Automatic Aggregation by Joint Modeling of Aspects and Values
We present a model for aggregation of product review snippets by joint aspect identification and sentiment analysis. Our model simultaneously identifies an underlying set of ratable aspects presented in the reviews of a product (e.g., sushi and miso for a Japanese restaurant) and determines the corr...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Association for the Advancement of Artificial Intelligence
2013
|
| Online Access: | http://hdl.handle.net/1721.1/78670 https://orcid.org/0000-0002-2921-8201 |
| _version_ | 1826199155895173120 |
|---|---|
| author | Sauper, Christina Barzilay, Regina |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Sauper, Christina Barzilay, Regina |
| author_sort | Sauper, Christina |
| collection | MIT |
| description | We present a model for aggregation of product review snippets by joint aspect identification and sentiment analysis. Our model simultaneously identifies an underlying set of ratable aspects presented in the reviews of a product (e.g., sushi and miso for a Japanese restaurant) and determines the corresponding sentiment of each aspect. This approach directly enables discovery of highly-rated or inconsistent aspects of a product. Our generative model admits an efficient variational mean-field inference algorithm. It is also easily extensible, and we describe several modifications and their effects on model structure and inference. We test our model on two tasks, joint aspect identification and sentiment analysis on a set of Yelp reviews and aspect identification alone on a set of medical summaries. We evaluate the performance of the model on aspect identification, sentiment analysis, and per-word labeling accuracy. We demonstrate that our model outperforms applicable baselines by a considerable margin, yielding up to 32% relative error reduction on aspect identification and up to 20% relative error reduction on sentiment analysis. |
| first_indexed | 2024-09-23T11:15:34Z |
| format | Article |
| id | mit-1721.1/78670 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T11:15:34Z |
| publishDate | 2013 |
| publisher | Association for the Advancement of Artificial Intelligence |
| record_format | dspace |
| spelling | mit-1721.1/786702022-10-01T02:23:47Z Automatic Aggregation by Joint Modeling of Aspects and Values Sauper, Christina Barzilay, Regina Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Sauper, Christina Barzilay, Regina We present a model for aggregation of product review snippets by joint aspect identification and sentiment analysis. Our model simultaneously identifies an underlying set of ratable aspects presented in the reviews of a product (e.g., sushi and miso for a Japanese restaurant) and determines the corresponding sentiment of each aspect. This approach directly enables discovery of highly-rated or inconsistent aspects of a product. Our generative model admits an efficient variational mean-field inference algorithm. It is also easily extensible, and we describe several modifications and their effects on model structure and inference. We test our model on two tasks, joint aspect identification and sentiment analysis on a set of Yelp reviews and aspect identification alone on a set of medical summaries. We evaluate the performance of the model on aspect identification, sentiment analysis, and per-word labeling accuracy. We demonstrate that our model outperforms applicable baselines by a considerable margin, yielding up to 32% relative error reduction on aspect identification and up to 20% relative error reduction on sentiment analysis. National Science Foundation (U.S.) (CAREER Grant IIS-0448168) National Institutes of Health (U.S.) (Grant 5-R01-LM009723-02) United States. Air Force Research Laboratory (Contract FA8750-09-C-017) 2013-05-02T15:03:37Z 2013-05-02T15:03:37Z 2013-01 2012-03 Article http://purl.org/eprint/type/JournalArticle 1943-5037 1076-9757 http://hdl.handle.net/1721.1/78670 Sauper, Christina, and Regina Barzilay. "Automatic Aggregation by Joint Modeling of Aspects and Values." Journal of Artificial Intelligence Research 46 (2013): 89-127. ©2013 AI Access Foundation, Inc. https://orcid.org/0000-0002-2921-8201 en_US http://dx.doi.org/10.1613/jair.3647 Journal of Artificial Intelligence Research Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Association for the Advancement of Artificial Intelligence AI Access Foundation |
| spellingShingle | Sauper, Christina Barzilay, Regina Automatic Aggregation by Joint Modeling of Aspects and Values |
| title | Automatic Aggregation by Joint Modeling of Aspects and Values |
| title_full | Automatic Aggregation by Joint Modeling of Aspects and Values |
| title_fullStr | Automatic Aggregation by Joint Modeling of Aspects and Values |
| title_full_unstemmed | Automatic Aggregation by Joint Modeling of Aspects and Values |
| title_short | Automatic Aggregation by Joint Modeling of Aspects and Values |
| title_sort | automatic aggregation by joint modeling of aspects and values |
| url | http://hdl.handle.net/1721.1/78670 https://orcid.org/0000-0002-2921-8201 |
| work_keys_str_mv | AT sauperchristina automaticaggregationbyjointmodelingofaspectsandvalues AT barzilayregina automaticaggregationbyjointmodelingofaspectsandvalues |