N. Srinivasa Gupta

Bio: N. Srinivasa Gupta is an academic researcher from VIT University. The author has contributed to research in topics: Sentiment analysis & Deep learning. The author has an hindex of 1, co-authored 5 publications receiving 2 citations.

Detection of Depression Related Posts in Tweets Using Classification Methods – A Comparative Analysis

Abstract: A 15-step pre-processing procedure is proposed to improve the accuracy of sentiment mining of depression related posts in the tweets. Perhaps, for the first time, converting emoticons in the depression related tweets into text form is proposed during the pre-processing stage. In this paper, Term Frequency-Inverse Document Frequency with n-grams is used for feature extraction. Sentiment analysis conducted on a dataset consisting of 1.6 million depression related tweets using the proposed pre-processing module with feature extraction using Term Frequency-Inverse Document Frequency with n-grams and Logistic Regression (LR) for classification resulted in 81% of accuracy in detecting depression related tweets.

A Comparative Study on the Performance of Deep Learning Algorithms for Detecting the Sentiments Expressed in Modern Slangs

Abstract: Sentiment analysis is a text investigation technique that distinguishes extremity inside the text, regardless of whether an entire document, sentence, etc. Understanding individuals’ feelings are fundamental for organizations since customers can communicate their considerations and emotions more transparently than any other time in recent memory. In this paper, the proposed model is the sentimental analysis on Twitter slangs, i.e., tweets that contain words that are not orthodox English words but are derived through the evolution of time. To do so, the proposed model will find the root words of the slangs using a snowball stemmer, vectorizing the root words, and then passing it through a neural network for building the model. Also, the tweets would pass through six levels of pre-processing to extract essential features. The tweets are then classified to be positive, neutral, or negative. Sentiment analysis of slangs used in 1,600,000 tweets is proposed using long short-term memory (LSTM) network, logistic regression (LR), and convolution neural network (CNN) algorithms for classification. Among these algorithms, the LSTM network gives the highest accuracy of 78.99%.

Prediction of Type 2 Diabetes Using Hybrid Algorithm

Abstract: Diabetes is a major metabolic disease which affects the entire body system. It becomes a lifelong disease if not handled properly in the early stages of diagnosis. People produce a hormone called insulin. It is a hormone which is used to convert glucose to energy. People suffering from diabetes type 2 produce this hormone, but according to doctors are not able to use it as well as they should. Data Mining tools can be used to mine the famous Indian Pima diabetes for more accurate prediction of diabetes. Many attempts have been made by researchers to improve the efficiency of various models.

CARIMO - A heuristic approach to machine-part cell formation

Abstract: This paper presents a correlation analysis-based heuristic for the machine-part cell formation in the context of cellular manufacturing systems. Two new indices, viz. “mean correlation index” for forming the part families and “relevance index-modified” for identifying the appropriate machine cells are proposed. The machine-part cells formed by the proposed heuristic resulted in a higher grouping efficacy (GE) for 14.3% of the test instances gathered from the literature, and it performed equal to the best in class heuristics available in the literature for 80% of the test instances. The method presented in this paper has set a new benchmark GE for 5 of the 35 test instances used by the researchers in the context of machine-part cell formation without singletons.

A Comparative Study on the Performance of Deep Learning Algorithms for Detecting the Sentiments Expressed in Modern Slangs

Abstract: Sentiment analysis is a text investigation technique that distinguishes extremity inside the text, regardless of whether an entire document, sentence, etc. Understanding individuals’ feelings are fundamental for organizations since customers can communicate their considerations and emotions more transparently than any other time in recent memory. In this paper, the proposed model is the sentimental analysis on Twitter slangs, i.e., tweets that contain words that are not orthodox English words but are derived through the evolution of time. To do so, the proposed model will find the root words of the slangs using a snowball stemmer, vectorizing the root words, and then passing it through a neural network for building the model. Also, the tweets would pass through six levels of pre-processing to extract essential features. The tweets are then classified to be positive, neutral, or negative. Sentiment analysis of slangs used in 1,600,000 tweets is proposed using long short-term memory (LSTM) network, logistic regression (LR), and convolution neural network (CNN) algorithms for classification. Among these algorithms, the LSTM network gives the highest accuracy of 78.99%.

A Classification Algorithm Based on Linear Regression and Linear Programming for Predicting the Breast Cancer

Abstract: In this study, we consider the classification problem in the healthcare domain, and the objective of the problem is to diagnose the breast cancer (binary classification: malignant/benign) based on the number of different features with respect to tumor characteristics. We propose an algorithm to solve this classification problem. The proposed algorithm first eliminates the linear dependency between the input features and then focuses on enriching the feature set based on these independent attributes in order to capture the curvilinear function between the classes. The proposed algorithm, subsequently, uses the multiple linear regression (MLR) model on the training data set to capture the relationship between the response/class variable (malignant/benign) and the set of features (from the enriched feature set). Thereafter, the proposed linear programming (LP) model makes use of the set of relatively influential attributes from the MLR based on the absolute values of the coefficients, to find the classification function/expression for predicting the breast cancer using the training data set. As a part of this process, LP model also finds the initial thresholds where b and b + 1 are thresholds for malignant and benign classes, respectively. In the final phase, the proposed algorithm fine-tunes the thresholds obtained in the LP model through a search process, to determine the exact boundary between the target classes (malignant/benign) using the validation data set. In order to evaluate the performance of the algorithm, we use the Wisconsin diagnostic breast cancer (WDBC) data set from the University of California—Irvine machine learning repository, and we compare the performance with the existing algorithm in the literature. From the results, we observe that the proposed algorithm performs better in terms of the accuracy to predict the breast cancer.

A Novel Neural Network Based Model for Diabetes Prediction Using Multilayer Perceptron and Jrip Classifier

Abstract: AbstractPeople should be conscious on good health and hygiene. Diabetes is another epidemic that threatens not only elderly but even youngsters due to sedentary lifestyles. Managing blood sugar level is vital to avoid further health complications, particularly for Type 2 Diabetes. Early prediction will help people to change their eating habits. Dataset containing Age, Gender, Height, Weight, Insulin Level, Fat Level for various age groups were collected for prediction. The JRip technique was employed to generate rules for the above parameters and the data was tested using Multi-Layer Perceptron (MLP) for improved accuracy. MLP model with 10-fold Jrip cross validation predicted the class labels with 99.8% accuracy.KeywordsDiabetesJRipMulti-layer perceptronPredictionAccuracy

Sustainable Consumption: An Approach to Achieve the Sustainable Environment in India

Abstract: AbstractIn last few years, research on sustainable environment has motivated to unfold the problems through different marketing and consumption patterns. This claims to provide an alternative path to conceptualize the dynamic nature of society to speak about the sustainability. Most of the conceptual–practical research focus on routine problems of people neglecting the need of protection of environment for future generation. The core issues had been unaddressed by behavioural researchers like role of consumers in sustainable development. This research article aims to examine the determinants of consumer behaviour linked with sustainable consumption. The focus would remain on sustainable consumption and how dream of protection of sustainable environment can be achieved through sustainable consumption. The research makes an attempt to find out the determinants and effects of demographic variables on sustainable consumption.KeywordsSustainable environmentSustainable consumptionHuman behaviourSustainable development