K. Govinda

Bio: K. Govinda is an academic researcher from VIT University. The author has contributed to research in topics: Cloud computing & Data security. The author has an hindex of 8, co-authored 64 publications receiving 251 citations.

Chronic Heart Disease Prediction Using Data Mining Techniques

Abstract: Cardiovascular disease has become one of the most widespread diseases in the world at present. It is estimated to have caused around 17.9 million deaths in 2017 which constitutes about 15% of all natural deaths. One major type of cardiovascular disease is chronic heart disease. CHD can be detected at the initial stages by measuring the levels of various health parameters like blood pressure, cholesterol level, heart rate and glucose level. Other characteristics of a person like number of cigarettes smoked per day and BMI level also help to diagnose CHD. This paper focuses on utilizing data mining techniques to predict whether a person is suffering from CHD based on data about various symptoms of CHD. This paper proposes two prediction models namely XGBoost algorithm and logistic regression approach to predict the CHD values. This prediction of CHD at its early stage will help in reducing the risk of CHD on a person.

Stock market prediction using data mining techniques

Abstract: Stock market prediction has been an area of interest for investors as well as researchers for many years due to its volatile, complex and regularly changing in nature, making it difficult to make reliable predictions This paper proposes an approach towards prediction of stock market trends using machine learning models like Random Forest model and Support Vector Machine. The Random Forest model is an ensemble learning method that has been an exceedingly successful model for classification and regression. Support vector machine is a machine learning model for classification. However, this model is mostly used for classification. These techniques are used to forecast whether the price of a stock in the future will be higher than its price on a given day, based on historical data while providing an in-depth understanding of the models being used.

Microsoft and Amazon: A comparison of approaches to cloud security

Abstract: Research has shown that data security has always been an important aspect of quality of service for data service providers; but cloud computing poses new and challenging security threats. The most common security concerns for users of cloud storage are data confidentiality, integrity and availability. Microsoft has considered these concerns and responded with the Azure virtual private storage based on Searchable Encryption. Amazon has also responded to these security issues with its Amazon Web Services. In this paper, we investigate and compare in depth the features of Microsoft Azure and Amazon Web Services deemed to provide security with a particular focus on confidentiality, integrity and availability of data.

Sentiment Analysis on Movie Reviews

Abstract: Movie reviews help users decide if the movie is worth their time. A summary of all reviews for a movie can help users make this decision by not wasting their time reading all reviews. Movie-rating websites are often used by critics to post comments and rate movies which help viewers decide if the movie is worth watching. Sentiment analysis can determine the attitude of critics depending on their reviews. Sentiment analysis of a movie review can rate how positive or negative a movie review is and hence the overall rating for a movie. Therefore, the process of understanding if a review is positive or negative can be automated as the machine learns through training and testing the data. This project aims to rate reviews using two classifiers and compare which gives better and more accurate results. Classification is a data mining methodology that assigns classes to a collection of data in order to help in more accurate predictions and analysis. Naive Bayes and decision tree classifications will be used and the results of sentiment analysis compared.

Twitter Sentiment Analysis Based on US Presidential Election 2016

Abstract: Twitter is among the most popular social networking Web sites today [1], with approximately 317 million monthly active users (Quarter 3 2016). Of these, 67 million users are from the USA. Twitter being a micro-blogging platform is widely used by people to express their opinions. Approximately, 500 million tweets are posted in a day, which is around 6000 tweets per second. Assuming, even one-tenth of these tweets reflect an emotion that results in a lot of people-generated data, which can prove to be a treasure trove of information if studied carefully. We intend to perform sentimental analysis on Twitter data of the US Presidential Election 2016 and then overlay our findings with respect to the two main candidates: Hillary Clinton and Donald Trump with the actual election result, to be able to categorically state whether Twitter can be used as a proper indication of any election.

Beyond the Usefulness and Ease of Use : Extending the TAM for a World - Wide - Web Context

Abstract: Ease of use and usefulness are believed to be fundamental in determining the acceptance and use of various, corporate ITs. These beliefs, however, may not explain the user’s behavior toward newly emerging ITs, such as the World-Wide-Web (WWW). In this study, we introduce playfulness as a new factor that reflects the user’s intrinsic belief in WWW acceptance. Using it as an intrinsic motivation factor, we extend and empirically validate the Technology Acceptance Model (TAM) for the WWW context. # 2001 Elsevier Science B.V. All rights reserved.

