| Presentation | 2015-05-14 Opinion Target Extraction for TV Viewers' Spontaneous Messages based on Sentiment Terms in Past Tense Nanae Kinoshita, Takeshi Kobayakawa, Atsushi Matsui, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In recent years, spontaneous messages at social networking service (SNS), e.g. Twitter, often reflect individual opinions about socially shared topics, including TV broadcast programs. TV broadcasters have been seeking for such informative text resources, especially for subjective phrases with some opinions to the TV contents. In order to extract such words from the gigantic logs of SNS, particularly specific nouns associated with the TV contents in concern, we propose an opinion target extraction technique based on sentiment terms described in past tense. Experimental comparisons were performed in terms of the number of extracted nouns, opinion targets, against the following two types of conventional approaches: 1). Opinion target extraction by sentiment terms described in any types of tense, 2). Opinion target extraction by sentiment terms described in present tense or future tense. Experimental results show the effectiveness of our proposed method. We will also mention some possible reasons for those improvements in opinion mining performances. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Twitter / Opinion Mining / Opinion Target Extraction / Past Tense / Natural Language Processing |
| Paper # | LOIS2015-2 |
| Date of Issue | 2015-05-07 (LOIS) |
| Conference Information | |
| Committee | LOIS / IPSJ-GN / IPSJ-SPT |
|---|---|
| Conference Date | 2015/5/14(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Tsuda College, Kodaira Campus |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Manabu Okamoto(NTT) / Satoshi Ichimura(*) / Kanta Matsuura(*) |
| Vice Chair | Hiroyuki Nishi(Sojo Univ.) |
| Secretary | Hiroyuki Nishi(NTT) / (Tsuda College) / (*) |
| Assistant | Tomoki Watanabe(NTT) |
| Paper Information | |
| Registration To | Technical Committee on Life Intelligence and Office Information Systems / Special Interest Group on Groupware and Network Services / Special Interest Group on Security Psychology and Trust |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Opinion Target Extraction for TV Viewers' Spontaneous Messages based on Sentiment Terms in Past Tense |
| Sub Title (in English) | |
| Keyword(1) | |
| Keyword(2) | Opinion Mining |
| Keyword(3) | Opinion Target Extraction |
| Keyword(4) | Past Tense |
| Keyword(5) | Natural Language Processing |
| 1st Author's Name | Nanae Kinoshita |
| 1st Author's Affiliation | RECRUIT STAFFING CO. LTD(Recruit Staffing) |
| 2nd Author's Name | Takeshi Kobayakawa |
| 2nd Author's Affiliation | Japan Broadcasting Corp.(NHK) |
| 3rd Author's Name | Atsushi Matsui |
| 3rd Author's Affiliation | Japan Broadcasting Corp.(NHK) |
| Date | 2015-05-14 |
| Paper # | LOIS2015-2 |
| Volume (vol) | vol.115 |
| Number (no) | LOIS-27 |
| Page | pp.pp.29-34(LOIS), |
| #Pages | 6 |
| Date of Issue | 2015-05-07 (LOIS) |