Language Relevance Scoring via Click Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately identifying languages relevant to resources on the internet, particularly for non-textual content and multilingual resources, making it difficult to determine language relevance for users with different linguistic preferences.
Innovation Solution
A method involving a data processing apparatus that selects a resource, accesses click data to identify search engine user interface languages, determines language relevance scores based on request weights, and selects languages as relevant if scores exceed a threshold, considering factors like user understanding, visit duration, and search result positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional language detection methods are used, then the process is simple, but the accuracy of identifying relevant languages is low
Solution Approach 1:
The patent uses search engine user interface language as an intermediary to infer the relevant languages for resources. Instead of directly analyzing resource content, the system observes which language interfaces users employ when accessing resources, using this behavioral data as a mediator to determine language relevance with high accuracy.
Solution Approach 2:
The system implements feedback loops where user interaction data (click data, visit duration, search result positions) is continuously collected and used to refine language relevance scores. This feedback mechanism allows the system to improve language identification accuracy over time by learning from actual user behavior patterns.
2Measurement precision
If click data analysis is performed for each request, then language relevance accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary analysis by pre-calculating and storing language relevance scores based on historical click data. When a search query is made, the system can quickly retrieve pre-computed language relevance information rather than analyzing all click data in real-time, significantly reducing processing time while maintaining accuracy.
Solution Approach 2:
The patent changes the parameter of analysis from individual request-level detailed analysis to aggregated statistical analysis of click patterns. By transforming raw click data into summarized language relevance scores and probabilities, the system reduces the computational complexity and processing time while preserving the essential information needed for accurate language identification.
3Measurement precision
If multiple language relevance scores are calculated, then multilingual resource disambiguation improves, but computational complexity increases
Solution Approach 1:
The system applies local quality by calculating language relevance scores specifically for each identified search engine user interface language rather than uniformly analyzing all possible languages. This targeted approach focuses computational resources on the languages actually relevant to the resource, improving disambiguation accuracy while reducing unnecessary computational complexity.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying languages that are relevant to resources. In one aspect, a method includes selecting in a data processing apparatus a first resource; accessing click data that identifies, for each of a plurality of requests for the first resource, a respective search engine user interface from which the request was received; identifying a search engine user interface language for each of the plurality of requests based on the click data; determining a respective language relevance score for the first resource for each identified search engine user interface language; and selecting one or more languages as being relevant to the first resource based on the language relevance scores.