Inverse Document Frequency & KNN Machine Learning Approach based Novel Text Semantic Analysis

In the modern era each and every thing is digital. E- commerce takes the place of the physical market. People buy anything not going physically anywhere and they are used the E-commerce platform. When The people buy, initially read the reviews provided by the previous customer regarding the product and decide they buy or not. So, the reviews are most essential part of the customer as well as for the customer. This paper delineates the sentiment analysis (SEAN) of the cell amazon real data values initially, use the NLTK, Porter Stemmer and various processing for pre-process the data and get the clean and desired data. Secondly, use the term frequency inverse document frequency (TFIDF) method for extracting the feature and divide the data using the train and test split and last apply the K-nearest neighbor method to train the machine. The artificial model predicts the positive and negative reviews. The experiment outcome result shows the model accuracy, precision, recall of the novel developed algorithm on the real text set.