Ramadass Suguna
Research Subjects
Ramadass Suguna has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Munisamy Shyamala Devi
Department of Robot and Smart System Engineering, National Satellite Information Research Institute, Kyungpook National University, Daegu, Korea
9 shared publications P. Anandhakumar
Anna University
2 shared publicationsRincy Merlin Mathew
2 shared publications- Rupali Amit Bagate
Army Institute of Technology
2 shared publications Zakir Mujeeb Shaikh
Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology
2 shared publications- Zakir Mujeeb Shaikh
Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology
2 shared publications
Latest Publications
- Open Access
- 1 July 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 35 (1) , 263-273
Unveiling deep learning powers: LSTM, BiLSTM, GRU, BiGRU, RNN comparison
Deep learning algorithms have revolutionized various fields by achieving remarkable results in time series analysis. Among the different architectures, recurrent neural networks (RNNs) have played a significant role in sequential data processing. This study presents a comprehensive comparison of prominent RNN variants: long short-term memory (LSTM), Bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), bidirectional GRU (BiGRU), and RNN, to analyze their respective strengths and weaknesses of national stock exchange India (NSEI). The Python application developed for this research aims to evaluate and determine the most effective algorithm among the variants. To conduct the evaluation, data from the public domain covering the period from 1/1/2004 to 30/06/2023 is collected. The dataset considers significant events such as demonetization, market crashes, the COVID-19 pandemic, downturns in the automobile sector, and rises in unemployment. Stocks from various sectors including banking, automobile, oil and gas, metal, and Pharma are selected for analysis. Finally, the results reveal that algorithm performance varies across different stocks. Specifically, in certain cases, BiLSTM outperforms, while in others, both BiGRU and LSTM are surpassed. Notably, the overall performance of simple RNN is consistently the lowest across all stocks.
- 1 July 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 35 (1) , 263-273
Open AccessUnveiling deep learning powers: LSTM, BiLSTM, GRU, BiGRU, RNN comparison
Deep learning algorithms have revolutionized various fields by achieving remarkable results in time series analysis. Among the different architectures, recurrent neural networks (RNNs) have played a significant role in sequential data processing. This study presents a comprehensive comparison of prominent RNN variants: long short-term memory (LSTM), Bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), bidirectional GRU (BiGRU), and RNN, to analyze their respective strengths and weaknesses of national stock exchange India (NSEI). The Python application developed for this research aims to evaluate and determine the most effective algorithm among the variants. To conduct the evaluation, data from the public domain covering the period from 1/1/2004 to 30/06/2023 is collected. The dataset considers significant events such as demonetization, market crashes, the COVID-19 pandemic, downturns in the automobile sector, and rises in unemployment. Stocks from various sectors including banking, automobile, oil and gas, metal, and Pharma are selected for analysis. Finally, the results reveal that algorithm performance varies across different stocks. Specifically, in certain cases, BiLSTM outperforms, while in others, both BiGRU and LSTM are surpassed. Notably, the overall performance of simple RNN is consistently the lowest across all stocks.
- Open Access
- 1 March 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 33 (3) , 1903-1914
Monte carlo simulation with bilstm for day-ahead stock portfolio management
Predicting stock price movement and optimizing day-ahead stock portfolios are challenging tasks due to the inherent complexity and volatility of financial markets. This study proposes a novel approach that combines bidirectional long short-term memory (BiLSTM) neural networks with monte carlo simulation (MCS) to enhance day-ahead stock portfolio management. In the proposed methodology, historical data of the top-performing 10 stocks from different sectors of the National Stock Exchange of India (NSEI) is obtained from 1 January 2004 to 30 June 2023 and utilized to train a BiLSTM model. This model effectively extracts intricate patterns and trends from the time series, leading to more accurate and robust stock price predictions. MCS generates different scenarios, considering various market conditions and uncertainties. These scenarios provide a comprehensive view of the portfolio’s performance under different conditions, thus mitigating the risk of relying solely on a single prediction. The study evaluates the proposed framework and compares its performance against traditional portfolio management strategies. Results demonstrate that the MCS with the BiLSTM approach outperforms traditional methods in terms of risk-adjusted returns and portfolio stability.
