Opinion Mining in Conversational Content within Web Discussions and Commentaries

Abstract : The paper focuses on the problem of opinion classification related to web discussions and commentaries. It introduces various approaches known in this field. It also describes novelty methods, which have been designed for short conversational content processing with emphasis on dynamic analysis. This dynamic analysis is focused mainly on processing of negations and intensifiers within the opinion analysis. The contribution describes implementations of these methods for the Slovak language. The Slovak dictionaries have been created to support these implementations of lexicon based approach. In addition, the paper describes test results of the presented implementations and discussion of these results as well.
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/hal-01506792
Contributor : Hal Ifip <>
Submitted on : Wednesday, April 12, 2017 - 11:19:18 AM
Last modification on : Thursday, April 13, 2017 - 1:06:47 AM
Long-term archiving on: Thursday, July 13, 2017 - 12:38:05 PM

File

978-3-642-40511-2_11_Chapter.p...
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

  • HAL Id : hal-01506792, version 1

Citation

Kristína Machová, Lukáš Marhefka. Opinion Mining in Conversational Content within Web Discussions and Commentaries. 1st Cross-Domain Conference and Workshop on Availability, Reliability, and Security in Information Systems (CD-ARES), Sep 2013, Regensburg, Germany. pp.149-161. ⟨hal-01506792⟩

Share

Metrics

Record views

148

Files downloads

131