Multi-Agent Systems: A Survey

Abstract: Multi-agent systems (MASs) have received tremendous attention from scholars in different disciplines, including computer science and civil engineering, as a means to solve complex problems by subdividing them into smaller tasks. The individual tasks are allocated to autonomous entities, known as agents. Each agent decides on a proper action to solve the task using multiple inputs, e.g., history of actions, interactions with its neighboring agents, and its goal. The MAS has found multiple applications, including modeling complex systems, smart grids, and computer networks. Despite their wide applicability, there are still a number of challenges faced by MAS, including coordination between agents, security, and task allocation. This survey provides a comprehensive discussion of all aspects of MAS, starting from definitions, features, applications, challenges, and communications to evaluation. A classification on MAS applications and challenges is provided along with references for further studies. We expect this paper to serve as an insightful and comprehensive resource on the MAS for researchers and practitioners in the area.

Fractional Neuro-Sequential ARFIMA-LSTM for Financial Market Forecasting

Abstract: Forecasting of fast fluctuated and high-frequency financial data is always a challenging problem in the field of economics and modelling. In this study, a novel hybrid model with the strength of fractional order derivative is presented with their dynamical features of deep learning, long-short term memory (LSTM) networks, to predict the abrupt stochastic variation of the financial market. Stock market prices are dynamic, highly sensitive, nonlinear and chaotic. There are different techniques for forecast prices in the time-variant domain and due to variability and uncertain behavior in stock prices, traditional methods, such as data mining, statistical approaches, and non-deep neural networks models are not suited for prediction and generalized forecasting stock prices. While autoregressive fractional integrated moving average (ARFIMA) model provides a flexible tool for classes of long-memory models. The advancement of machine learning-based deep non-linear modelling confirms that the hybrid model efficiently extracts profound features and model non-linear functions. LSTM networks are a special kind of recurrent neural network (RNN) that map sequences of input observations to output observations with capabilities of long-term dependencies. A novel ARFIMA-LSTM hybrid recurrent network is presented in which ARFIMA model-based filters having the linear tendencies better than ARIMA model in the data and passes the residual to the LSTM model that captures nonlinearity in the residual values with the help of exogenous dependent variables. The model not only minimizes the volatility problem but also overcome the over fitting problem of neural networks. The model is evaluated using PSX company data of the stock market based on RMSE, MSE and MAPE along with a comparison of ARIMA, LSTM model and generalized regression radial basis neural network (GRNN) ensemble method independently. The forecasting performance indicates the effectiveness of the proposed AFRIMA-LSTM hybrid model to improve around 80% accuracy on RMSE as compared to traditional forecasting counterparts.

Predicting Stock Market Trends Using Machine Learning and Deep Learning Algorithms Via Continuous and Binary Data; a Comparative Analysis

Abstract: The nature of stock market movement has always been ambiguous for investors because of various influential factors. This study aims to significantly reduce the risk of trend prediction with machine learning and deep learning algorithms. Four stock market groups, namely diversified financials, petroleum, non-metallic minerals and basic metals from Tehran stock exchange, are chosen for experimental evaluations. This study compares nine machine learning models (Decision Tree, Random Forest, Adaptive Boosting (Adaboost), eXtreme Gradient Boosting (XGBoost), Support Vector Classifier (SVC), Naive Bayes, K-Nearest Neighbors (KNN), Logistic Regression and Artificial Neural Network (ANN)) and two powerful deep learning methods (Recurrent Neural Network (RNN) and Long short-term memory (LSTM). Ten technical indicators from ten years of historical data are our input values, and two ways are supposed for employing them. Firstly, calculating the indicators by stock trading values as continuous data, and secondly converting indicators to binary data before using. Each prediction model is evaluated by three metrics based on the input ways. The evaluation results indicate that for the continuous data, RNN and LSTM outperform other prediction models with a considerable difference. Also, results show that in the binary data evaluation, those deep learning methods are the best; however, the difference becomes less because of the noticeable improvement of models' performance in the second way.

Deep Learning for Stock Market Prediction

Abstract: The prediction of stock groups values has always been attractive and challenging for shareholders due to its inherent dynamics, non-linearity, and complex nature. This paper concentrates on the future prediction of stock market groups. Four groups named diversified financials, petroleum, non-metallic minerals, and basic metals from Tehran stock exchange were chosen for experimental evaluations. Data were collected for the groups based on 10 years of historical records. The value predictions are created for 1, 2, 5, 10, 15, 20, and 30 days in advance. Various machine learning algorithms were utilized for prediction of future values of stock market groups. We employed decision tree, bagging, random forest, adaptive boosting (Adaboost), gradient boosting, and eXtreme gradient boosting (XGBoost), and artificial neural networks (ANN), recurrent neural network (RNN) and long short-term memory (LSTM). Ten technical indicators were selected as the inputs into each of the prediction models. Finally, the results of the predictions were presented for each technique based on four metrics. Among all algorithms used in this paper, LSTM shows more accurate results with the highest model fitting ability. In addition, for tree-based models, there is often an intense competition between Adaboost, Gradient Boosting, and XGBoost.