Incorporating Chinese Characters of Words for Lexical Sememe Prediction

Jin, Huiming; Zhu, Hao; Liu, Zhiyuan; Xie, Rong; Sun, Maosong; Lin, Feng; Lin, Leyu

doi:10.48550/arxiv.1806.06349

Cited by 2 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SPWE+SPSE [22] 40.25 CSP [23] 46.53 LD + seq2seq [35] 25.85 SCorP [27] 56.32 ASPSW [25] 61.02 SFCW 67.70…”

Section: Methods F1mentioning

confidence: 99%

“…Bai [26] and Du [27] introduced dictionary definitions into sememe prediction, enabling the selection of more accurate sememes for words with given definitions. Jin [23], Lyu [24], and Kang [25] improved sememe prediction models by incorporating the character information of Chinese characters. Wang et al [29] utilized the expert-annotated knowledge base CilinE for sememe prediction, achieving better performance.…”

Section: Sememe Predictionmentioning

confidence: 99%

“…• SPWE [22]: A word embedding-based method that recommends sememes based on similar words in the word vector space. • CSP [23]: An ensemble model consisting of four sub-models, including SPWCF and SPCSE using internal word information and SPWE and SPSE using external word information. • LD + seq2seq [35]: A sequence-to-sequence model utilizing the text definition.…”

Section: Baselinesmentioning

confidence: 99%

“…Later approaches enhanced prediction accuracy by incorporating additional semantic information during training. Jin [23], Lyu [24], and Kang [25], among others, utilized the character information of Chinese characters during prediction tasks. Bai [26] and Du [27] introduced dictionary definitions, and Qi [28] integrated word embeddings from other languages.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Sememe Prediction Method Based on the Central Word of a Semantic Field

Luo,

Cui

2024

Electronics

View full text Add to dashboard Cite

A “sememe” is an indivisible minimal unit of meaning in linguistics. Manually annotating sememes in words requires a significant amount of time, so automated sememe prediction is often used to improve efficiency. Semantic fields serve as crucial mediators connecting the semantics between words. This paper proposes an unsupervised method for sememe prediction based on the common semantics between words and semantic fields. In comparison to methods based on word vectors, this approach demonstrates a superior ability to align the semantics of words and sememes. We construct various types of semantic fields through ChatGPT and design a semantic field selection strategy to adapt to different scenario requirements. Subsequently, following the order of word–sense–sememe, we decompose the process of calculating the semantic sememe similarity between semantic fields and target words. Finally, we select the word with the highest average semantic sememe similarity as the central word of the semantic field, using its semantic primes as the predicted result. On the BabelSememe dataset constructed based on the sememe knowledge base HowNet, the method of semantic field central word (SFCW) achieved the best results for both unstructured and structured sememe prediction tasks, demonstrating the effectiveness of this approach. Additionally, we conducted qualitative and quantitative analyses on the sememe structure of the central word.

show abstract

“…SPWE+SPSE [22] 40.25 CSP [23] 46.53 LD + seq2seq [35] 25.85 SCorP [27] 56.32 ASPSW [25] 61.02 SFCW 67.70…”

Section: Methods F1mentioning

confidence: 99%