CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding
The numerical transformation of text is a key step in natural language processing tasks, among which the word embedding model is the most representative. However, the word embedding model is insufficient in representing unregistered and low-frequency words, and the character-level embedding model ma...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9536576/ |
| _version_ | 1819293418175594496 |
|---|---|
| author | Xiaoming Fan Tuo Shi Jiayan Cai Binjun Wang |
| author_facet | Xiaoming Fan Tuo Shi Jiayan Cai Binjun Wang |
| author_sort | Xiaoming Fan |
| collection | DOAJ |
| description | The numerical transformation of text is a key step in natural language processing tasks, among which the word embedding model is the most representative. However, the word embedding model is insufficient in representing unregistered and low-frequency words, and the character-level embedding model makes up for it. Most Chinese character-level models focus on the independent use of Chinese character features such as strokes, radicals, and pinyin, or the shallow correlations between some features, while the inherent correlations among different features such as pronunciation, glyph, stroke order, and word frequency are not fully utilized. Through the statistical analyses of various features of Chinese characters, this paper proposes a precoding method based on Character Helix Alternative Representation Model (CHARM), which can realize the reversible mapping of Chinese characters or words to English-like sequences, and the advantage of this method is verified in three tasks: text classification, named entity recognition and machine translation. Experimental results on several test sets show that the model performs well, and can be a replacement character-level corpus for the original Chinese text. |
| first_indexed | 2024-12-24T04:10:06Z |
| format | Article |
| id | doaj.art-8118bc555003430c9ba7a11aeec18d20 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-24T04:10:06Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-8118bc555003430c9ba7a11aeec18d202022-12-21T17:16:06ZengIEEEIEEE Access2169-35362021-01-01912953912955110.1109/ACCESS.2021.31121909536576CHARM: An Improved Method for Chinese Precoding and Character-Level EmbeddingXiaoming Fan0https://orcid.org/0000-0003-2933-8884Tuo Shi1Jiayan Cai2https://orcid.org/0000-0002-9276-7385Binjun Wang3School of Information and Cyber Security, People’s Public Security University of China, Beijing, ChinaBeijing Police College, Beijing, ChinaBeijing Police College, Beijing, ChinaSchool of Information and Cyber Security, People’s Public Security University of China, Beijing, ChinaThe numerical transformation of text is a key step in natural language processing tasks, among which the word embedding model is the most representative. However, the word embedding model is insufficient in representing unregistered and low-frequency words, and the character-level embedding model makes up for it. Most Chinese character-level models focus on the independent use of Chinese character features such as strokes, radicals, and pinyin, or the shallow correlations between some features, while the inherent correlations among different features such as pronunciation, glyph, stroke order, and word frequency are not fully utilized. Through the statistical analyses of various features of Chinese characters, this paper proposes a precoding method based on Character Helix Alternative Representation Model (CHARM), which can realize the reversible mapping of Chinese characters or words to English-like sequences, and the advantage of this method is verified in three tasks: text classification, named entity recognition and machine translation. Experimental results on several test sets show that the model performs well, and can be a replacement character-level corpus for the original Chinese text.https://ieeexplore.ieee.org/document/9536576/Chinese textcharacter levelrepresentationprecoding |
| spellingShingle | Xiaoming Fan Tuo Shi Jiayan Cai Binjun Wang CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding IEEE Access Chinese text character level representation precoding |
| title | CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding |
| title_full | CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding |
| title_fullStr | CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding |
| title_full_unstemmed | CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding |
| title_short | CHARM: An Improved Method for Chinese Precoding and Character-Level Embedding |
| title_sort | charm an improved method for chinese precoding and character level embedding |
| topic | Chinese text character level representation precoding |
| url | https://ieeexplore.ieee.org/document/9536576/ |
| work_keys_str_mv | AT xiaomingfan charmanimprovedmethodforchineseprecodingandcharacterlevelembedding AT tuoshi charmanimprovedmethodforchineseprecodingandcharacterlevelembedding AT jiayancai charmanimprovedmethodforchineseprecodingandcharacterlevelembedding AT binjunwang charmanimprovedmethodforchineseprecodingandcharacterlevelembedding |