A Comprehensive Survey on Naive Bayes Algorithm: Advantages, Limitations and Applications

A well-known and frequently used machine learning method, the Naive Bayes algorithm has attracted a lot of interest in a number of fields. The Naive Bayes algorithm, along with its benefits, drawbacks, and practical applications, will be covering in this survey study. The Naive Bayes algorithm's fundamentals are first studied in the survey, with special attention paid to how heavily it relies on probability theory and the notion of feature independence. The Naive Bayes method is further examined, as well as its applicability to both continuous and discrete data. The study then looks into the benefits of the Naive Bayes method, emphasizing its tolerance to irrelevant characteristics, great computing efficiency, and capacity for handling high- dimensional data. The algorithm's interpretability and ability to work with different application domains are also highlighted. The survey article looks at various restrictions, such as the feature independence assumption and algorithm's sensitivity to missing data and its incapacity to recognize intricate correlations between variables. The survey article offers a variety of real-world situations where the Naive Bayes algorithm has been effectively used in order to give a thorough viewpoint. Text categorization, spam filtering, sentiment analysis, recommender systems, medical diagnosis, and financial fraud detection are some of these uses. This survey article provides academics, practitioners, and enthusiasts with an invaluable resource for comprehending the Naive Bayes algorithm's capabilities and concerns by compiling the algorithm's benefits, drawbacks, and applications. It serves as a starting point for further investigation and the creation of enhanced versions and modifications of the Naive Bayes algorithm.