Cross-Lingual Product Search Using Translatability Classification

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 multilingual environments where words can have different meanings or varieties.

Innovation Solution

A system and method that processes cross-lingual search queries using a translatability classifier module to determine user intent and a language translator module to provide accurate translations or maintain original queries based on class labels, addressing ambiguity through a cache of class-translation pairs and user interaction data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a conventional search engine guesses the language of cross-lingual queries without user context, then the system can operate with simple monolingual processing, but the language identification becomes inaccurate and produces irrelevant results

Engineering Contradiction:

Improvesearch engine processing complexityVSAvoidlanguage identification accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system performs preliminary language identification and translatability classification before executing the search query. By analyzing the query in advance and determining the source language and translatability class, the system prepares the appropriate translation and search strategy beforehand, ensuring accurate language handling without adding complexity to the core search engine.

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If the system translates all cross-lingual queries to the primary language, then search results can be consistently generated in the primary language, but user intent may be misinterpreted due to language ambiguity and polysemy

Engineering Contradiction:

Improvesearch result consistencyVSAvoiduser intent interpretation accuracy

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system applies different processing qualities to different types of queries based on their translatability class. For queries with clear meanings (Ignore class), the original query is used directly. For queries with ambiguous meanings (OptIn/OptOut classes), translation is applied with user control options. This localized approach ensures that translation is applied only where appropriate, maintaining both consistency and accuracy.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system incorporates user feedback through control elements that allow users to indicate whether they want translation or not. This feedback mechanism enables the system to learn from user preferences and adjust its translation behavior, improving the reliability of user intent interpretation over time while maintaining operational consistency.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If the system provides translation options and control elements to users, then user control over translation is enhanced, but the search process becomes more complex and time-consuming

Engineering Contradiction:

Improveuser control flexibilityVSAvoidsearch process complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system determines the translatability class and prepares translation options in advance before presenting them to the user. By performing this classification work preliminarily, the system reduces the complexity of the user interface and the search process, as the heavy lifting of language analysis is already completed before user interaction begins.

Inventive Principle:

Principle #10Preliminary action

4Productivity

If the system automatically translates cross-lingual queries without user input, then search processing speed is maintained, but the system cannot handle polysemy and cross-language ambiguity effectively

Engineering Contradiction:

Improvesearch processing speedVSAvoidsearch result relevance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system changes the parameter of user involvement based on the translatability class of the query. For Ignore class queries, the system operates in automatic mode with high productivity. For OptIn/OptOut class queries, the system transitions to a mode that allows user control, accepting a slight reduction in processing speed in exchange for significantly improved result relevance and reliability.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12517935B2System and method for performing cross-lingual product searches

Publication Date: 2026.01.06 WALMART APOLLO LLC

AI summary

A method can include determining a translatability class label and a class-associated search query for a cross-lingual search query. Determining the translatability class label and the class-associated search query can include determining whether a first class-translation pair for the cross-lingual search query exists in a memory module. If the first class-translation pair exists in the memory module, the method further can include retrieving the first class-translation pair for the cross-lingual search query from the memory module. When no class-translation pair for the cross-lingual search query exists in the memory module, the method additionally can include: determining the translatability class label, and determining the class-associated search query based on the translatability class label. The method further can include determining search results for the cross-lingual search query based on the translatability class label and the class-associated search query, and transmitting the search results to a user based on the translatability class label. Other embodiments are disclosed.