Foetal Health Classification Using Input Dataset and Fine Tuning it using K-nearest Neighbour, Naïve Bayes and Decision Tree Classifier

• Published in: 2023 First International Conference on Advances in Electrical, Electronics and Computational Intelligence (ICAEECI) pp. 1 - 5
• Main Authors: Gill, Kanwarpartap Singh, Anand, Vatsala, Gupta, Rupesh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 19.10.2023
• Subjects: Classification; CNN; Confusion matrix; Decision support systems; Decision Tree classifier; Decision trees; Deep Learning; Dimensionality reduction; Feature extraction; Foetal Health Classification; K-Nearest Neighbours; Medical services; Model Training; Naive Bayes classifier; Naive Bayes methods; Risk management
• DOI: 10.1109/ICAEECI58247.2023.10370853

Classifying foetal health is a crucial responsibility in healthcare, and innovative machine learning techniques can be utilised to help. In this instance, K-nearest neighbour, Nave Bayes, and Decision Tree classifiers are used to fine-tune an input picture dataset. The adjustment is done on these classifiers using methods like feature selection, cross-validation, and hyperparameter tweaking. Through the identification of the ideal parameters and feature combinations for your dataset, these strategies assist in improving the performance of the classifiers. To enhance the sustainable effectiveness of the classifier, you may also think about pre-processing methods like normalisation or dimensionality reduction. For the categorization of foetal health in this solution, three models are used and compared, with K-Nearest Neighbours, Naive Bayes classifier, and Decision Tree classifier showing 90%, 84.5%, and 93% accuracy, respectively when machine learning optimization techniques are applied using pre-trained model.