- 1 March 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 33 (3) , 1903-1914
Open AccessMonte carlo simulation with bilstm for day-ahead stock portfolio management
Predicting stock price movement and optimizing day-ahead stock portfolios are challenging tasks due to the inherent complexity and volatility of financial markets. This study proposes a novel approach that combines bidirectional long short-term memory (BiLSTM) neural networks with monte carlo simulation (MCS) to enhance day-ahead stock portfolio management. In the proposed methodology, historical data of the top-performing 10 stocks from different sectors of the National Stock Exchange of India (NSEI) is obtained from 1 January 2004 to 30 June 2023 and utilized to train a BiLSTM model. This model effectively extracts intricate patterns and trends from the time series, leading to more accurate and robust stock price predictions. MCS generates different scenarios, considering various market conditions and uncertainties. These scenarios provide a comprehensive view of the portfolio’s performance under different conditions, thus mitigating the risk of relying solely on a single prediction. The study evaluates the proposed framework and compares its performance against traditional portfolio management strategies. Results demonstrate that the MCS with the BiLSTM approach outperforms traditional methods in terms of risk-adjusted returns and portfolio stability.
- Open Access
- 1 November 2023
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 32 (2) , 968-976
Analyzing interconnections between national stock exchange India and global markets using data analytics
Financial networks have become increasingly complex and interconnected in the global economy. In this context, the use of data analytics has emerged as a valuable tool for understanding and analyzing the relationships between different financial markets. This paper presents a study using Pearson and partial correlation analysis to explore the interconnections between the National Stock Exchange India (NSEI) and global stock markets. The analysis aims to provide insights into the dynamics and patterns or cross-markets interactions and to identify potential opportunities for investment and risk management that to considering the world’s 43 major indices based on imports and exports. The historical data from 1st January 2008 to 30th June 2022 was obtained from the public domain. In this study, a Python framework was developed to investigate Pearson and partial correlation-based networks in financial markets. The results showed that these two types of networks are significantly different from each other. Furthermore, the study discovered that the NSEI is closely connected to the Singapore market and is integrated with other markets. These findings emphasize the complexities to financial market relationships and have important implications for investors and policymakers.
- 1 November 2023
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 32 (2) , 968-976
Open AccessAnalyzing interconnections between national stock exchange India and global markets using data analytics
Financial networks have become increasingly complex and interconnected in the global economy. In this context, the use of data analytics has emerged as a valuable tool for understanding and analyzing the relationships between different financial markets. This paper presents a study using Pearson and partial correlation analysis to explore the interconnections between the National Stock Exchange India (NSEI) and global stock markets. The analysis aims to provide insights into the dynamics and patterns or cross-markets interactions and to identify potential opportunities for investment and risk management that to considering the world’s 43 major indices based on imports and exports. The historical data from 1st January 2008 to 30th June 2022 was obtained from the public domain. In this study, a Python framework was developed to investigate Pearson and partial correlation-based networks in financial markets. The results showed that these two types of networks are significantly different from each other. Furthermore, the study discovered that the NSEI is closely connected to the Singapore market and is integrated with other markets. These findings emphasize the complexities to financial market relationships and have important implications for investors and policymakers.
- Open Access
- 31 October 2022
- Published by Perpetual Innovation Media Pvt. Ltd. in INTERNATIONAL JOURNAL OF NEXT-GENERATION COMPUTING
Sarcasm Detection on Tweets: Ensemble Approach
In the era of web 2.0 social media data is getting generated at huge amount of value to different organizations. Twitter is one of the platforms where different communities express their opinion on social media platform. These different opinion leads to tremendous amount information to work on such as opinion mining, sentiment analysis, sarcasm detection on various social media platform. Sentiment analysis is a natural language processing where author find contextual meaning of text to identify the sentiment of people on digital platform. Sentiment analysis is study of people’s behaviour. Now a day’s people use Sarcasm in many forms to convey their feelings on many social media platforms. Therefore, Sarcasm detection which is sub branch of sentiment analysis becomes a challenging job. Our research is broadly focusing on sarcasm detection of tweets using different machine learning model and neural network model. To enhance the result of classification author have used ensemble model such as staking. Proposed model is inputted with training dataset of tweeter which approximately divided into 70% into training and 30 % into testing dataset. Ensemble model is trained on cleaned dataset and our model is generating an acceptable accuracy which can classify the sarcastic tweets accurately.
- 31 October 2022
- Published by Perpetual Innovation Media Pvt. Ltd. in INTERNATIONAL JOURNAL OF NEXT-GENERATION COMPUTING
Open AccessSarcasm Detection on Tweets: Ensemble Approach
In the era of web 2.0 social media data is getting generated at huge amount of value to different organizations. Twitter is one of the platforms where different communities express their opinion on social media platform. These different opinion leads to tremendous amount information to work on such as opinion mining, sentiment analysis, sarcasm detection on various social media platform. Sentiment analysis is a natural language processing where author find contextual meaning of text to identify the sentiment of people on digital platform. Sentiment analysis is study of people’s behaviour. Now a day’s people use Sarcasm in many forms to convey their feelings on many social media platforms. Therefore, Sarcasm detection which is sub branch of sentiment analysis becomes a challenging job. Our research is broadly focusing on sarcasm detection of tweets using different machine learning model and neural network model. To enhance the result of classification author have used ensemble model such as staking. Proposed model is inputted with training dataset of tweeter which approximately divided into 70% into training and 30 % into testing dataset. Ensemble model is trained on cleaned dataset and our model is generating an acceptable accuracy which can classify the sarcastic tweets accurately.
