TY - GEN
T1 - System for supporting web-based public debate using transcripts of face-to-face meeting
AU - Shiramatsu, Shun
AU - Takasaki, Jun
AU - Zidrasco, Tatiana
AU - Ozono, Tadachika
AU - Shintani, Toramatsu
AU - Okuno, Hiroshi G.
PY - 2010
Y1 - 2010
N2 - We propose a public debate support system through iterative alternation of face-to-face meetings and Web-based debates. Web-based facilitation of public debates is suitable for complementary question and answering when the debate in face-to-face meetings is insufficient due to time restriction. Since transcripts of public meetings tend to be lengthy and specialized, sharing public concerns among citizens and stakeholders requires much time and effort. To solve this problem, we propose two approaches. First, supporting the sharing of public concerns with structuration of public debate based on rhetorical structure theory (RST). It enables a user to manually specify his/her intention of a question, e.g., the question requires more evidence. An intention tag is effective for facilitating public debate. Second, supporting reading with visualization of topic transition, using our developed SalienceGraph. Moreover SalienceGraph retrieves passages related to a transient topic from past meeting records or documents. These approaches support citizens and stakeholders in finding, tracking, and sharing public concerns.
AB - We propose a public debate support system through iterative alternation of face-to-face meetings and Web-based debates. Web-based facilitation of public debates is suitable for complementary question and answering when the debate in face-to-face meetings is insufficient due to time restriction. Since transcripts of public meetings tend to be lengthy and specialized, sharing public concerns among citizens and stakeholders requires much time and effort. To solve this problem, we propose two approaches. First, supporting the sharing of public concerns with structuration of public debate based on rhetorical structure theory (RST). It enables a user to manually specify his/her intention of a question, e.g., the question requires more evidence. An intention tag is effective for facilitating public debate. Second, supporting reading with visualization of topic transition, using our developed SalienceGraph. Moreover SalienceGraph retrieves passages related to a transient topic from past meeting records or documents. These approaches support citizens and stakeholders in finding, tracking, and sharing public concerns.
UR - http://www.scopus.com/inward/record.url?scp=79551534541&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79551534541&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-13033-5_32
DO - 10.1007/978-3-642-13033-5_32
M3 - Conference contribution
AN - SCOPUS:79551534541
SN - 3642130321
SN - 9783642130328
VL - 6098 LNAI
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 311
EP - 320
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
T2 - 23rd International Conference on Industrial Engineering and Other Applications of Applied Intelligence Systems, IEA/AIE 2010
Y2 - 1 June 2010 through 4 June 2010
ER -