Automatic Indexing of Patent Right-claiming Document Based on Deep Learning

摘要

In recent years, there have been more and more applications of deep learning in natural language processing, and people have paid more and more attention to the value embodied in patent. In this paper, based on the value of right-claiming document in the patent, a deep learning tool word2vec is used to convert text information into a set of word embeddings. The word embeddings carry semantic information, so that the quantified metrics the relationships between words. Then, the k-means clustering method is used to extract words whose distance between words is closer to the center of the cluster, so as to achieve the purpose of automatic indexing the right-claiming document.