ML-Based Implementation Plan Generation for IT Service Requests
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for evaluating client requests in IT service delivery are complex, subjective, and prone to delays and cost overruns due to reliance on human experts, often failing to accurately address client requirements and resulting in sub-optimal solutions.
Innovation Solution
A machine learning-based method that uses a curated dataset to train classifiers and generate feature graphs, enabling the extraction of requirements and identification of relevant use cases from client documents, which are then used to create optimized implementation plans and resource profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human experts are used to evaluate client requests and create implementation plans, then the understanding of client requirements can be deep and nuanced, but the process becomes complex, subjective, costly, and time-consuming
Solution Approach 1:
The patent replaces the mechanical system of human expert evaluation with an automated machine learning system that uses trained classifiers and feature graph models to objectively analyze client requests, extract requirements, and generate implementation plans, thereby eliminating subjectivity and reducing process complexity
Solution Approach 2:
The system enables self-service by allowing the machine learning model to autonomously evaluate client requests, extract requirements, and generate implementation plans without requiring human expert intervention, thus reducing costs and delays while maintaining consistent evaluation quality
2Reliability
If human experts manually create architectural artifacts and implementation plans, then detailed understanding can be achieved, but costs increase and delays occur
Solution Approach 1:
The system performs preliminary action by pre-training classifiers and feature graph models on curated datasets of client requests and use cases before actual evaluation, enabling rapid and accurate analysis of new requests without requiring time-consuming manual expert review
Solution Approach 2:
The patent replaces the time-consuming mechanical process of manual expert analysis with automated machine learning inference that can rapidly evaluate client requests, extract requirements, and generate implementation plans in minutes rather than days
3Adaptability or versatility
If extensive effort by subject matter experts is deployed to analyze client requests, then comprehensive solutions can be developed, but added costs and delays result
Solution Approach 1:
The system achieves universality by creating a multi-functional machine learning platform that can handle various types of client requests across different domains through a single unified architecture, using the same classifiers and feature graph models to extract requirements and generate implementation plans for diverse use cases
Solution Approach 2:
The system applies parameter changes by transforming the unstructured text of client requests into structured feature vectors and graph representations, enabling the machine learning model to efficiently process and analyze comprehensive requirements while maintaining high speed delivery
4Measurement precision
If human experts are used to derive cost cases and pricing for client requests, then accurate resource estimation can be achieved, but the process becomes subjective and costly
Solution Approach 1:
The patent replaces the subjective mechanical process of human expert cost estimation with an objective machine learning system that uses trained classifiers and feature graph models to automatically derive resource requirements and cost cases from client requests, eliminating human bias and subjectivity
Solution Approach 2:
The system incorporates feedback by using curated datasets of historical client requests and their corresponding resource estimations and cost cases to train the machine learning models, allowing the system to learn from past experiences and continuously improve the accuracy of resource estimation while maintaining objectivity
Data Source
AI summary
Techniques for text evaluation are provided. A curated dataset comprising a plurality of textual documents is received. A tree of classifiers is trained, based on the curated dataset, to identify use cases. A feature graph model is generated, based on the curated dataset, to determine textual similarity. A new document is received, and a plurality of requirements is extracted from the new document. For each requirement, one or more vector scores are generated by evaluating the requirement using the tree of classifiers, one or more feature scores are generated by evaluating the requirement using the feature graph model, and one or more use cases are identified for the new textual document based on the one or more vector scores and the one or more feature scores. An implementation is generated for the new document based on the one or more use cases.