Keyword Classification Model Using Word Vector Clustering

Solution Overview

Problem

Content providers face challenges in selecting appropriate keywords for search results, leading to irrelevant outcomes due to broad application across multiple units, resulting in less relevant search results.

Innovation Solution

A classification model is trained to associate keywords with categories by converting them into word vectors, mapping them to a vector space, and performing cluster analysis to identify the nearest cluster, which corresponds to specific entities related to the content provider, such as sub-organizations or departments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If broad terms are selected to apply to multiple units, then the setup process becomes simpler and faster, but the relevance of search results decreases

Engineering Contradiction:

Improveease of keyword selectionVSAvoidrelevance of search results

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent segments the content provider's units into distinct categories and uses a classification model to assign keywords to specific categories rather than applying them broadly across all units. This segmentation allows for more precise keyword-targeted matching while maintaining ease of setup through automated classification.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs self-service by automatically classifying keywords into appropriate categories using a trained classification model, eliminating the need for manual keyword selection and assignment by the content provider. The model autonomously determines the most relevant categories based on the keywords provided.

Inventive Principle:

Principle #25Self-service

2Reliability

If specific plans are developed for each unit, then the relevance of search results improves, but the time and resources required for setup increases

Engineering Contradiction:

Improverelevance of search resultsVSAvoidtime for plan development

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-training a classification model using a comprehensive training set before actual keyword classification is needed. This pre-training enables the system to quickly and accurately classify new keywords into appropriate categories without requiring time-consuming manual plan development for each unit.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system performs self-service by automatically classifying keywords into appropriate categories using a trained classification model, eliminating the need for manual keyword selection and assignment by the content provider. The model autonomously determines the most relevant categories based on the keywords provided.

Inventive Principle:

Principle #25Self-service

3Measurement precision

If manual keyword selection is performed, then the precision of keyword targeting improves, but the complexity and time required for setup increases

Engineering Contradiction:

Improveprecision of keyword targetingVSAvoidcomplexity of plan setup

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs self-service by automatically classifying keywords into appropriate categories using a trained classification model, eliminating the need for manual keyword selection and assignment by the content provider. The model autonomously determines the most relevant categories based on the keywords provided.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual process of keyword selection and category assignment with an automated machine learning-based classification system. The classification model uses neural networks and training data to automatically determine keyword categories, substituting manual mechanical operations with automated intelligent processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10599731B2Method and system of determining categories associated with keywords using a trained model

Publication Date: 2020.03.24 BAIDU USA LLC

AI summary

Described is a technique for associating words used in a search query with categories. This technique aims to produce potentially more relevant search results by improving the associations with words used for a search. A machine learning technique is implemented to train a classification model, which may include a word embedding model. The classification model is trained to receive words as input and to create vectors of the words as output. These word vectors may then be mapped to a vector space and the technique may then perform a cluster analysis of the vectors. Based on the cluster analysis, clusters may be identified and each cluster may be associated with a corresponding category.