Language Proficiency Inference via Text and Profile Score Aggregation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for determining a user's language proficiency are inadequate, relying on noisy language classifiers and counters, which fail to accurately assess proficiency in multiple languages or sets of languages, especially when user data is limited and behavior is unreliable.

Innovation Solution

A language proficiency inference system that calculates both text-based and profile-based probability scores using user data and text engagement, aggregating these scores to infer proficiency in multiple languages, providing a more accurate assessment of language skills.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If language classifiers and counters are used to determine user language proficiency, then the system can identify user language preferences, but the accuracy of proficiency assessment deteriorates due to noisy and unreliable user behavior data

Engineering Contradiction:

Improvelanguage proficiency assessment accuracyVSAvoiduser behavior data reliability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent segments the language proficiency assessment into multiple independent components: language classifier results, text-based probability scores from user-authored content, and profile-based probability scores from user attributes. Each component is calculated separately and then aggregated to form the final proficiency assessment, reducing the impact of noise in any single data source

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent merges multiple probability score sources (language classifier, text-based analysis of user content, and profile-based inference) into a unified language proficiency assessment. By combining these independent assessments, the system achieves more reliable and accurate results than any single method could provide alone

Inventive Principle:

Principle #5Merging (Combining)

2Measurement precision

If multiple data sources are aggregated to improve language proficiency inference, then the accuracy of proficiency determination improves, but the system complexity increases

Engineering Contradiction:

Improveproficiency determination accuracyVSAvoidinference system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements a universal probability scoring framework that handles multiple data sources (language classifiers, text analysis, profile data) through a common aggregation mechanism. This multi-functional approach allows the same core algorithm to process different types of input data, reducing overall system complexity despite handling diverse data sources

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent transforms various input data types (classifier outputs, text content, profile attributes) into a unified parameter format (probability scores). By changing all inputs to the same parameter type, the aggregation process becomes simpler and more manageable, even though multiple data sources are being processed

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If user profile data and text engagement are used to calculate probability scores, then the language proficiency inference becomes more accurate, but the data processing requirements and computational resources increase

Engineering Contradiction:

Improveproficiency inference accuracyVSAvoidcomputational resource consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements partial action by selectively processing only the most relevant data sources based on data availability and quality. The system calculates probability scores from available inputs (profile data, text engagement) without requiring all possible data types, reducing computational overhead while maintaining inference accuracy where sufficient data exists

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11403463B2Language proficiency inference system

Publication Date: 2022.08.02 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Disclosed are systems, methods, and non-transitory computer-readable media for a language proficiency inference system used to determine a user's proficiency in one or more languages. The language proficiency inference system determines both text-based probability scores and profile-based probability scores indicating a probability that a user speaks a language or set of languages. The text-based probability score is based on text associated with the first user, whereas the profile-based probability score is based profile data of the user. The language proficiency inference system determines aggregated probability scores based on the corresponding text-based and profile-based probability scores. For example, the aggregated probability score is the sum of the text and profile-based probability scores. The language proficiency inference system uses the aggregated scores to determine the languages in which the user is proficient.