of E-Negotiation Users and Processes with Inspire & Negoisst Datasets
Concordia University, S. Szpakowicz, University
of Ottawa, R. Vetschera, University of Vienna
University of Ottawa, S. Koszegi, University of
Vienna, H. Lai, National Sun Yat-sen University,
M. Schoop, Lehrstuhl Informatik V, Rheinisch Westfalishe
This project builds upon the experience gained in analysing data collected
from the Inspire e-negotiation system to determine the relationships between
user characteristics, the negotiation process and its outcomes. The Inspire
ENS has been used in teaching and training since 1996. To date we have
collected data from over 5,000 users from 40 countries. This project will
examine qualitative, qualita-tive and text data collected from the Inspire
negotiations. Exploratory research has already uncovered several relationships
between culture, gender and other users’ characteristics, and the
use of the ENS, e-negotiation process and outcomes. These influences will
be studied now with the complete dataset and with the text, the latter
using both methods coming from natural language processing and the qualitative
research. Lexical and semantic models of the domain will be formulated
to study how the negotiators' language skills affect their performance.
To validate the models and to show how the re-sults may be generalized,
they will be applied to other ENSs, such as Negoisst, for possible adapta-tions
and reassessments. The preliminary studies also indicated the necessity
to re-design and better validate the measurement and analysis instruments
used. These instruments will be used in the ex-periments with Inspire
and other ENSs.
- Build tools to extract and categorize the textual elements of the
Inspire negotiation tran-scripts.
- Organize textual elements into a morphologically tagged corpus; link
individual texts to ne-gotiation histories; construct a domain model
and ontology, and a domain-specific semantic lexicon.
- Development and initial empirical analysis of a model for the assessment
of web-based sup-port systems (AMIS) based on existing Inspire data.
- Design and build a dedicated syntactic analyzer (or adapt an existing
analyzer) and apply it to the corpus.
- Study of the social, cultural, gender and other user characteristics
on the e-negotiation proc-ess and outcomes and the use of Inspire.
- Re-design and validation of measurement instruments for user characteristics,
user attitudes and negotiation processes.
- Development of a consistent framework for the statistical analysis
of negotiation data.
- Classify segments of texts by relevance to the underlying history;
identify references to negotiation steps.
- Classify negotiation utterances into speech-act based communication
types; identify refer-ences to negotiation steps and the resulting effects
such as obligations.
- E-negotiation controlled experiments with the Inspire and Negoisst
- Propose a preliminary catalogue of phrases associated with winning,
losing and abandoning negotiation.