Ensemble classifier based approach for code-mixed cross-script question classification
With an increasing popularity of social-media, people post updates that aid other users in finding answers to their questions. Most of the user-generated data on social-media are in code-mixed or multi-script form, where the words are represented phonetically in a non-native script. We address the p...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Conference Paper |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/87322 http://hdl.handle.net/10220/49465 http://ceur-ws.org/Vol-1737/ |
| _version_ | 1826127234269708288 |
|---|---|
| author | Bhattacharjee, Debjyoti Bhattacharya, Paheli |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Bhattacharjee, Debjyoti Bhattacharya, Paheli |
| author_sort | Bhattacharjee, Debjyoti |
| collection | NTU |
| description | With an increasing popularity of social-media, people post updates that aid other users in finding answers to their questions. Most of the user-generated data on social-media are in code-mixed or multi-script form, where the words are represented phonetically in a non-native script. We address the problem of Question-Classfication on social-media data. We propose an ensemble classifier based approach towards question classification when the questions are written in mixedscript, specifically, the Roman script for the Bengali language. We separately train Random Forests, One-Vs-Rest and k-NN classifiers and then build an ensemble classifier that combines the best from the three worlds. We achieve an accuracy of 82% approximately, suggesting that the method works well in the task. |
| first_indexed | 2024-10-01T07:05:28Z |
| format | Conference Paper |
| id | ntu-10356/87322 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:05:28Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | ntu-10356/873222019-12-06T16:39:30Z Ensemble classifier based approach for code-mixed cross-script question classification Bhattacharjee, Debjyoti Bhattacharya, Paheli School of Computer Science and Engineering FIRE 2016 - Forum for Information Retrieval Evaluation Question Answering System DRNTU::Engineering::Computer science and engineering Mixed Script Information Retrieval With an increasing popularity of social-media, people post updates that aid other users in finding answers to their questions. Most of the user-generated data on social-media are in code-mixed or multi-script form, where the words are represented phonetically in a non-native script. We address the problem of Question-Classfication on social-media data. We propose an ensemble classifier based approach towards question classification when the questions are written in mixedscript, specifically, the Roman script for the Bengali language. We separately train Random Forests, One-Vs-Rest and k-NN classifiers and then build an ensemble classifier that combines the best from the three worlds. We achieve an accuracy of 82% approximately, suggesting that the method works well in the task. Published version 2019-07-25T01:34:37Z 2019-12-06T16:39:30Z 2019-07-25T01:34:37Z 2019-12-06T16:39:30Z 2016 Conference Paper Bhattacharjee, D., & Bhattacharya, P. (2016). Ensemble classifier based approach for code-mixed cross-script question classification. FIRE 2016 - Forum for Information Retrieval Evaluation, 1737, 119-121. https://hdl.handle.net/10356/87322 http://hdl.handle.net/10220/49465 http://ceur-ws.org/Vol-1737/ en © 2016 The Author(s). All rights reserved. This paper was published in CEUR Workshop Proceedings and is made available with permission of 2016 The Author(s). 3 p. application/pdf |
| spellingShingle | Question Answering System DRNTU::Engineering::Computer science and engineering Mixed Script Information Retrieval Bhattacharjee, Debjyoti Bhattacharya, Paheli Ensemble classifier based approach for code-mixed cross-script question classification |
| title | Ensemble classifier based approach for code-mixed cross-script question classification |
| title_full | Ensemble classifier based approach for code-mixed cross-script question classification |
| title_fullStr | Ensemble classifier based approach for code-mixed cross-script question classification |
| title_full_unstemmed | Ensemble classifier based approach for code-mixed cross-script question classification |
| title_short | Ensemble classifier based approach for code-mixed cross-script question classification |
| title_sort | ensemble classifier based approach for code mixed cross script question classification |
| topic | Question Answering System DRNTU::Engineering::Computer science and engineering Mixed Script Information Retrieval |
| url | https://hdl.handle.net/10356/87322 http://hdl.handle.net/10220/49465 http://ceur-ws.org/Vol-1737/ |
| work_keys_str_mv | AT bhattacharjeedebjyoti ensembleclassifierbasedapproachforcodemixedcrossscriptquestionclassification AT bhattacharyapaheli ensembleclassifierbasedapproachforcodemixedcrossscriptquestionclassification |