it.
(Project Director: Noriko Arai [National Institute of Informatics])
Objective of the Project
We aim to build a mechanism, the “Science 3.0 platform,” which will help researchers to comprehend the rapidly increasing amount of research resources and information and promote cross-disciplinary research without being hindered by the field-specific cultures or spatial constraints. More specifically, in order to make the research resources and information machine-readable, research and development of large-scale data mining technology and search technology, semantic web technology for linking existing academic databases, information linkage technology that connects heterogeneous databases, and ontology technology shall be promoted.
Project Promotion Structure
Implementation scheme of research and collaboration with universities and other research institutions (Implementation system towards the realization of the project)
The target of this study is the research community that aims to solve problems in academic research or in society on the basis of data. This study provides a research environment through visualization, modeling, and simulation of the structure behind the challenges by advanced utilization of data provided by a data infrastructure improvement project, thus promoting appropriate problem solving. In particular, the aspects of (1) visualization and search support technology of data structures and (2) modeling of knowledge-based technology are being considered. The semantic web technology, large-scale data mining and ontology technology, and the linkage technology and search technology that connect heterogeneous databases are indispensable to the implementation of the present study, so that it could be coordinated with existing academic databases. Therefore, this research project is divided into three sub-themes. The research and development of semantic web technology will be carried out in the Takeda sub-project and that of large-scale data mining and ontology technology in the Aizawa sub-project. The Arai sub-project integrates these two technical components and carries out research and development of infrastructure in order to serve researchers by incorporating visualization and search support technology of data structures. The project thus promotes research empirically by actually offering research results to universities and major domestic research groups, academic societies, etc.
As part of the “Transdisciplinary Research Integration Center” project that was carried out prior to the implementation of the present study, the trial version of the Science 2.0 infrastructure service “Researchmap” intended for researchers in Japan was made public and became operational. Researchmap was integrated with ReaD in the year 2011 and has grown dramatically into the world’s largest research community, with 230,000 participating researchers. This service has a monthly average of 350,000 cases of unique access and has risen to third place in the search rank on both Google and Yahoo by the keyword “researchers.” In addition, as many as 14 million papers have been published on this service (more than twice as compared to last year), and these research resources are organized in a manner that forms the ideal base to start building an infrastructure for the interdisciplinary sharing and collaboration of research resources, which is the objective of this study.
Introduction of Sub-themes
1. Establishment of information recommendation platform related to research resources
In this sub-theme, we have thus far been working on a paper recommendation method and development of a demonstration system, under the Transdisciplinary Research Integration Center project. Based on the findings of this research, in this sub-theme, the study has been expanded to support up to “search for information that is written in papers” from “search for papers as a unit” and is targeted to be established as one of the technologies for the research commons platform. The research is advanced by focusing on the two aspects of (1) observation and analysis of the document reading process and modeling and (2) establishment of a document analysis platform through the integration of the document structure and the semantic structure as essential technical components for this. In addition, the development and demonstration of the paper browsing system to support information access of researchers is promoted through both (3) the technology of information presentation with high readability as well as information search support technology based on semantic analysis.
(Principal researcher: Akiko Aizawa [National Institute of Informatics]
2. Establishment of Open Social Semantic Web infrastructure for academic resources
The goal of this sub-theme is to build a mechanism that will enable the sharing of academic information in the scholarly community and in society. The Internet, and particularly the Web, is important as an infrastructure that accomplishes a broad and flexible academic information distribution as mentioned above. However, since the Web in its current form does not incorporate such a mechanism, it is difficult to achieve such a broad and flexible academic information distribution. Therefore, we aim to build an information distribution infrastructure using techniques such as the semantic web. The meaning of the information in the semantic web is described using the mechanisms of URI and ontology. Exchange of information in different fields can be made possible by using the semantic description of this information. Furthermore, the bottom-up information structure can be realized by introducing the concept of the social Web. Generation and structuring of information in the Social Web is done through the participation of a wide range of people. In this study, a centralized mechanism, rather than a distributed mechanism, is introduced in building and mapping ontology. Wider distribution of information can be thus achieved. Moreover, all such information should be available without restriction. Hence, we aim to build a mechanism that will publish information in an openly accessible format. The result of the present study is to build and make available an information infrastructure that will help realize flexible and extensive information distribution. Using this information infrastructure, sharing and distribution of information in the more minute areas is made possible besides facilitating information distribution beyond the boundaries of academic discipline.
The results of the research in this sub-theme shall be directly provided to the researchers as required through systems such as Researchmap so that improvements could be made using their feedback.
(Principal researcher: Hideaki Takeda [National Institute of Informatics])
3. Establishment of information sharing cloud services to accelerate integrated research
In this sub-theme, the current state of the study was visualized by combining a variety of static information relating to research such as research papers and patents, competitive funding situations, and social action programs with dynamic information known as research activities. We aim at building a recycling-oriented research information infrastructure leading to integrated research and collaboration of industry, academia, and government in addition to the resolution of social issues utilizing the research. First, we shall establish the terms of the researchers’ entry by using the results of semantic web technology research in the Takeda project and carry out the research and development of the technique of linkage of multi-typed information and the methodology of visualization. Further, the OpenDepo system shall be built for the accumulation of primary data including research papers; this is given an ID link with Researchmap. On the basis of these data, the search and recommendation of information given in papers that are developed by the Aizawa project will be implemented, and, subsequently, the support interface for reading the papers will be developed and provided by this sub-project.
Thus, it will be possible to provide personalized service to each researcher by sifting useful academic and research information from the explosive information space.
We plan to offer the software Researchmap that has been developed in the present study to JST any time after empirical trials and provide it to researchers in Japan as a researcher service ReaD & Researchmap.
(Principal Researcher: Noriko Arai [National Institute of Informatics])
Research View 026
Toward Open Web Data
[e-Science Fundamental Technology] Hideaki Takeda (Professor, National Institute of Informatics)
Research View 024
Artificial Intelligence: our wait for machine-readable Big Data
[e-Science Fundamental Technology] Noriko Arai (Professor, National Institute of Informatics)