Project Name

Natural Language Processing Project


T. Tossavainen, K. Nakano, A. Sugiyama


Twitter, Linked Open Data, facilitation, visualization, annotation, mining, consensus building


Twitter Analysis and Applications


A characteristic of Twitter is its real-time nature and is restricted to 140 characters. We use news articles and Wikipedia entries to extract twitter context information. Furthermore, we suggest a scheme of context information as Linked Data and a Web interface for accessing context information.

Facilitating discussion is important for building consensus between stakeholders in various communities. This problem is one of the most important application of natural language processing techniques. For example, public involvement in building consensus for community development requires much effort and time for sharing public concerns among citizens and stakeholders. Hence, supporting discussion by structuring the meeting transcripts and visualizing the structure is indispensable.

We aim to develop a novel discussion support system that newcomers can join a discussion easily. A lively discussion need to join many people. The system supports a discussion between newcomers and stagers.

Copyright (c) 2014 Shintani Lab. All rights reserved.