Title | ||
---|---|---|
Semantic Relationship between Abbreviations and the Original Words Based on Word Vectors |
Abstract | ||
---|---|---|
Based on word vectors, this paper studies the semantic relationship between abbreviations and the original words from a quantitative perspective. First, we train word vectors according to the corpus of the People's Daily. Then, from the perspectives of intrinsic evaluation and extrinsic evaluation, the research on the relationship between abbreviations and the original words is carried out. In the intrinsic evaluation, after vectorizing the abbreviations and their original words, we measure their semantic similarity, and compare with the semantic similarity of synonym collection; In the extrinsic evaluation, we propose an abbreviation recognition task, like named entity recognition. Through this task, the degree of mutual substitution between abbreviations and the original words can be described. The above experiments show that after the abbreviations and their original words are expressed by word vectors, there is still a strong semantic relationship between them.
|
Year | DOI | Venue |
---|---|---|
2019 | 10.1145/3319921.3319938 | Proceedings of the 2019 3rd International Conference on Innovation in Artificial Intelligence |
Keywords | DocType | ISBN |
abbreviation, abbreviation recognition, semantic similarity, word vector | Conference | 978-1-4503-6128-6 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jianyu Zheng | 1 | 0 | 0.34 |
Jin Sun | 2 | 0 | 0.34 |
Xin'ge Xiao | 3 | 0 | 0.34 |
Lijiao Yang | 4 | 0 | 1.69 |