Open AccessPower patent classification method based on deep neural network
Author(s) -
Min Hua
Publication year - 2021
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 - 10.1088/1742-6596/1848/1/012048
Subject(s) - artificial neural network , preprocessor , artificial intelligence , computer science , power (physics) , construct (python library) , deep learning , patent visualisation , key (lock) , deep neural networks , pattern recognition (psychology) , machine learning , data mining , physics , computer security , quantum mechanics , programming language
Aiming at the problems in power patent classification, a power patent classification method based on deep neural network is proposed. Firstly, the power patent is analyzed for preprocessing. Then, the key words in the patent abstract are extracted to construct a power vocabulary dictionary. Then, the power patent classification model is built according to the deep neural network. Finally, the power patent classification model based on deep neural network is used to classify the power patents, and the accuracy is close to 98%. It can be seen that the deep neural network method can not only realize the classification of power patent, but also has good classification effect.