Cross-Lingual Query Classification for Search Intent Disambiguation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional search engines struggle with cross-lingual queries due to language ambiguity, leading to incorrect guesses and irrelevant results without user context, especially in regions with diverse language populations.

Innovation Solution

A system and method that includes a translatability classifier module to determine user intent and a language translator module to process cross-lingual search queries, using modules like translatability classifier module 3120 and language translator module 3130 to classify and translate queries based on user intent and historical data, ensuring accurate search results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a conventional search engine guesses the language of cross-lingual queries, then it can process the queries without user input, but it produces incorrect guesses and irrelevant results due to language ambiguity

Engineering Contradiction:

Improvequery processing without user inputVSAvoidlanguage identification accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary classification system that categorizes cross-lingual queries into different types (code-switched, ambiguous, mixed-language) before processing. This intermediary step acts as a mediator between the raw query and the search engine, enabling more accurate language identification by recognizing query patterns and applying appropriate disambiguation strategies for each category.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary analysis of the search query to detect language mixing patterns, identify code-switching points, and classify the query type before executing the search. This preliminary action includes analyzing word sequences, detecting language boundaries, and preparing disambiguation rules in advance, which significantly improves language identification accuracy compared to guessing.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If the search engine processes cross-lingual queries directly without classification, then the system remains simple, but it cannot handle language ambiguity and produces irrelevant results

Engineering Contradiction:

Improvesearch result relevanceVSAvoidsystem structure

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the query processing system into distinct functional modules: a query analysis module that detects language mixing, a classification module that categorizes query types, and a search execution module that applies appropriate handling strategies. This segmentation allows each module to specialize in specific tasks, improving overall reliability while keeping individual modules relatively simple.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system dynamically adjusts its processing approach based on the detected query type. For code-switched queries, it applies specific language boundary detection; for ambiguous queries, it uses disambiguation rules; for mixed-language queries, it segments and processes different language portions separately. This dynamic adaptation improves search result relevance without requiring a completely complex static system.

Inventive Principle:

Principle #15Dynamics

3Adaptability or versatility

If the system translates all cross-lingual queries, then it can provide consistent language output, but it loses original language context and may misrepresent user intent

Engineering Contradiction:

Improvelanguage handling flexibilityVSAvoidoriginal language context

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent applies different processing qualities to different portions of the query based on local language characteristics. Instead of uniformly translating or preserving the entire query, it identifies specific segments in different languages and applies appropriate handling to each segment, maintaining original language context where needed while providing translation only where necessary for search compatibility.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

Rather than translating the entire query upfront, the system inverts the approach by first analyzing and preserving original language segments, then selectively translating only the portions necessary for search execution. This inversion maintains original language context and user intent while still enabling cross-lingual search functionality.

Inventive Principle:

Principle #13The other way round (Inversion)

Data Source

PatentUS12517899B2System and method for processing cross-lingual search queries

Publication Date: 2026.01.06 WALMART APOLLO LLC

AI summary

A method can include determining, via a translatability classifier module, a translatability class label for a cross-lingual search query received, via a computer network, from a user device for a user. The translatability classifier module can be trained to determine the translatability class label among multiple translatability class labels associated with user intentions for the cross-lingual search query based on a respective probability of the cross-lingual search query being associated with each of the multiple translatability class labels. The method further can include determining, via a language translator module, a class-associated search query for the cross-lingual search query based on the translatability class label, as determined. The method additionally can include transmitting, via the computer network, the class-associated search query to a monolingual search engine. Other embodiments are disclosed.