ABSTRACT
The herbal drug industry is expanding its horizon being one of the most widely used approaches for therapeutic benefits due to lesser side effects and folklore usage. The major limitations in commercial utilization of herbal drugs are unavailability of scientific methodologies for its identification and characterization due to its dependency on traditional knowledge.
With the advent of artificial-intelligence- and machine-learning-based techniques, scientists across the globe perceive a ray of hope for developing new methodologies for recognition and quality evaluation of raw herbal materials from natural sources.
The present chapter will offer novel ideas and practical answers to this complex challenge by discussing the rationale of using various decision-making modules based on machine learning models with specific attention to herbal drugs for proper understanding, critical review of various possible classifiers and machine learning models like convolutional neural network, artificial neural network, recurrent neural network, random forests etc., along with developing confusion heat matrix for multiclass problems.
It will also cover various steps and key factors to consider while developing and optimizing artificial-intelligence-based decision-making models for reproducible and precise identification of herbal drugs/plants in definitive and systematic manner with acceptable accuracy.
