Keyword Prediction in Legacy Application Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Legacy applications, such as those with mainframe COBOL program artifacts, pose challenges in business rule mining due to their complexity and the difficulty for business analysts to understand program artifacts, requiring a time-consuming manual process to identify keywords for business logic.
Innovation Solution
A computer-implemented method and system that uses a classification model to predict keywords by tokenizing program artifacts into terms, building feature vectors, and determining confidence levels indicating the likelihood of each term being a keyword, thereby automating the discovery of keywords.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual process is used to identify keywords in legacy applications, then business analysts can understand and validate business logic, but the process is time-consuming and requires significant human effort
Solution Approach 1:
The system enables self-service keyword identification by training a classification model to automatically analyze program artifacts and predict keywords with confidence levels, eliminating the need for manual analysis while maintaining accuracy through the model's learning capability from training data
Solution Approach 2:
The patent replaces the mechanical manual process of keyword identification with an automated machine learning-based classification model that processes program artifacts, extracts features, and predicts keywords algorithmically, substituting human effort with computational automation
2Ease of operation
If program artifacts are analyzed directly by business analysts, then business logic can be understood, but the complexity of legacy application syntax and structure makes analysis difficult
Solution Approach 1:
The classification model serves as an intermediary between program artifacts and business analysts, automatically identifying and predicting keywords that bridge the gap between complex code syntax and understandable business logic, making the artifacts more accessible without requiring analysts to directly interpret complex programming structures
Solution Approach 2:
The system extracts keywords from program artifacts by analyzing features and using the classification model to predict which terms are most relevant, separating the essential business logic indicators from the surrounding complex syntax and structure, thereby simplifying the analysis for business analysts
3Productivity
If automated keyword prediction is implemented, then time and effort are reduced, but the complexity of training and deploying machine learning models increases
Solution Approach 1:
The classification model is trained in advance on relevant data before deployment, performing the complex learning and pattern recognition work beforehand, so that during actual keyword identification, the model can quickly make predictions without requiring complex real-time processing or intervention
Data Source
AI summary
A computer-implemented method includes training a classification model to produce a mapping from a set of feature vectors to a set of confidence levels. Each feature vector describes a respective term, and each confidence level indicates a likelihood that the respective term is a keyword. A plurality of program artifacts are tokenized into a plurality of terms. For each term in the plurality of terms, a respective feature vector is determined for describing the term, based on the plurality of program artifacts. For each term in the plurality of terms, based on the respective feature vector, using the mapping, a respective confidence level is determined for indicating a likelihood that the term is a keyword of the plurality of program artifacts.