One of the prominent uses of Predictive Analytics is Health care for more accurate predictions based on proper analysis of cumulative datasets. Often times the datasets are quite imbalanced and sampling techniques like Synthetic Minority Oversampling Technique (SMOTE) give only moderate accuracy in such cases. To overcome this problem, a two-step approach has been proposed. In the first step, SMOTE is modified to reduce the class imbalance in terms of Distance-based SMOTE (D-SMOTE) and Bi-phasic SMOTE (BP-SMOTE) which were then coupled with selective classifiers for prediction. An increase in accuracy is noted for both BP-SMOTE and D-SMOTE compared to basic SMOTE. In the second step, Machine learning, Deep Learning and Ensemble algorithms were used to develop a Stacking Ensemble Framework which showed a significant increase in accuracy for Stacking compared to individual machine learning algorithms like Decision Tree, Naïve Bayes, Neural Networks and Ensemble techniques like Voting, Bagging and Boosting. Two different methods have been developed by combing Deep learning with Stacking approach namely Stacked CNN and Stacked RNN which yielded significantly higher accuracy of 96–97% compared to individual algorithms. Framingham dataset is used for data sampling, Wisconsin Hospital data of Breast Cancer study is used for Stacked CNN and Novel Coronavirus 2019 dataset relating to forecasting COVID-19 cases, is used for Stacked RNN.

Extraction of positive or negative opinions from any online content has received more consideration from researchers during the past decade, since the number of internet users that actively use online review sites, social networks and personal blogs to express their opinions has been growing. Sentiment analysis is the study of automated techniques for extracting sentiments from written languages making use of natural language processing tasks to thoroughly pre-process the data and extract polarity from the data. Customers who want to purchase products or services as well as business organizations, often rely on online reviews for knowing the overall user sentiment. Based on the sentiment, customers can choose whether or not to purchase a product while the organizations get an overall picture of their product. Aspect-based sentiment analysis helps in extraction of important features called aspects because knowing the polarity only is not sufficient. The proposed aspect-based sentiment analysis model uses polarity classification and sentiment extraction on reviews, and extracts the most interesting polarity aspects preferred by the customers automatically using both machine learning and deep learning algorithms. A search engine to pull out tweets and reviews relevant to user specified keyword is developed and corresponding interesting aspects are displayed.

Social media content on the internet is increasing day by day. Since media knowledge helps people in making decisions, web based businesses give their clients an opportunity to express their opinions about items available on the web in the form of surveys and reviews. Sentiment analysis can be used on product reviews or tweets, comments, blogs to infer individual’s feelings or attitudes. Here Aspect Based Sentiment Analysis is used to extract most interesting aspect of a particular product from unlabeled text. We have developed two models for aspect/feature extraction.Model1 uses POS tagging whereas Model2 utilizes TFIDF .In Model 1 we start with noun phrase algorithm and extend it to adjectives and adverbs to extract all the aspect terms. In model2 after data preprocessing TDIDF technique is used. The relative importances of the aspects are calculated and the most important positive, negative and neutral aspects are presented to the user. Naïve Bayes, Support Vector machine, Decision Tree, KNN were used to classify the sentiment polarity of the generated aspects

No abstract

No abstract