A5059577984- Geographic Information Systems Studies
- Data Management and Algorithms
- Human Mobility and Location-Based Analysis
- Natural Language Processing Techniques
- Semantic Web and Ontologies
- Web Data Mining and Analysis
- Topic Modeling
- Environmental Monitoring and Data Management
- Public Relations and Crisis Communication
- Innovation in Digital Healthcare Systems
- Geochemistry and Elemental Analysis
- Web Applications and Data Management
- Advanced Image and Video Retrieval Techniques
- Energy Efficient Wireless Sensor Networks
- Augmented Reality Applications
- Mathematics, Computing, and Information Processing
- 3D Modeling in Geospatial Applications
- Research Data Management Practices
- Mobile Agent-Based Network Management
- Handwritten Text Recognition Techniques
- Recommender Systems and Techniques
- Biomedical Text Mining and Ontologies
- Context-Aware Activity Recognition Systems
The University of Tokyo
2001-2017
National Disaster Management Research Institute
2017
Infosys (India)
2013-2017
University of Electro-Communications
2017
One of the issues in extracting natural language sentences from PDF documents is identification non-textual elements a sentence. In this paper, we report our preliminary results on in-line mathematical expressions. We first construct manually annotated corpus and apply conditional random field (CRF) for math-zone using both layout features, such as font types, linguistic context n-grams, obtained documents. Although method naive uses small amount training data, achieved an 88.95% F-measure...
Weather is a typical topic of daily conversations, so it natural idea to use social data observe weather. Geotagging key for weather applications because highly localized phenomenon on the earth. Hence we developed software called GeoNLP toponym-based geotagging, and applied Twitter stream find toponyms (place names) from Japanese tweets talking about precipitation events. We observed that less than 10 percent contain toponym information, but can capture events each place. also show temporal...
In this paper, we introduce a cross-lingual recommender system for research papers based on multiple-facets in Japanese. The is the first Japanese paper and recommends international simply by typing keywords. Academic search engine users vary from undergraduate students to researchers. As have different backgrounds preferences, they will aim at kind of papers. order consider various users' introduces eight viewpoints: internationality, similarity, state art, serendipity, contextual analysis...
The Great East Japan Earthquake, which took place in area on the 11 March 2011, had a broad impact land and infrastructure as well human life. City Bureau, MLIT, has made survey of tsunami damage for local government systematically comprehensively. This study develops data archiving this result using various open source GIS, order to keep raw including GIS available, report document. archive about 1,000 registered users 114,000 files that totally amounts 200GByte, such flood data, building...
Most information includes some kinds of spatial data such as the address a restaurant and position person carrying portable phone. The are useful meta multimedia because they provide connections between data. called keys join different contents with relationships. media fusion project started last year (2000) to construct framework content circulation based on keys. A geographic coordinate (x,y) is one kind data, but there other referenced which can be converted coordinates. We particularly...
日本語の技術的な文章中の専門用語から，英語ウィキペディア項目へのリンクを付与することができれば，専門用語について豊富な情報を得ることができる．そこで本研究では，日本語文章中の専門用語から英語ウィキペディア項目へリンクする言語横断エンティティリンキングの実現を目指す．本稿では，そのために必要な語義曖昧性解消手法を提案する．また，手動で構築した評価データを用いて，提案手法の有効性を実験により示す．
The CODATA Data Science Journal is a peer-reviewed, open access, electronic journal, publishing papers on the management, dissemination, use and reuse of research data databases across all domains, including science, technology, humanities arts. scope journal includes descriptions systems, their implementations publication, applications, infrastructures, software, legal, reproducibility transparency issues, availability usability complex datasets, with particular focus principles, policies...
The purpose of this study was to support research studies on geospatial information sciences by examining an Internet-based spatial data infrastructure and data-sharing system. original system’s make available academic researchers. However, after 10 years service, it experiencing problems, such as increased human costs system operation inconvenience user operations, which no longer were negligible. In response, the rebuilt in 2010 reviewing changing its design address problems. Two...
This paper proposes a spatial document management system (SDMS) that can easily and directly manage documents. A is an electronic includes reference information, such as HTML file, Excel text file so on. Our provides simple interface operating documents in own personal computer on the web manages textual information about Point of Interest (POI) these When user load by “drag drop” operation, converts address expressions into POI displays distribution map. SDMS will support mapping actions...
This study demonstrates a framework for joint research at university center that supports in the geospatial information sciences. Here we examine improvements an Internet-based spatial data infrastructure and data-sharing system through its long-term operation. In original this (the Spatial Data Infrastructure System or SDIS), purpose was to make accessing easier academic researchers. However, after ten years of service, it still experiencing challenges, such as increased human costs...
学術的用語には利用される分野によって意味が異なるものがあるため，他言語の文書を検索したり，翻訳する際には適切な対訳を選択する必要があるが，その分野の専門家以外には難しい．そこで，一つの論文に含まれる日本語と英語の抄録がパラレルコーパスであることに着目し，多数の論文抄録からLDA潜在トピック空間のモデルを構築した．このモデルを用いて，学術的な文章に含まれる学術用語の多義性を解消し，適切な対訳を選択する手法を提案する．
As a joint usage/research center, Center for Spatial Information Science has been supporting researchers by establishing spatial data infrastructure academic research and making it available proposed projects based on the strict review inter-university researchers. We reconstruct information system managing whole process above due to expansion of increase number users. In this paper, we summarize problems solutions in developing evaluate usage log. especially specific issues from viewpoint...