Attention-Based Chinese Word Embedding

Liang, Yue; Zhang, Wei; Yang, Kehua

doi:10.1007/978-3-030-00015-8_24

Cited by 3 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhu et al [34] proposed a text classification method based on the dependencies and the Tongyici Cilin semantic dictionary. Recent studies have shown that the internal composition of the Chinese word provides rich semantic information for Chinese word representation [35]. Chinese characters have semantic information.…”

Section: Introductionmentioning

confidence: 99%

Target Mining and Recognition of Product Form Innovation Design Based on Image Word Similarity Model

Zhang

Liu

Zhang

et al. 2022

Advances in Mathematical Physics

View full text Add to dashboard Cite

Product Kansei image design is one of the research hotspot of product emotional design. Due to the subjectivity, low efficiency, and low level of intelligence in the existing product form innovation design methods in the mining of design goals. This study combines the semantic dictionary of Tongyici Cilin with Kansei engineering theory and uses clustering analysis algorithm, semantic difference method, and word similarity calculation method to realize product Kansei image design. Tongyici Cilin is a computable Chinese semantic dictionary. In this study, we innovatively introduced Tongyici Cilin into the image word similarity calculation in product image design. First, the product image design process based on Tongyici Cilin is proposed. Then, we establish a model of image word similarity calculation using the common distance, difference distance, common adjustment parameter, and differential adjustment parameter. By comparing with international standard data, it is confirmed that the image word similarity calculation model proposed in this article is effective and efficient. Using the sedan image design of middle-aged, middle-income men as an example, the sedan form style of each target image was successfully derived from the Internet-based questionnaire. Based on the case studies, we determined that it is effective to use the Tongyici Cilin semantic dictionary to determine the target image and improve the efficiency of product image design.

show abstract

Section: Introductionmentioning

confidence: 99%

Target Mining and Recognition of Product Form Innovation Design Based on Image Word Similarity Model

Zhang

Liu

Zhang

et al. 2022

Advances in Mathematical Physics

View full text Add to dashboard Cite

show abstract

Learning Chinese Word Embeddings By Discovering Inherent Semantic Relevance in Sub-characters

Zhang

Yuan

et al. 2022

Proceedings of the 31st ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

Using NLP in openEHR archetypes retrieval to promote interoperability: a feasibility study in China

Sun

Zhang

et al. 2021

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

Background With the development and application of medical information system, semantic interoperability is essential for accurate and advanced health-related computing and electronic health record (EHR) information sharing. The openEHR approach can improve semantic interoperability. One key improvement of openEHR is that it allows for the use of existing archetypes. The crucial problem is how to improve the precision and resolve ambiguity in the archetype retrieval. Method Based on the query expansion technology and Word2Vec model in Nature Language Processing (NLP), we propose to find synonyms as substitutes for original search terms in archetype retrieval. Test sets in different medical professional level are used to verify the feasibility. Result Applying the approach to each original search term (n = 120) in test sets, a total of 69,348 substitutes were constructed. Precision at 5 (P@5) was improved by 0.767, on average. For the best result, the P@5 was up to 0.975. Conclusions We introduce a novel approach that using NLP technology and corpus to find synonyms as substitutes for original search terms. Compared to simply mapping the element contained in openEHR to an external dictionary, this approach could greatly improve precision and resolve ambiguity in retrieval tasks. This is helpful to promote the application of openEHR and advance EHR information sharing.

show abstract

Attention-Based Chinese Word Embedding

Cited by 3 publications

References 10 publications

Target Mining and Recognition of Product Form Innovation Design Based on Image Word Similarity Model

Target Mining and Recognition of Product Form Innovation Design Based on Image Word Similarity Model

Learning Chinese Word Embeddings By Discovering Inherent Semantic Relevance in Sub-characters

Using NLP in openEHR archetypes retrieval to promote interoperability: a feasibility study in China

Contact Info

Product

Resources

About