A Novel Method to Predict Sales Price of Domestic Vehicles using News Sentiment Analysis with Random Forest Algorithm

A highly scalable computer environment nowadays makes it possible to perform a variety of tasks involving data-intensive machine learning and natural language processing. One of these is the sales price prediction of home autos with recent concerns that many data scientists have looked at. In this research, the in-memory computing platform Apache Spark-which implements Naive Bayes, Novel Random Forest, Decision Tree, Support Vector Machines, and Logistic Regression are some of the classifiers that the authors examine. This study compares the classification accuracy of several classifiers based on the size of training data sets and the number of n-grams. Tests analyzed quick Amazonl product reviews. Techniques and resources: With 102 samples, the Random Forest Classifier was used on a dataset of 2943 stock sentiment scores. New Random Forest classifiers have been presented and developed as an alternative to Naive Bayes classifiers as a framework for stock market prediction. The classifiers' accuracy was assessed and noted. The Findings and Discussion: The Naive Bayes classifier produces 87% in predicting the future stock share prices on the data set used, whereas the Random forest classifier predicts the same at the rate of 92%. The Random Forest and the Naive Bayes have statistically significant differences from one other (p<0.003). The classification accuracy of the suggested model may be analyzed from the computational analysis results, and it appears that Novel Random Forest is more accurate than Naive Bayes.