SYSTEMS AND METHODS FOR VALIDATING DOMAIN SPECIFIC MODELS

Patent №

US 11,055,200

Granted

2021-07-06

Filed 2019

Owner

TATA CONSULTANCY SERVICES LIMITED

Lab

AI components

4

ml · nlp · kr · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

16599517

Model driven engineering (MDE) approaches necessitate verification and validation (V&V) of the models used. Balancing usability of modeling languages with verifiability of the specification presents several challenges. Conventional modeling languages have automated solvers but are hard to interpret and use. Implementations of present disclosure provide systems and methods for validating domain specific models wherein rules and vocabularies in domain specific model are translated to machine interpretable rules (MIR) and machine interpretable vocabularies (MIV) which are processed (via logic programming technique) to generate a logic programming representation (LPR) of the MIR and MIV based on which solution scenarios are generated for validating the domain specific model. Validation of the model involves verifying LPR using a set of ground facts. During validation of solution scenarios of model, system also checks for inconsistences in rules, if any. These rules are corrected and solution scenarios are re-generated to obtain anomaly free solution scenarios.

AI classification

Natural language1.00
Knowledge representation1.00
Machine learning0.94
AI hardware0.79
Planning0.43
Vision0.00
Speech0.00
Evolutionary computation0.00

Ownership

TATA CONSULTANCY SERVICES LIMITED

assignment · 506880895

Assignors

PRABHU KHOLKAR, DEEPALI, MULPURU, DUSHYANTHI, KULKARNI, VINAY VASANT

On an employer assignment, the assignors are typically the inventors.

From the same owner

© 2026 NYSGPT2525 LLC