ABSTRACT:
Due to the complexity of data in computational processes, interpretation of pattern or extraction of information becomes difficult. Therefore, machine learning (ML) becomes useful in teaching machines how to handle data more efficiently. The learning algorithm provides predictive solution to environmental challenges. Descriptive methodology is introduced using a programmatic approach for ML illustrations. Naïve Bayes algorithm is employed for feature extraction and predictions. At the implementation, two secondary datasets namely knowledge discovery data (KDD) and worldwide web (WWW) are downloaded from GitHub and imported into a python environment. Five different stages of Bayes algorithm are transformed using python scripts. When the machine is trained with the cleaned data and tested data for keyphrase extraction and classification, the ML results indicate 94% and 91% accuracy in prediction for KDD and WWW respectively.
Keywords:
Machine Learning, Algorithm, Dataset, Naïve Bayes, Supervised Learning. electronic payment.
