NLP-Based Software Tagging for Dynamic Resource Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

As the number of computing resources and discovery patterns grows, it becomes impractical to maintain up-to-date discovery patterns due to changes such as updates, configuration changes, and new use cases, especially when textual data does not adhere to standardized formats, making it difficult to identify specific attributes of computing resources.

Innovation Solution

Utilizing natural language processing (NLP) techniques to generate tags for computing resources by processing textual data, including adjusting letter case, removing characters, and determining root forms, which allows for the selection of candidate tags based on frequency weights, enabling efficient mapping and relationship identification between resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional discovery patterns are used to map computing resources, then mapping accuracy can be maintained for standardized resources, but maintenance becomes impractical as the number of resources grows and formats vary

Engineering Contradiction:

Improveadaptability to varying data formatsVSAvoidcomplexity of maintaining discovery patterns

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces manual pattern-matching mechanisms with natural language processing technology. Instead of maintaining complex discovery patterns for each resource type, the system uses NLP to automatically extract and standardize attributes from unstructured textual data, transforming the mechanical pattern-matching process into an intelligent language-based system that adapts to varying formats without requiring pattern updates

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the parameter of data processing from fixed pattern-matching rules to dynamic NLP-based extraction. By transforming unstructured textual data into standardized attributes through language processing, the system adapts to different data formats and resource types without changing the underlying discovery mechanism, thereby maintaining versatility while reducing maintenance complexity

Inventive Principle:

Principle #35Parameter changes

2Reliability

If discovery patterns are updated to accommodate changes in computing resources, then mapping accuracy is improved, but the maintenance burden increases significantly

Engineering Contradiction:

Improveaccuracy of resource mappingVSAvoidtime required for pattern maintenance

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The NLP-based system performs self-updating by automatically adapting to new resource formats and attributes. When new computing resources or formats are introduced, the system independently processes the textual data, extracts relevant attributes, and updates its mapping capabilities without requiring manual pattern maintenance, thereby maintaining high accuracy while eliminating time-consuming updates

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary processing of textual data by pre-extracting and standardizing attributes before formal mapping is required. This advance processing creates a ready-to-use standardized representation that can be quickly mapped to computing resources, ensuring accuracy is maintained without requiring time-consuming pattern updates when resources change

Inventive Principle:

Principle #10Preliminary action

3Productivity

If non-NLP methods are used to process textual data, then processing speed is faster, but relationships between resources with varying formats cannot be identified

Engineering Contradiction:

Improvespeed of text processingVSAvoidloss of relationships in non-standardized data

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The NLP process acts as an intermediary between raw textual data and standardized resource attributes. It translates unstructured text into a standardized format that preserves relationships between resources, serving as a bridge that maintains both processing efficiency and relationship integrity by converting rather than losing information

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11632303B2Enhanced service mapping based on natural language processing

Publication Date: 2023.04.18 SERVICENOW INC

AI summary

A computing system includes persistent storage configured to store representations of software applications that include textual data respectively indicative of attributes of the software applications, and a mapping application configured to perform operations. The operations include retrieving a representation corresponding to a software application and, based thereon, identifying character strings present within textual data associated therewith. The operations also include generating, for each character string, a corresponding weight based on a frequency of the character string within the particular textual data and a frequency of the character string within textual data associated with at least a subset of the software applications. The operations additionally include selecting, from the character strings and based on the corresponding weights, candidate tags for the particular software application. The operations further include generating a mapping between the software application and a computing resource based on the candidate tags, and storing a representation of the mapping.