Automatic Document Tagging via Weighted Co-occurrence Probability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for adding tags to documents are inefficient, as manual tagging is labor-intensive for large volumes, keyword tagging may not select suitable tags, and socialized tagging leads to inconsistent and disordered tags due to varying user standards.
Innovation Solution
A method and apparatus that determine candidate tag words by calculating the weighted co-occurrence probability with characteristic words from a corpus, allowing for intelligent selection of tags beyond just key words, using modules for candidate tag determination, co-occurrence probability calculation, weight assignment, and tag word addition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual tagging is used, then tag accuracy can be ensured, but productivity deteriorates due to labor-intensive processes for large volumes of documents
Solution Approach 1:
The patent introduces an automatic tagging system that acts as an intermediary between manual tagging and document processing. The system uses co-occurrence probability calculations and characteristic word analysis to generate candidate tags, which are then presented to users for confirmation. This intermediary approach maintains high tag accuracy through user verification while dramatically improving productivity by automating the initial tag generation process.
Solution Approach 2:
The system enables documents to essentially tag themselves by automatically analyzing their content and generating relevant tags based on co-occurrence probabilities with characteristic words. This self-service mechanism reduces the need for manual intervention while maintaining reasonable tagging accuracy, allowing the system to handle large volumes of documents efficiently.
2Productivity
If keyword tagging is used, then productivity is improved through automatic abstraction, but measurement precision deteriorates as only key words occurring in the document are chosen
Solution Approach 1:
The patent extends the tagging dimension beyond just words present in the document by incorporating co-occurrence probability analysis. The system calculates probabilities based on characteristic words and their relationships with candidate tags, adding a probabilistic dimension to tag selection. This allows the system to suggest tags that may not literally appear in the document but are highly relevant based on statistical relationships, thereby improving tag relevance while maintaining automatic processing speed.
3Adaptability or versatility
If socialized tagging is used, then adaptability is improved as users add their own tags, but measurement precision deteriorates due to inconsistent standards from different users
Solution Approach 1:
The patent implements homogeneity by establishing a standardized tagging framework based on co-occurrence probability calculations. All users interact with the same systematic approach to tag generation and selection, ensuring consistent tagging standards across different users and documents. The system presents a unified set of candidate tags derived from objective statistical analysis, eliminating the inconsistency that arises when users independently choose tags based on personal judgment.
Data Source
AI summary
A method and apparatus for automatically adding a tag to a document are provided. The method comprises: determining a plurality of candidate tag words corresponding to the document; determining a corpus comprising a plurality of texts; selecting commonly-used words from the corpus as characteristic words; determining, for each of the characteristic words and each of the candidate tag words, a probability for co-occurrence of the candidate tag word with the characteristic word; abstracting characteristic words from the document, and calculating a weight for each of the abstracted characteristic words; and calculating, in the corpus, a weighted probability for co-occurrence of each of the candidate tag words with all of the characteristic words abstracted from the document; selecting the candidate tag word with a high weighted co-occurrence probability as a tag word to be added to the document.