A patent keywords extraction method using TextRank model with prior public knowledge

Abstract

For large amount of patent texts, how to extract their keywords in an unsupervised way is a very important problem. In existing methods, only the own information of patent texts is analyzed. In this study, an improved TextRank model is proposed, in which prior public knowledge is effectively utilized. Specifically, two following points are first considered: (1) a TextRank network is constructed for each patent text, (2) a prior knowledge network is constructed based on public dictionary data, in which network edges represent the prior interpretation relationship among all dictionary words in dictionary entries. Then, an improved node rank value evaluation formula is designed for TextRank networks of patent texts, in which prior interpretation information in prior knowledge network are introduced. Finally, patent keywords can be extracted by finding top-k node words with higher node rank values. In our experiments, patent text clustering task is used to examine the performance of proposed method, wherein several comparison experiments are executed. Corresponding results demonstrate that, new method can markedly obtain better performance than existing methods for patent keywords extraction task in an unsupervised way.

Bibliometric analysis of Journal of Money Laundering Control: emerging trends and a way forward

“…The supporting information for the network of keywords is shown in Table 7. Every keyword is represented by a node, where the size of the node depicts its occurrence (Huang and Xie, 2022). It means the higher the size of the node, the greater the occurrences of the keyword.…”

Research on the construction method of knowledge graph for programming language

“…TextRank algorithm [13] , derived from PageRank, is a typical graph-based keyword extraction method. It extracts keywords by using co-occurrence information between words in documents, and extracts key sentences of the text by using extractive automatic abstracting method.…”

Multifeature Fusion Keyword Extraction Algorithm Based on TextRank

“…However, it is poorly effective for new words and words that are not included on Wikipedia. Huang Z [10] constructed a prior knowledge network by using public dictionary data and improved the scoring function in the TextRank network by using some prior relations in this network. Kazemi A [11] optimized the state transfer matrix by randomly resetting the weights during iterations of the TextRank graph, which improved the effect of the keyword extraction.…”

“…The final node scores are only related to the state transfer matrix between nodes rather than the initial values of nodes. The final state transfer matrix of the nodes in the optimized word graph is shown in Formula (10), where P ij denotes the probability of node v i transferring to v j . n is the number of nodes.…”