STATEMENT MAP Project
Inui and Okazaki Laboratory, Tohoku University
STATEMENT MAP: Mapping statements of interest onto the logical space of an argument
The importance of the internet as a source of information cannot be disputed.
A recent poll by the Pew Research Center Pew Research found that among Americans
the internet has overtaken newspapers as a news outlet and rivaled television for those surveyed
under the age of thirty.
Recent research reports that people are turning to the internet for information on important decisions like
health care, medical information, and large purchases; However, we must not accept as given the reliability of
all information on the Web because there is a lot of incorrect information on the Web.
So, although users are at risk of believing incorrect information, they often lack the knowledge
necessary to evaluate the credibility of online information.
To evaluate the credibility of information on the Web, it is necessary to gather sentences and summarize
them, organize opinions on a topic into different viewpoints, and show users the evidence.
Metzger pointed out not only the quality of description but also citing to scientific information
in other Web sites, as factors to be utilized for evaluating the credibility of information on the
Web. Meola also indicated that comparing the focused Web information with others is able
to reveal specific areas of a topic that are controversial and that need special attention and verification,
and corroborating information is to verify it against one or more different sources.
The goal of the STATEMENT MAP project is to assist internet users with evaluating the credibility
of online information by presenting them with a comprehensive survey of opinions on a topic and
showing how they relate to each other. However,
because real text on the Web is often complex in nature, we target a simpler and more fundamental
unit of meaning which we call the “statement.”
To summarize opinions for the statement map users, we first convert all sentences into statements and then,
organize them into groups of agreeing and conflicting opinions that show the logical support for each group.
Consider the case of an anxious user who is worried about whether vaccines are really safe for his or her child.
The figure shows the results of a similar query ”Do vaccines cause autism?” would produce with STATEMENT MAP.
Publications (in English)
- Organizing Information on the Web through Agreement-Conflict Relation Classification. Junta Mizuno, Eric Nichols, Yotaro Watanabe and Kentaro Inui. In Proceedings of the Eighth Asia Information Retrieval Societies Conference (AIRS 2012), 2012. (to appear)
- Toward Evidence Search. Eric Nichols, Junta Mizuno, Yotaro Watanabe, and Kentaro Inui. Seventeenth Annual Meeting of the Association for Natural Language Processing. 2011.
- Recognizing confinement in web texts. Megumi Ohki, Eric Nichols, Suguru Matsuyoshi, Koji Murakami, Junta Mizuno, Shouko Masuda, Kentaro Inui and Yuji Matsumoto. 9th International Conference on Computational Semantics (IWCS 2011). pp.215-224. 2011
- Statement Map: Reducing Web Information Credibility Noise through Opinion Classification. Koji Murakami, Eric Nichols, Junta Mizuno, Yotaro Watanabe, Shouko Masuda, Hayato Goto, Megumi Ohki, Chitose Sao, Suguru Matsuyoshi, Kentaro Inui and Yuji Matsumoto. Fourth Workshop on Analytics for Noisy Unstructured Text Data (AND 2010). pp.59-66. 2010
- Automatic Classification of Semantic Relations between Facts and Opinions. Koji Murakami. Eric Nichols, Junta Mizuno, YotaroWatanabe, Hayato Goto, Megumi Ohki, Suguru Matsuyoshi, Kentaro Inui, Yuji Matsumoto. The Second International Workshop on NLP Challenges in the Information Explosion Era (NLPIX 2010). pp.21-30. 2010
- Annotating Event Mentions in Text with Modality, Focus, and Source Information. Suguru Matsuyoshi, Megumi Eguchi, Chitose Sao, Koji Murakmai, Kentaro Inui, Yuji Matsumoto, In Proceedings of The seventh international conference on Language Resources and Evaluation, 2010
- Constructing a Scientific Blog Corpus for Information Credibility Analysis. Eric Nichols, Koji Murakami, Kentaro Inui, Yuji Matsumoto, In Proceedings of PACLING 2009, 2009. (to appear)
- Annotating Semantic Relations Combining Facts and Opinions. Koji Murakami, Shouko Masuda, Suguru Matsuyoshi, Eric Nichols, Kentaro Inui, Yuji Matsumoto, In Proceedings of The Third Linguistic Annotation Workshop (LAW III), pp.150-153, 2009.
- Statement Map: Assisting Information Credibility Analysis by Visualizing Arguments. Koji Murakami, Eric Nichols, Suguru Matsuyoshi, Asuka Sumida, Shouko Masuda, Kentaro Inui, Yuji Matsumoto, In Proceedings of the 3rd Workshop on Information Credibility on the Web (WICOW2009), pp.43-50, Madrid, Spain, 2009.
- Constructing a Scientific Blog Corpus for Information Credibility Analysis. Eric Nichols, Koji Murakami, Kentaro Inui, Yuji Matsumoto, Anuual meeting of Japanese NLP Society, pp. 176-179, 2009.
- A Database of Relations between Predicate Argument Structures for Recognizing Textual Entailment and Contradiction. Suguru Matsuyoshi, Koji Murakami, Yuji Matsumoto, Kentaro Inui, In Proceedings of the Second International Symposium on Universal Communication, pp.366-373, Osaka, Japan, 2008.
- Experience Mining: Building a Large-Scale Database of Personal Experiences and Opinions from Web Documents. Kentaro Inui, Shuya Abe, Hiraku Morita, Megumi Eguchi, Asuka Sumida, Chitose Sao, Kazuo Hara, Koji Murakami, and Suguru Matsuyoshi, In Proceedings of the 2008 IEEE/WIC/ACM International Conference on Web Intelligence, pp.314-321, Sydney, 2008.
- Constructing a Corpus of Logical Refutations from Science Blogs. Eric Nichols, NLP Forum 2008.