In this paper, we designed a novel method to extract comparative
relations from customer opinion data, to build comparative relation
maps for aiding enterprise managers in identifying the potential
operation risks and supporting strategy decisions. The two-level CRF
model with unfixed interdependencies can better extract the
comparative relations, by utilizing the complicated dependencies
between relations, entities and words, and the unfixed interdependencies
among relations. The empirical evaluation demonstrated the
effectiveness of this model. The comparative relation map is
potentially a very effective tool to support enterprise risk management
and decision making. The contributions of this paper include the
following: 1) To the best of our knowledge, this is the first work on
using comparison opinion as information sources in CI for enterprise
risk management; 2) the proposed graphical model can achieve better
performance for relation extraction by modeling the unfixed interdependencies
among relations, which is not covered by the existing
methods; and 3) the empirical evaluation shows that the performance
of the comparative relation extraction is quite promising, and it
implies the feasibility of mining the comparison opinions for CI.
In the future, we plan to conduct an empirical evaluation of the
proposed model on a larger scale with other product types. We also
plan to extend the model to jointly recognize the comparative
relations and entities so as to reduce the errors accumulated in the
pipeline process. In addition, the comparative relation map will be
aligned with the product market shares, for facilitating the analysis of
customer opinions' influence on the product sales and better support
of enterprise decisions. Managers from the industry will be invited to
use and evaluate the system
0
In this paper, we designed a novel method to extract comparativerelations from customer opinion data, to build comparative relationmaps for aiding enterprise managers in identifying the potentialoperation risks and supporting strategy decisions. The two-level CRFmodel with unfixed interdependencies can better extract thecomparative relations, by utilizing the complicated dependenciesbetween relations, entities and words, and the unfixed interdependenciesamong relations. The empirical evaluation demonstrated theeffectiveness of this model. The comparative relation map ispotentially a very effective tool to support enterprise risk managementand decision making. The contributions of this paper include thefollowing: 1) To the best of our knowledge, this is the first work onusing comparison opinion as information sources in CI for enterpriserisk management; 2) the proposed graphical model can achieve betterperformance for relation extraction by modeling the unfixed interdependenciesamong relations, which is not covered by the existingmethods; and 3) the empirical evaluation shows that the performanceof the comparative relation extraction is quite promising, and itimplies the feasibility of mining the comparison opinions for CI.In the future, we plan to conduct an empirical evaluation of theproposed model on a larger scale with other product types. We alsoplan to extend the model to jointly recognize the comparativerelations and entities so as to reduce the errors accumulated in thepipeline process. In addition, the comparative relation map will bealigned with the product market shares, for facilitating the analysis ofcustomer opinions' influence on the product sales and better supportof enterprise decisions. Managers from the industry will be invited touse and evaluate the system0
การแปล กรุณารอสักครู่..
