Machine Learning Process Model Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently processing large amounts of heterogeneous data from various sources, particularly in converting unstructured data into structured formats that can be easily accessed and analyzed for process optimization.
Innovation Solution
The use of machine learning algorithms trained by a computing server to automatically extract relevant data from heterogeneous sources, identify optimization opportunities in process models, and generate improved process models that can replace existing ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual identification and categorization techniques are used to process unstructured data, then data processing accuracy can be maintained, but labor intensity and processing time increase significantly
Solution Approach 1:
The patent replaces manual mechanical data processing with automated machine learning systems. The ML algorithms automatically extract, categorize, and structure unstructured data from documents and files, eliminating the need for manual identification and classification while maintaining high accuracy through trained models.
Solution Approach 2:
The system enables self-service data processing where the machine learning models autonomously perform data extraction, validation, and structuring tasks. The automated system serves itself by continuously learning from data patterns and improving processing accuracy without human intervention, significantly reducing processing time.
2Reliability
If conventional data conversion techniques are used to convert unstructured files to structured data, then some data can be converted, but the conversion cost is too high to capture all potential changes consistently
Solution Approach 1:
The patent changes the fundamental parameters of data conversion by using machine learning models that can dynamically adapt to different document formats and structures. The ML system processes and converts unstructured data at scale with consistent reliability, capturing all potential changes without the prohibitive costs of conventional techniques.
Solution Approach 2:
The machine learning system provides universal data conversion capability that handles multiple document types, formats, and structures through a single automated platform. This multi-functional approach maintains consistent conversion reliability across diverse data sources while dramatically increasing processing throughput compared to conventional specialized techniques.
3Adaptability or versatility
If data from various heterogeneous sources is collected, then comprehensive data coverage is achieved, but data integration and linking difficulty increases
Solution Approach 1:
The patent introduces machine learning algorithms as intermediary components that bridge heterogeneous data sources. The ML system automatically extracts relevant features, standardizes data formats, and creates meaningful links between data from diverse sources, reducing integration complexity while maintaining comprehensive coverage.
Solution Approach 2:
The system segments the complex data integration task into manageable components: data extraction, feature identification, data structuring, and relationship mapping. Each segment is handled by specialized ML algorithms, making the overall integration process more manageable and less complex while achieving comprehensive heterogeneous data coverage.
Data Source
AI summary
A computing server may receive master data, transaction data, and one or more existing process models of a domain. The computing server may aggregate, based on domain knowledge ontology of the domain, the master data and the transaction data to generate a fact table. For example, entries in the fact table may be identified as relevant to the target process model and include attributes and facts that are extracted from master data or transaction data. The computing server may convert the entries in the fact table into vectors. The computing server inputting vectors into one or more machine learning algorithms to generate one or more algorithm outputs. One or more algorithm outputs may correspond to one or more improved process models that are optimized compared to the existing process models. The computing server may provide the improved process model to the domain to replace one of the existing process models.