Rights statement: ©2015 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Accepted author manuscript, 419 KB, PDF document
Final published version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
}
TY - GEN
T1 - Exploring fine-grained sentiment values in online product reviews
AU - Teh, Phoey Lee
AU - Rayson, Paul
AU - Pak, Irina
AU - Piao, Scott
N1 - ©2015 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
PY - 2015
Y1 - 2015
N2 - We hypothesise that it is possible to determine a fine-grained set of sentiment values over and above the simple three-way positive/neutral/negative or binary Like/Dislike distinctions by examining textual formatting features. We show that this is possible for online comments about ten different categories of products. In the context of online shopping and reviews, one of the ways to analyse consumers' feedback is by analysing comments. The rating of the ???like??? button on a product or a comment is not sufficient to understand the level of expression. The expression of opinion is not only identified by the meaning of the words used in the comments, nor by simply counting the number of ???thumbs up???, but it also includes the usage of capital letters, the use of repeated words, and the usage of emoticons. In this paper, we investigate whether it is possible to expand up to seven levels of sentiment by extracting such features. Five hundred questionnaires were collected and analysed to verify the level of ???like??? and ???dislike??? value. Our results show significant values on each of the hypotheses. For consumers, reading reviews helps them make better purchase decisions but we show there is also value to be gained in a finer-grained sentiment analysis for future commercial website platforms.
AB - We hypothesise that it is possible to determine a fine-grained set of sentiment values over and above the simple three-way positive/neutral/negative or binary Like/Dislike distinctions by examining textual formatting features. We show that this is possible for online comments about ten different categories of products. In the context of online shopping and reviews, one of the ways to analyse consumers' feedback is by analysing comments. The rating of the ???like??? button on a product or a comment is not sufficient to understand the level of expression. The expression of opinion is not only identified by the meaning of the words used in the comments, nor by simply counting the number of ???thumbs up???, but it also includes the usage of capital letters, the use of repeated words, and the usage of emoticons. In this paper, we investigate whether it is possible to expand up to seven levels of sentiment by extracting such features. Five hundred questionnaires were collected and analysed to verify the level of ???like??? and ???dislike??? value. Our results show significant values on each of the hypotheses. For consumers, reading reviews helps them make better purchase decisions but we show there is also value to be gained in a finer-grained sentiment analysis for future commercial website platforms.
KW - sentiment value
KW - value expression
KW - like and dislike
U2 - 10.1109/ICOS.2015.7377288
DO - 10.1109/ICOS.2015.7377288
M3 - Conference contribution/Paper
SN - 9781467394338
SP - 114
EP - 118
BT - 2015 IEEE Conference on Open Systems
PB - IEEE
T2 - 2015 IEEE Conference on Open Systems
Y2 - 24 August 2015 through 26 August 2015
ER -