Feature word extraction method and device, electronic equipment, computer readable storage medium and computer program product

By determining the weight based on the number and total number of words in the text in the feature word extraction method, and adjusting the weight in combination with the preset threshold, the problem of low accuracy in feature word extraction in the prior art is solved, and a more accurate text similarity calculation is achieved.

CN119692344APending Publication Date: 2025-03-25MASHANG CONSUMER FINANCE CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202411669886.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-20
Publication Date
2025-03-25

AI Technical Summary

Technical Problem

In the prior art, the word frequency-inverse document frequency algorithm is used to extract feature words for texts with low accuracy, especially when dealing with the influence of new words and word order.

Method used

In the feature word extraction method, the first weight is determined based on the number and total number of words in the first text, and when the words are not included in the text set, the second weight is determined according to the preset threshold value, so that the weight of the word is dynamically adjusted, and the feature words are extracted in combination with the first weight and the second weight.

Benefits of technology

Improve the accuracy of feature word extraction, so that important but perhaps not widely used words can also be selected as feature words, enhancing the accuracy of text similarity calculation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a feature word extraction method and device, electronic equipment, a computer readable storage medium and a computer program product. The method comprises the steps of determining first weights of first words in a first text based on a first number of the first words in the first text and a total number of the words in the first text; under the condition that the first word is not contained in the text set, determining a second weight of the first word in the first text based on the first quantity and a preset first quantity threshold value; the number of texts in the text set is greater than a preset second number threshold; and extracting feature words from the first text based on the first weight and the second weight. According to the invention, the accuracy of feature word extraction can be improved.
Need to check novelty before this filing date? Find Prior Art