Natural Semantic Digitization of Enterprise Process Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current enterprise modeling techniques face challenges in transforming spontaneous, unstructured oral conversations into formalized computerized models, particularly due to the media break between physical whiteboards or flipcharts and digital modeling software, which is error-prone and resource-intensive, and lacks effective handling of complex semantic concepts and domain-specific terminology.
Innovation Solution
A user-interactive system that utilizes a combination of general and domain-specific grammars to recognize semantic concepts from audio input, allowing for multi-step processing and user review, thereby transforming unstructured natural language into a standardized modeling language, reducing the need for physical media and specialized technical knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual transference from whiteboard/flipchart to modeling software is used, then model creation flexibility is maintained, but productivity is reduced and error rate increases
Solution Approach 1:
The patent replaces the manual mechanical process of transcribing models from whiteboard to software with an automated optical recognition system. The system captures images of hand-drawn process models, automatically recognizes the graphical elements and their relationships, and generates corresponding digital model files, eliminating the error-prone manual transference step while maintaining the flexibility of informal sketching.
2Loss of time
If mobile computing with modeling software is used during meetings, then the media break is removed, but device complexity and ease of operation deteriorate
Solution Approach 1:
The patent extracts the complex modeling software interface from the meeting process, replacing it with simple image capture and verbal description. Participants only need to sketch models informally on whiteboards or flipcharts and provide verbal descriptions, while the system handles the complex tasks of recognition, interpretation, and digital model generation, thus eliminating media break without burdening participants with technical complexity.
3Measurement precision
If static shape-based recognition is used, then text recognition accuracy is maintained, but semantic concept recognition capability is limited
Solution Approach 1:
The patent combines multiple recognition approaches into a composite system: optical character recognition for text accuracy, graphical element recognition for shape identification, and natural language processing for semantic concept extraction. This composite approach integrates the strengths of different methods, maintaining text recognition precision while adding comprehensive semantic understanding of process models, roles, artifacts, and relationships.
4Manufacturing precision
If domain-specific grammars are implemented, then manufacturing precision of model recognition is improved, but device complexity increases
Solution Approach 1:
The patent segments the grammar system into hierarchical levels: general process modeling grammars for basic structures, domain-specific grammars for industry particulares, and artifact-type grammars for specific model elements. This segmentation allows the system to apply appropriate grammar complexity only where needed, improving recognition precision for domain-specific concepts while managing overall system complexity through modular organization.
Data Source
AI summary
Certain example embodiments relate to techniques for creating computerized models usable with enterprise modeling platforms implementing formalized modeling languages. Audio input of an orally-described model having semantic concepts associable with the formalized language but following a natural language pattern rather than an input pattern expected by the formalized language is received. At least some of the semantic concepts are recognizable from a domain-specific grammar that includes possible semantic concepts that are arranged hierarchically and associated with a domain to which the computerized model being created belongs. At least some others are recognizable from a general grammar that includes other possible semantic concepts that are relevant to the computerized model and that are arranged hierarchically but that are domain-independent. A digitized iteratively-reviewed version of the orally-described model is transformed into the computerized model via rules defining relationships between elements therein, and the formalized language.