Method for automatically extracting key phrases of patent documents

Abstract

The invention provides a method for automatically extracting key phrases of patent documents. The method includes steps of 1, preprocessing texts; 2, recognizing topic types of patented inventions; 3, extracting alternative key phrases and filtering the same; 4, subjecting the alternative key phrases to weight calculation and selecting the key phrases.

Description

Technical field [0001] The present invention relates to text information processing technology, and more specifically, to a method for automatically extracting key phrases of patent documents. Background technique [0002] With the rapid growth of the number of patent documents, the professional and socialized search of patent documents has become more and more common. Realizing the recall and accuracy of patent document data has become the difficulty and focus of patent document information retrieval. For a long time, the retrieval of patent information using original patent data has often resulted in poor recall and precision and often contradicted each other. Since the original information of patent documents comes from the applicant’s original submissions, there are often a large number of relevant technical information and citation techniques, so in the search process, in order to ensure the recall rate, too many documents will be introduced, a lot of noise data or Noise li...

AI Technical Summary

Problems solved by technology

However, because data processing requires a lot of time, manpower and financial resources, the construction cost is very high, and the processing efficiency is not satisfactory at present

[0004] However, there is currently a lack of professional and accurate indexing tools, and most of them use manual indexing to improve the accuracy rate, making indexing work even more difficult to meet the needs of the current increasing number of patent applications

Chinese invention patent CN1818906A provides an indexing method for patent documents. This method establishes technical classification and keyword correspondence, and provides corrections to improve the accuracy rate. However, this method still relies on manual work and is not fully automatic. The data processed by the method is large and difficult to be practical

Images