Open AccessIntelligent Text Scoring System Based on Deep Learning
Author(s) -
Lin Qi
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1486/3/032052
Subject(s) - computer science , lexicon , sentiment analysis , support vector machine , field (mathematics) , text segmentation , cluster analysis , artificial intelligence , process (computing) , word (group theory) , segmentation , service (business) , natural language processing , information retrieval , listing (finance) , data mining , linguistics , philosophy , mathematics , economy , finance , pure mathematics , economics , operating system
In view of the lack of unified standards for emerging local service industries and the impact on users’ reasonable choices, this thesis takes the field of B&B as an example to design an analysis and scoring system based on massive online data. The system first obtains massive online data, including text-type data, through web crawlers. Then analyze and pre-process the captured information to build a corpus; use Chinese word segmentation and label clustering to mine the keywords in the listing introduction and comments. The sentiment polarity of the tenant reviews was analyzed based on the support vector machine (SVM) and sentiment lexicon. Finally, an optimized scoring system was obtained.