A REVIEW DEVELOPMENT OF RECOMMENDATION SYSTEM USING MACHINE LEARNING

Abstract

E-commerce and entertainment to social media and online education, automated suggestions help to improve customer experience throughout the system. Although the earlier recommendation methods such as collaborative filtering or content-based filtering have set the precedent, these methods have been overtaken by more effective ones. The methods have been applied with deep learning, machine learning, reinforcement learning, and graph-based algorithms, which improves the prediction accuracy, reduces bias, and enhances the satisfaction of the users. The addition of contextual information, user modelling, and feedback loop considerably improves the relevance and accuracy of the recommendations is analysed. The recommendation systems try to fulfil the users with information, products, or content that resonates with the personal interests. Improvements within the real time recommendations system explainability and fairness also tries to address the ever-present bias and opacity within the systems. The additional integration of text, images, and user behavior data also helps in addressing the improvement of the system biases within the algorithms. The more intelligent and user-cantered systems that this recommendation algorithms provide will almost predict the future suggesting the systems with intelligent reasoning will be the most helpful for people. The recommendation system finds itself in a system of prediction where complexity and user reasoning are the determinants of its performance. The information system tries to predict user behaviour, and complexity of the information is based on the behaviour prediction. The presence of these recommendation heuristics, will try to predict systems that are emerging based on complexity and user content. The more the recommendation systems are used, the more intelligent they will be. The more intelligent, the more they will save the user time and effort. The aim is to save the user excessive content. Predict systems that will keep users of recommendation systems engaged highly. Recommender System helps in increase product sales, improve user experience, improve business decision making etc. Machine learning algorithms such as K-Nearest Neighbours (KNN), Convolutional Neural Networks (CNN), SVM (Support Vector Machine) analyses interactions like user data, product features to detect patterns and make suggestions for product accordingly also improve the accuracy of recommendations.