Keyword Recommendation System Using Contextual Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engines can only recommend keywords that are similar in meaning to the search term, failing to connect semantically dissimilar terms that may be relevant in context, leading to user frustration when accurately describing search queries is difficult.
Innovation Solution
A method and apparatus that receive and process multiple search terms, updating a keyword library by determining a similarity coefficient based on the presence of one term in the search results of another, allowing for the recommendation of contextually related keywords even if they are not similar in meaning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If keywords are recommended based on semantic similarity to the search term, then the recommendation accuracy for clearly defined searches is improved, but the system fails to connect contextually relevant terms that are semantically dissimilar
Solution Approach 1:
The patent introduces a new dimension for keyword recommendation by incorporating user behavior data (search history, click-through rates, dwell time) alongside traditional semantic similarity. This multi-dimensional approach allows the system to connect semantically dissimilar terms that are contextually relevant based on user interactions, resolving the contradiction between precision and versatility.
Solution Approach 2:
The system implements feedback mechanisms by analyzing user search behavior patterns and using this information to refine keyword recommendations. User interactions (clicks, views, subsequent searches) provide feedback that helps the system learn and adapt, enabling it to recommend contextually relevant keywords even when semantic similarity is low, thus improving both accuracy and versatility over time.
2Productivity
If the system only recommends semantically similar keywords, then the recommendation process is simple and fast, but users must manually filter through search results to modify search terms when they cannot accurately describe what they are searching for
Solution Approach 1:
The system performs self-service by automatically analyzing user search behavior and proactively recommending relevant keywords without requiring manual filtering. The system monitors user interactions with search results and autonomously generates personalized keyword recommendations, reducing user effort while maintaining high search efficiency.
Solution Approach 2:
The system performs preliminary actions by pre-processing and analyzing user search patterns, click behavior, and dwell time data in advance. This preparation enables the system to quickly generate accurate keyword recommendations when users perform new searches, improving both ease of operation and productivity by eliminating the need for manual filtering.
3Adaptability or versatility
If the keyword library is updated with new terms based on user behavior, then the system can recommend contextually relevant keywords, but the complexity of maintaining and updating the keyword library increases
Solution Approach 1:
The keyword library performs self-service by automatically updating itself based on user behavior data. The system continuously monitors search patterns, clicks, and dwell times, and autonomously adds new contextually relevant keywords to the library without requiring manual curation. This self-updating mechanism increases library coverage while minimizing maintenance complexity.
Solution Approach 2:
The system uses feedback from user interactions to continuously refine and update the keyword library. User behavior data (what users search for, what they click on, how long they stay on pages) provides feedback that automatically triggers library updates, enabling the system to adapt to changing user needs while keeping maintenance complexity manageable through automated processes.
Data Source
AI summary
A method for recommending keywords can receive a first search term entered by a user, search a keyword library comprising a plurality of keywords and retrieve a preset number of keywords based on a similarity coefficient between each keyword and the first search term. After receiving a second search term entered by the user, the method obtains a correlation value between the second search term and the first search term based on whether a webpage in a search result of the first search term visited by the user includes the second search term, and determines the similarity coefficient between the second search term and the first search term in accordance with the correlation value. And then, the method updates the keyword library to save the similarity coefficient between the second search term and the first search term.