http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Word2vec based Latent Semantic Analysis (W2V-LSA): 새로운 토픽 모델링을 통한 블록체인 기술 연구 트렌드 분석
김수현(Suhyeon Kim),박해청(Haecheong Park),이정혜(Junghye Lee) 대한산업공학회 2018 대한산업공학회 추계학술대회논문집 Vol.2018 No.11
It is an urgent task to analyze trends in Blockchain technology to help establish action plans based on Blockchain, which is one of the core technologies in Industry 4.0. This study provides the trend analysis on Blockchain based on topic modeling of text mining for 231 abstracts of Blockchain-related papers published over the past five years. We developed a new topic modeling method called Word2vec-based Latent Semantic Analysis (W2V-LSA), which is based on Word2vec and Spherical k-means clustering to capture the context of corpus in a better representation. We used W2V-LSA to perform the annual trend analysis of Blockchain research by country and compared the results with Probabilistic LSA. We demonstrated the usefulness of W2V-LSA in terms of accuracy and diversity of the topics captured for the documents. It is believed that W2V-LSA will be a useful alternative for better topic modeling and the trend analysis of W2V-LSA will provide insight and show the direction for the future research on Blockchain technology.