Adding a Capability to Extract Sentiment from Text using HanDles

PDF

Authors
  1. Dennis, S.
  2. Stone, B.
  3. Hamm, J.
  4. Kwantes, P.
Corporate Authors
Defence R&D Canada - Toronto, Toronto ONT (CAN);Defence R&D Canada - Ottawa, Ottawa ONT (CAN)
Abstract
HanDles is a document visualization tool developed by Ohio State University for DRDC Toronto. One aspect of documents that might be of interest to analysts is the extent to which they express positive or negative opinion or sentiment toward some issue or group. In this report, we describe how HanDles was extended to include the ability to classify documents as containing predominantly positive or negative sentiment. To do so, we trained the semantic model underlying HanDles' understanding of the document collection to distinguish positive from negative documents. Our tests of the system suggested that its ability to discriminate positive from negative documents would be greatly improved by selecting a training collection that is similar in nature and content to the documents that will be evaluated in operational settings.

Il y a un résumé en français ici.

Keywords
Handles;document visualization;sentiment analysis;opinion mining
Report Number
DRDC-TORONTO-TM-2012-063 — Technical Memorandum
Date of publication
01 May 2012
Number of Pages
31
DSTKIM No
CA036892
CANDIS No
536633
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: