Evaluating style transfer in natural language
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/119734 |
| _version_ | 1826189865892446208 |
|---|---|
| author | Matthews, Nicholas (Nicholas J.) |
| author2 | Regina Barzilay. |
| author_facet | Regina Barzilay. Matthews, Nicholas (Nicholas J.) |
| author_sort | Matthews, Nicholas (Nicholas J.) |
| collection | MIT |
| description | Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018. |
| first_indexed | 2024-09-23T08:25:54Z |
| format | Thesis |
| id | mit-1721.1/119734 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T08:25:54Z |
| publishDate | 2018 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1197342019-04-09T18:17:23Z Evaluating style transfer in natural language Matthews, Nicholas (Nicholas J.) Regina Barzilay. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Electrical Engineering and Computer Science. Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Cataloged from student-submitted PDF version of thesis. Includes bibliographical references (pages 46-47). Style transfer is an active area of research growing in popularity in the Natural Language setting. The goal of this thesis is present a comprehensive review of style transfer tasks used to date, analyze these tasks, and delineate important properties and candidate tasks for future methods researchers. Several challenges still exist, including the difficulty of distinguishing between content and style in a sentence. While some state of the art models attempt to overcome this problem, even tasks as simple as sentiment transfer are still non-trivial. Problems of granularity, transferability, and distinguishability have yet to be solved. I provide a comprehensive analysis of the popular sentiment transfer task along with a number of metrics that highlight its shortcomings. Finally, I introduce possible new tasks for consideration, news outlet style transfer and non-parallel error correction, and provide similar analysis for the feasibility of using these tasks as style transfer baselines. by Nicholas Matthews. M. Eng. 2018-12-18T19:47:45Z 2018-12-18T19:47:45Z 2018 2018 Thesis http://hdl.handle.net/1721.1/119734 1078688749 eng MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission. http://dspace.mit.edu/handle/1721.1/7582 47 pages application/pdf Massachusetts Institute of Technology |
| spellingShingle | Electrical Engineering and Computer Science. Matthews, Nicholas (Nicholas J.) Evaluating style transfer in natural language |
| title | Evaluating style transfer in natural language |
| title_full | Evaluating style transfer in natural language |
| title_fullStr | Evaluating style transfer in natural language |
| title_full_unstemmed | Evaluating style transfer in natural language |
| title_short | Evaluating style transfer in natural language |
| title_sort | evaluating style transfer in natural language |
| topic | Electrical Engineering and Computer Science. |
| url | http://hdl.handle.net/1721.1/119734 |
| work_keys_str_mv | AT matthewsnicholasnicholasj evaluatingstyletransferinnaturallanguage |