Topic-Based NLP Result Caching Under Memory Constraints

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Information retrieval systems face inefficiencies due to the computationally expensive nature of second-order NLP tasks, which hinder optimization and performance improvement.

Innovation Solution

Selective caching of second-order NLP task results in a database cache, allowing reuse of these results for subsequent queries, thereby reducing the need for repeated processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If second-order NLP tasks are performed for each user query to ensure accurate text analysis, then measurement precision is improved, but productivity deteriorates due to repeated computationally expensive processing

Engineering Contradiction:

Improvetext analysis accuracyVSAvoidquery processing throughput

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system performs NLP task results caching in advance by storing results of second-order NLP tasks in a database cache before they are needed, allowing subsequent queries to retrieve pre-computed results instead of repeating the expensive processing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates cached copies of NLP task results and stores them in a database cache, enabling multiple queries to use these copies instead of requiring original processing for each query

Inventive Principle:

Principle #26Copying

2Productivity

If all NLP task results are cached to maximize query speed, then productivity is improved, but device complexity increases due to memory management requirements

Engineering Contradiction:

Improvequery processing speedVSAvoidcache management complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system implements dynamic cache management where the caching behavior adapts based on query demand patterns, using learned models to determine which topics and NLP tasks should be cached rather than using static caching rules

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system uses query demand data as feedback to continuously improve caching decisions, analyzing patterns in user queries to optimize which NLP results should be retained in cache memory

Inventive Principle:

Principle #23Feedback

3Loss of energy

If comprehensive caching is implemented to reduce processing costs, then loss of energy is reduced, but device complexity increases due to cache selection algorithms

Engineering Contradiction:

Improvecomputational energy consumptionVSAvoidcache selection algorithm complexity

Core Design Contradiction:

Loss of energyVSDevice complexity

Solution Approach 1:

The system replaces complex rule-based cache selection mechanics with machine learning models that automatically learn optimal caching strategies from query patterns, reducing the need for hand-crafted complex selection algorithms

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12505144B2Caching of text analytics based on topic demand and memory constraints

Publication Date: 2025.12.23 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

An embodiment includes analyzing text content of a user query to identify via natural language processing (NLP) a query topic. The embodiment maps the query topic to a topic cluster at a node of a hierarchical model of a text database. The embodiment generates query demand data indicative of demand for the topic cluster based on user queries. The embodiment identifies the topic cluster as a topic-cache candidate based on the query demand data. The embodiment compares an amount of memory required for storing text associated with the first topic cluster to available cache memory. The embodiment caches the text of the topic cluster candidate upon determining that there is sufficient available cache memory space.