Research Article
Prediction of Patent Value Based on Machine Learning Algorithms
@INPROCEEDINGS{10.4108/eai.2-6-2023.2334602, author={Haiying Ren and Chuangchuang Sun}, title={Prediction of Patent Value Based on Machine Learning Algorithms}, proceedings={Proceedings of the 2nd International Conference on Information Economy, Data Modeling and Cloud Computing, ICIDC 2023, June 2--4, 2023, Nanchang, China}, publisher={EAI}, proceedings_a={ICIDC}, year={2023}, month={8}, keywords={patent value prediction; knowledge network embeddedness; machine learning; extratrees}, doi={10.4108/eai.2-6-2023.2334602} }- Haiying Ren
Chuangchuang Sun
Year: 2023
Prediction of Patent Value Based on Machine Learning Algorithms
ICIDC
EAI
DOI: 10.4108/eai.2-6-2023.2334602
Abstract
Predicting the value of patents is crucial for individuals and enterprises to make informed decisions during the patent application and commercialization process. But little research has considered the role of prior knowledge play in the prediction of patent value. This paper selects and designs variables of the knowledge network embeddedness to represent the association between focal patents’ knowledge and the prior domain knowledge from the knowledge recombination perspective. Then use multiple machine learning models to predict patent value proxied by patent transfers. The feasibility of this method is tested with sample patents in the neural network field. The results show that the ExtraTrees achieves the best prediction accuracy of 84.4%.
