Text Classification via Spherical Space Vector Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current text classification technologies are inefficient due to lengthy calculations and lack of semantic specificity, resulting in inaccurate classification of texts.
Innovation Solution
The method employs a spherical space model with pre-constructed dimensions equal to the number of categories, using normalized word frequency values to determine the distance between word vectors and category vectors, thereby classifying texts into categories with the shortest distance, optimizing the classification process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the cosine angle method is used to measure similarity between text vectors and category vectors, then the classification can be performed, but the computation becomes lengthy and time-consuming
Solution Approach 1:
The patent changes the mathematical parameter from cosine angle calculation to spherical distance calculation. By representing text and categories as vectors on a unit sphere and computing the spherical distance between them, the system achieves classification without the computationally intensive cosine angle calculations, thus reducing time consumption while maintaining accuracy
Solution Approach 2:
The patent replaces the conventional vector space model with a spherical space model. This substitution transforms the classification mechanism from Euclidean space operations to spherical geometry operations, enabling faster computation through simplified distance metrics while preserving semantic relationships
2Productivity
If conventional text classification methods are used, then the basic classification function is achieved, but the semantic representation lacks specificity and classification accuracy is insufficient
Solution Approach 1:
The patent introduces a spherical dimension to the traditional vector space model. By mapping text vectors onto the surface of a unit sphere and utilizing spherical geometry for distance calculation, the system adds a new dimensional perspective that enhances semantic representation capability and improves classification accuracy while maintaining computational efficiency
Data Source
AI summary
The present disclosure provides a technique of text categorization to simplify and optimize the classification. In one aspect, a method parses a given text into one or more words; determines a word vector in a spherical space model for one of the one or more words, a number of dimensions of the spherical space being equal to a number of categories, each category corresponding to a spherical space category vector; for each category, determines a distance between a sum of word vectors of the one or more words and the respective category vector; and classifies the text into one or more categories with the shortest distance. The present disclosure also provides an apparatus used to implement the method.