Search Query Classification Using ML and Rules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of the World-Wide Web and the ever-changing nature of webpages make it difficult for search engines to accurately understand and match user search intentions, leading to challenges in getting resources highly ranked in organic search results.
Innovation Solution
A computer-implemented method for classifying search queries into categories such as informational, navigational, and transactional using predetermined rules and a machine learning module trained on similarity values, allowing for the classification of search queries into user intent and buying cycle categories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If search engines crawl and index an indefinite number of webpages with ever-changing contents, then the coverage and comprehensiveness of search results improve, but the difficulty of understanding and matching user search intention worsens
Solution Approach 1:
The patent segments search queries into distinct categories (informational, navigational, transactional) based on user intent. This segmentation allows the search engine to handle different types of queries with specialized processing methods, making it easier to understand and match user intentions despite the vast number of indexed webpages.
Solution Approach 2:
The patent changes the parameter of query classification by using machine learning models that analyze multiple features of search queries (such as query length, presence of question words, domain-specific terms) to dynamically categorize queries. This parameter-based classification improves the ability to understand user intent across a large volume of webpages.
2Reliability
If operators optimize webpages using various SEO techniques, then the ranking and visibility of resources improve, but the complexity of matching search queries with relevant resources worsens
Solution Approach 1:
The patent applies preliminary classification to search queries before the matching process. By categorizing queries into informational, navigational, and transactional types in advance, the system can apply appropriate matching strategies for each category, simplifying the overall matching process while improving ranking accuracy for different query types.
Solution Approach 2:
The patent introduces an intermediary classification layer between the search query and the webpage matching process. This intermediary system uses machine learning to determine query categories and applies different matching algorithms accordingly, acting as a mediator that simplifies the complex task of matching diverse SEO-optimized webpages with varied user intentions.
3Measurement precision
If the system classifies search queries manually, then the accuracy of user intent understanding improves, but the time and cost required for processing increases
Solution Approach 1:
The patent implements a self-service classification system where machine learning models automatically categorize search queries without human intervention. The system learns from training data and autonomously determines query categories, achieving high accuracy while eliminating the time and cost associated with manual classification.
Solution Approach 2:
The patent replaces the mechanical process of manual query classification with an automated machine learning system. The ML models process queries using computational algorithms, substituting human analysts with automated systems that can handle large volumes of queries quickly and consistently, maintaining high accuracy while reducing time and cost.
Data Source
AI summary
A computer-implemented method of classifying a search query in a network comprises: classifying a plurality of search queries into categories, comprising: applying predetermined rules to each of the plurality of search queries, wherein the predetermined rules are indicative of the categories and each of the plurality of search queries is associated with search results in the network; determining, for each of the plurality of search queries, similarity values indicating similarity to each of the categories based on the applied predetermined rules; and training a machine learning module, comprising: applying the machine learning module to a plurality of training sets to a plurality of training sets, wherein each of the plurality of training sets is based on one of the plurality of classified search queries and at least one of the respective one or more similarity values, a corresponding system, computing device and non-transitory computer-readable storage medium.