| Presentation | 2018-09-06 Latent co-occurrence words graph extraction using sparse structure estimation Norimitsu Kubono, Nozomi Hiyoshi, Daiju Akashi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We are considering application of "structural topic model" in order to extract customer insight from member questionnaire on job change site DODA. As a new method of extracting customer insight, we investigated extracting latent co-occurrence graph based on sparse structure estimation method from word co-occurrence graph. We made a word vector from four types of topic model (structural topic model, LDA) and distributed representation (word 2 vec, Glove) and evaluated using sparsity estimation regularization parameters and graph cluster. Furthermore, we interpreted the graph cluster as a topic, defined document vectors, and also applied feasibility study to document clustering. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Customer Insight / Structural Topic Model / co-occurrence graph / sparse structure learning / Graphical Lasso / feature selection / graph clustering / document clustering |
| Paper # | NLC2018-19 |
| Date of Issue | 2018-08-30 (NLC) |
| Conference Information | |
| Committee | NLC / IPSJ-DC |
|---|---|
| Conference Date | 2018/9/6(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Seikei University |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | The Thirteenth Text Analytics Symposium |
| Chair | Takeshi Sakaki(Hottolink) / Michiko Oba(Hitachi) |
| Vice Chair | Mitsuo Yoshida(Toyohashi Univ. of Tech.) / Kazutaka Shimada(Kyushu Inst. of Tech.) |
| Secretary | Mitsuo Yoshida(Ryukoku Univ.) / Kazutaka Shimada(NTT) / (Kyushu Univ.) |
| Assistant | Takeshi Kobayakawa(NHK) / Hiroki Sakaji(Univ. of Tokyo) |
| Paper Information | |
| Registration To | Technical Committee on Natural Language Understanding and Models of Communication / Special Interest Group on Document Communication |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Latent co-occurrence words graph extraction using sparse structure estimation |
| Sub Title (in English) | Comparison of word vectors between topic model and distributed representation |
| Keyword(1) | Customer Insight |
| Keyword(2) | Structural Topic Model |
| Keyword(3) | co-occurrence graph |
| Keyword(4) | sparse structure learning |
| Keyword(5) | Graphical Lasso |
| Keyword(6) | feature selection |
| Keyword(7) | graph clustering |
| Keyword(8) | document clustering |
| 1st Author's Name | Norimitsu Kubono |
| 1st Author's Affiliation | PERSOL CAREER(PERSOL CAREER) |
| 2nd Author's Name | Nozomi Hiyoshi |
| 2nd Author's Affiliation | PERSOL CAREER(PERSOL CAREER) |
| 3rd Author's Name | Daiju Akashi |
| 3rd Author's Affiliation | PERSOL CAREER(PERSOL CAREER) |
| Date | 2018-09-06 |
| Paper # | NLC2018-19 |
| Volume (vol) | vol.118 |
| Number (no) | NLC-210 |
| Page | pp.pp.51-56(NLC), |
| #Pages | 6 |
| Date of Issue | 2018-08-30 (NLC) |