Longest Common Subsequence Document Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for classifying documents in intranets are time-consuming and prone to errors, particularly when applying content management policies, as they often require manual labeling or rely on complex machine learning techniques that are difficult to understand and may produce false predictions.
Innovation Solution
The use of a longest common word sequence (LCS) as a content pattern to automatically classify documents, allowing for efficient and explainable grouping and labeling by identifying sequences of words common to similar documents, thereby reducing the need for manual labeling and minimizing false predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual labeling is used to classify documents, then classification accuracy can be maintained, but time consumption and labor costs increase significantly
Solution Approach 1:
The system enables documents to classify themselves automatically by extracting and comparing their own content features with predefined classification rules, eliminating the need for manual human intervention in the classification process while maintaining accuracy through rule-based automated decision making
Solution Approach 2:
The patent replaces the mechanical manual labeling process with an automated computer-based system that uses natural language processing and pattern matching algorithms to perform classification, substituting human cognitive work with computational processes
2Extent of automation
If machine learning models are used for document classification, then automation is improved, but the models become difficult to understand and may produce false predictions
Solution Approach 1:
The patent uses simple, interpretable classification rules rather than complex machine learning models, sacrificing the ability to learn from unlimited data in exchange for rules that are easy to understand, modify, and explain, effectively using simple disposable rules instead of complex persistent models
Solution Approach 2:
The system changes the approach from using complex statistical parameters in machine learning models to using simple text-based features and rule conditions that are directly interpretable, transforming the classification mechanism from black-box statistical analysis to transparent rule-based decision making
3Adaptability or versatility
If complex machine learning analysis is used, then classification capability is improved, but explainability of predictions deteriorates
Solution Approach 1:
The patent introduces classification rules as an intermediary layer between raw document content and classification results, making the decision-making process transparent and explainable by showing which specific rules matched the document features, rather than directly outputting results from complex statistical models
Data Source
AI summary
Computer systems, devices, and associated methods of content pattern based automatic document classification are disclosed herein. In one embodiment, a method includes receiving a document and a sequence of words corresponding to a document class having a class label from a network storage. The method also includes determining a longest common subsequence of words between the words in the document and the sequence of words and calculating a similarity percentage between the document and the sequence of words based on the determined longest common subsequence. When the calculated similarity percentage is above a threshold, the class label corresponding to the document class is automatically applied to the received document in the network storage.