Analysis and Synthesis of A Human Prakriti Identification System Based on Soft Computing Techniques

Title:Analysis and Synthesis of A Human Prakriti Identification System Based on Soft Computing Techniques

Volume: 13 Issue: 6

Author(s): Vishu Madaan*Anjali Goyal

Affiliation:

• Department of Computer Science Engineering, IKG-Punjab Technical University, Kapurthala,India

Keywords: Ayurveda system, decision making system, knowledge-based system, machine learning, medical expert system, E-health.

Abstract:

Background: The research done on the side effects of modern medicines motivates us to bring Ayurveda back in our modern lifestyle. All allopathic medicines are artificially created and the chemicals used are designed in such a way that they only cure the problem on the surface. This paper will discuss the how can we retain our health for longer time.

Objective: Building a trained and intelligent decision making system that can categorize any health or unhealthy human being into a suitable category of human prakriti dosha.

Methods: Proposed adaptive neuro-fuzzy inference system is trained using hybrid learning technique. Grid Partitioning method is used for membership functions. Total 28 parameters that identify human prakriti are reduced to 7 effective components to get maximum accuracy of results. System is trained with data of 346 healthy individuals to avoid biasing in the result.

Results: The resulting system can answer to any individual about his prakriti dosha, based on its output one can make changes in his lifestyle to avoid the effect of diseases in future. System is obtained with 94.23% accuracy for identifying prakriti dosha.

Conclusion: Building an ANFIS system trained with 346 individuals has shown the improved performance. Consideration of 28 input parameters have actually enhanced the robustness of the system aimed to identify human prakriti dosha.

Cite this article as:

Madaan Vishu*, Goyal Anjali, Analysis and Synthesis of A Human Prakriti Identification System Based on Soft Computing Techniques, Recent Advances in Computer Science and Communications 2020; 13 (6) . https://dx.doi.org/10.2174/2213275912666190207144831