![](bg7.png)
specific words such as “parent company”, “sub sys-
tem”, and so on. The proposed system has not found
the relations based on the domain-specific words. As
our future work, we will try to obtain the knowledge
including the domain-specific words from the other
resources, i.e. the documented case, the other lecture
book, wikipedia and so on. In addition, the size of
the graph gets larger than we expects. Some learn-
ers indicate difficulty in finding the past opinions and
their relations. So, contraction of the discussion struc-
ture is also necessary. Furthermore, coloring nodes of
important opinions will help the learners’ understand-
ings.
5 CONCLUSION
We proposed the method identifying opinions which
have relations by demonstrativesand connectives, and
complementing links with the separate nodes by con-
sidering the relation in the sequence of learners’ opin-
ions. In the experiment, on average, we confirmed
that the proposed method could improve both pre-
cision rate and recall rate by 5.6% and 23.8% re-
spectively compared with the conventional method.
Opinions including demonstratives and connectives
are linked with the previous opinion in this method.
However, it is necessary to create links by extracting
the content correctly, so we will improve the method
in the future. In addition, we will develop the method
in order to complement links more precisely.
ACKNOWLEDGEMENTS
This work was partially supported by KAK-
ENHI:JSPS (25730205).
REFERENCES
Brooke, S. L. (2006). Using the case method to teach on-
line classes: Promoting socratic dialogue and critical
thinking skills. International Journal of Teaching and
Learning in Higher Education, 18(2):142–149.
Conklin, J. and Begeman, M. L. (1988). gibis: A hypertext
tool for exploratory policy discussion. ACM Transac-
tions on Information Systems (TOIS), 6(4):303–331.
Fillmore, C. J. (2011). Pragmatically controlled zero
anaphora. In Proceedings of the annual meeting of
the Berkeley Linguistics Society, volume 12.
Gerard, S. and McGill, M. J. (1983). Introduction to Mod-
ern Information Retrieval. McGraw-Hill.
Gragg, C. I. (1951). Because wisdom can’t be told. Inter-
collegiate Case Clearing House.
Hammond, J. S. (1980). Learning by the case method. Har-
vard Business School.
Hearst, M. A. (1997). Texttiling: Segmenting text into
multi-paragraph subtopic passages. Computational
linguistics, 23(1):33–64.
Hobbs, J. R. (1990). Literature and Cognition. Center for
the Study of Language and Information Publication.
Lee, A. and Kawahara, T. (2009). Recent development of
open-source speech recognition engine julius. In Pro-
ceedings of APSIPA Annual Summit and Conference,
pages 131–137.
Matsumura, N., Kato, Y., Ohsawa, Y., and Ishizuka, M.
(2003). Visualization of discussion structure for dis-
covering and understanding discussion points. Japan
Society for Fuzzy Theory and Intelligent Informatics,
15(5):554–564.
Okumura, M. and Takeo, H. (1994). Word sense disam-
biguation and text segmentation based on lexical co-
hesion. In Proceedings of the 15th conference on
Computational linguistics-Volume 2, pages 755–761.
Rienks, R. J., Heylen, D., and van der Weijden. E. (2005).
Argument diagramming of meeting conversations. In
Proceedings of the 7th International Conference on
Multimodal Interfaces, pages 85–92.
Schwarz, R. (2002). The Skilled Facilitator: A Compre-
hensive Resource for Consultants, Facilitators, Man-
agers, Trainers, and Coaches. Jossey-Bass.
Werner, K. and Rittel, H. W. (1970). Issues as elements of
information systems. University of California.
Yamashita, M. (2000). An introduction of facilitation
graphic system for resident’s participation. Architec-
tural Institute of Japan, 73:395–396.
Zhao, Y., Matsumura, M., and Yoshida, M. (2006). Auto-
generation of discussion-map from speech-recognized
minute. In Proceedings of the 20th Annual Conference
of the Japanese Society for Artificial Intelligence.
KMIS2014-InternationalConferenceonKnowledgeManagementandInformationSharing
132