Performance Analysis and Classification using Naive bayes and Logistic Regression on Big Data

In the era of of fast growth of information technology create massive volume of datasets from social media, mobile the devices, internet of things and other sources. These generate large amounts of unstructured data; this data is known as Big Data. Big data handling is the effective control of the storage and processing of a very high volume of data. Data in organized and unstructured formats necessitate a distinct strategy to overall processing. In the field of data processing and analysis, a dataset may be a vast number of attributes that limit data usefulness and application; consequently, data set dimensions must be minimized. In the classification method, the Naive Bayes Classifier from the UCI repository is used. To explore the influence of feature selection methods, they are applied to various characteristics datasets to produce selected feature vectors, which are then categorized according to each dataset category. The design and implementation of this research are to implement the Naive Bayes Classification and Logistic Regression, we will utilize the well-known Iris Flower and Mushroom Dataset. Experimental results indicate the classification using the pairwise plot, and prediction values between naive bayes and logistic regression.