The major cause of death globally is heart disease and it is increasing day by day. It is difficult to diagnose the heart disease at early stages until a cardiac issue occurs. Huge amount of heart disease data is available in the health care sector such as in clinics, hospitals etc. However, this data is not intelligently handled to extract useful knowledge. Machine learning techniques help in turning this medical data into useful knowledge. Machine learning is used to design such decision support systems (DSS) that can learn and improve from their past experiences. Recently, deep learning has gained the interest of industry and academics. This research paper focuses on diagnosis of heart disease with high accuracy. In the proposed approach, Keras based deep learning model is used for the computation of results using dense neural network. The proposed model is tested with different combination of hidden layers in dense neural network, started from 3layers to 9layers. Each hidden layer uses 100 neurons and activation function as Relu. Multiple benchmark heart disease datasets have been used for the analysis. Individual as well as Ensemble models are evaluated on all heart disease datasets. Moreover, dense neural network is also evaluated on all datasets using accuracy, sensitivity, specificity and f-measure. Different combination of layers performed well on different datasets due to different categories of attributes. This extensive experimentation is performed to show the effectiveness of proposed model. The analysis of results shows that proposed model based on deep learning performed significantly better in terms of accuracy, sensitivity and specificity as compared to individual and other ensemble techniques on all heart disease datasets.

Feature selection is the process of identifying the most relevant features from the given data having a large feature space. Microarray datasets are comprised of high‐quality features and very few samples of data. Feature selection is performed on such datasets to identify the optimal feature subset. The major goal of feature selection is to improve the accuracy by identifying a minimal feature subset. For this purpose, the proposed research focused on analyzing and identifying effective feature selection algorithms. A novel framework is proposed which utilizes different feature selection methods from filters, wrappers, and embedded algorithms. Furthermore, classification is then performed on selected features to classify the data using a support vector machine (SVM) classifier. Two publically available benchmark datasets are used, i.e., the Microarray dataset and the Cleveland Heart Disease dataset, for experimentation and analysis, and they are archived from the UCI data repository. The performance of SVM is analyzed using accuracy, sensitivity, specificity, and f‐measure. The accuracy of 94.45% and 91% is achieved on each dataset, respectively.

A massive amount of medical data is available in healthcare industry, which can be utilized to extract useful knowledge. A Clinical Decision Support System (CDSS) is used to improve patient’s safety by minimizing medical errors. Heart disease is one of the major chronic maladies even in todays’ world. Many researchers have employed different data mining techniques to predict heart disease. The objective of proposed framework is to improve the accuracy of heart disease prediction. In this paper, an ensemble based voting scheme is proposed to efficiently predict heart disease. Four benchmark heart disease datasets from UCI repository have been utilized for experimentation and evaluation. The performance of the proposed ensemble is compared with individual classifiers as well as with five different ensemble schemes using various parameters in order to show the effectiveness of the proposed ensemble scheme. The evaluation of results shows that the proposed ensemble scheme has better average accuracy (83%) as compared to other ensemble schemes as well as individual classifiers.

Mobile communication has become a dominant medium of communication over the past two decades. New technologies and competitors are emerging rapidly and churn prediction has become a great concern for telecom companies. A customer churn prediction model can provide the accurate identification of potential churners so that a retention solution may be provided to them. The proposed churn prediction model is a hybrid model that is based on a combination of clustering and classification algorithms using an ensemble. First, different clustering algorithms (i.e. K-means, K-medoids, X-means and random clustering) were evaluated individually on two churn prediction datasets. Then hybrid models were introduced by combining the clusters with seven different classification algorithms individually and then evaluations were performed using ensembles. The proposed research was evaluated on two different benchmark telecom data sets obtained from GitHub and Bigml platforms. The analysis of results indicated that the proposed model attained the highest prediction accuracy of 94.7% on the GitHub dataset and 92.43% on the Bigml dataset. State of the art comparison was also performed using the proposed model. The proposed model performed significantly better than state of the art churn prediction models.

Online user generated reviews are now a vital source of product evaluation to both consumer and retailer. There is a need of knowing the factors, generally affecting the helpfulness of reviews and how to identify them. Various studies and researches have been conducted for finding the helpfulness value of online reviews in past recent years. In this paper we have summarized and then analyzed the past work of review helpfulness prediction in a systematic way. The paper provides brief of methodology of each study and how it is contributing towards this domain. It also emphasizes on the pros of the methods used in past and how they are lacking in determining few other aspects of the review helpfulness. The survey discovers that the most popular techniques used for helpfulness prediction are supervised ones and most frequently used are Regression Models and SVM.