The Internet is becoming the most important source of information for the creation of knowledge because the social activities of the state, companies, and individuals are recorded and released as information on the Internet. There is a strong tendency to access only self-bias information, making it a serious problem. Therefore, it is necessary to support information acquisition efficiently and without bias. In addition, there are many easy-to-understand but Web-based descriptions with only superficial descriptions. It will be effective for the user's learning and understanding support to complementarily present the pages explained in more detail using various materials.
Discovery and analysis of entities projected on information is important in order to obtain information for the creation of knowledge from the Internet. In our research, we identify, extract, and organize descriptions about entities. It also analyzes entity relationships and temporal transitions, and performs entity dynamic relationship mining.
Video integrity analysis
Using subtitles and surrounding text of video content, we search for related information of video content from other media sources, compare and analyze, and find differences (inconsistencies) in the description of video content. If only a specific part in the real world can be explained in the content, we clarified it, and their differences between other parts are presented to build a system that supports the user to determine the authenticity of the video contents.
Information complement system
Categorize subjective and objective descriptions of entities, and analyze the representativeness / typicality / complementary relationship of content from their correspondence. We will develop information complementation systems that search for content that has a complementary relationship with the content specified by the user, correct the bias of the information, and provide information to the user in a detailed and balanced way.
1. Makoto Yokoyama and Qiang Ma, “Topic Model-based Freshness Estimation Towards Diverse Tweet Recommendation”, BigComp2019:1-8, 2019.2
2. Keisuke Kiritoshi and Qiang Ma： Named Entity Oriented Difference Analysis of News Articles and Its Application. IEICE Transactions 99-D(4): 906-917, 2016
3. Shintaro Horie, Keisuke Kiritoshi, and Qiang Ma: Abstract-Concrete Relationship Analysis of News Events Based on a 5W Representation Model. DEXA (2) 2016: 102-117，2016
4. Keisuke Kiritoshi and Qiang Ma： A Diversity-Seeking Mobile News App Based on Difference Analysis of News Articles. DEXA (2) 2015: 73-81，2015
5. Hiroshi Ishii, Qiang Ma, and Masatoshi Yoshikawa： Incremental Construction of Causal Network from News Articles, Journal of Information Processing, Vol. 20, No.1, pp.207-215, 2012
6. Tatsuya Ogawa, Qiang Ma, and Masatoshi Yoshikawa: News Bias Analysis Based on Stakeholder Mining, IEICE Transactions on Information and Systems, E94-D, pp. 578-586, 2011
7. Shin Ishida, Qiang Ma, and Masatoshi Yoshikawa: Extraction of Characteristic Description for Analyzing News Agencies, Journal of Digital Information Management, Vol.8 Issue 6, pp.349-355, 2010
8. Qiang Ma, Akiyo Nadamoto, and Katsumi Tanaka: Complementary Information Retrieval for Cross-Media News Content, Information System Journal Vol. 31, No.7, pp. 659-678, 2006
9. Qiang Ma and Katsumi Tanaka: Topic-Structure-Based Complementary Information Retrieval and Its Application, ACM Transactions on Asia Language Information Processing, Vol. 4, No.4, pp.475-503, 2005
10. Akiyo Nadamoto, Qiang Ma, and Katsumi Tanaka: B-CWB: Bilingual Comparative Web Browser Based on Content-Synchronization and Viewpoint Retrieval, Journal of World Wide Web, Vol.8 No.3, pp. 347-367,2005.
• 科学研究費補助金 (若手（A）)，「エンティティマイニングに基づく情報補完機構に関する研究」，2013-2016
• 科学研究費補助金 （若手（B）），「情報補完のための検索方式とそのクロスメディア検索への応用」，2008-2010