- Open Access
- 1 August 2021
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 23 (2) , 993-1001
Sarcasm detection of tweets without #sarcasm: data science approach
Identifying sarcasm present in the text could be a challenging work. In sarcasm, a negative word can flip the polarity of a positive sentence. Sentences can be classified as sarcastic or non-sarcastic. It is easier to identify sarcasm using facial expression or tonal weight rather detecting from plain text. Thus, sarcasm detection using natural language processing is major challenge without giving away any specific context or clue such as #sarcasm present in a tweet. Therefore, research tries to solve this classification problem using various optimized models. Proposed model, analyzes whether a given tweet, is sarcastic or not without the presnece of hashtag sarcasm or any kind of specific context present in text. To achieve better results, we used different machine learning classification methodology along with deep learning embedding techniques. Our optimized model uses a stacking technique which combines the result of logistic regression and long short-term memory (LSTM) recurrent neural net feed to light gradient boosting technique which generates better result as compare to existing machine learning and neural network algorithm. The key difference of our research work is sarcasm detection done without #sarcasm which has not been much explored earlier by any researcher. The metrics used for evolutionis F1-score and confusion matrix.
- 1 August 2021
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 23 (2) , 993-1001
Open AccessSarcasm detection of tweets without #sarcasm: data science approach
Identifying sarcasm present in the text could be a challenging work. In sarcasm, a negative word can flip the polarity of a positive sentence. Sentences can be classified as sarcastic or non-sarcastic. It is easier to identify sarcasm using facial expression or tonal weight rather detecting from plain text. Thus, sarcasm detection using natural language processing is major challenge without giving away any specific context or clue such as #sarcasm present in a tweet. Therefore, research tries to solve this classification problem using various optimized models. Proposed model, analyzes whether a given tweet, is sarcastic or not without the presnece of hashtag sarcasm or any kind of specific context present in text. To achieve better results, we used different machine learning classification methodology along with deep learning embedding techniques. Our optimized model uses a stacking technique which combines the result of logistic regression and long short-term memory (LSTM) recurrent neural net feed to light gradient boosting technique which generates better result as compare to existing machine learning and neural network algorithm. The key difference of our research work is sarcasm detection done without #sarcasm which has not been much explored earlier by any researcher. The metrics used for evolutionis F1-score and confusion matrix.
- 1 January 2020
- Published by American Scientific Publishers in Journal of Computational and Theoretical Nanoscience
- Vol. 17 (1) , 519-525
Systematic Image Zooming and Panning of Graphical Images Using Fractional Replication
Scaling is the major operation performed in Transformation of images. The Scaling is an important operation for resizing and reshaping the images that are in digital form. Various operations can be performed with digital images out of which the shrinking and zooming are the most widely operations by any type of users in the world. The other name for shrinking is sub sampling and the zooming operation is also named as Oversampling. The purpose of zooming operation is to extend or enlarge the image in order to have a clear and efficient view. Zooming operations are mostly performed in our mobile for viewing the images in our gallery and this operation is the most frequently performed operation by the mobile users. By considering all these factors, graphical images are the most widely agent to convey any information to the transmitter in order to know the position, size of an object. With this in concern, we try to focus on graphical image zooming as it is very important process in image processing. In this paper, we have used Fractional Replication for Image Zooming that is processed by copying and replicating the pixels from the base image based on the inexact spatial correspondence between the base image and zoom image. And also, the pixel address of the zoom image is calculated fractionally from the pixel address of the base image. We have implemented all the zooming operations like Zoom out in subject of Y-Axis, Zoom out in subject of X-Axis, Zoom in subject of Y Axis, Zoom in subject of X Axis, Left Side Panning, Right Side Panning, Top Zoom Panning and Bottom Zoom Panning for a graphical image using C programming language. Here we attempt to process the various Zooming operation analysis of the graphical image.
- 1 January 2020
- Published by American Scientific Publishers in Journal of Computational and Theoretical Nanoscience
- Vol. 17 (1) , 519-525
Systematic Image Zooming and Panning of Graphical Images Using Fractional Replication
Scaling is the major operation performed in Transformation of images. The Scaling is an important operation for resizing and reshaping the images that are in digital form. Various operations can be performed with digital images out of which the shrinking and zooming are the most widely operations by any type of users in the world. The other name for shrinking is sub sampling and the zooming operation is also named as Oversampling. The purpose of zooming operation is to extend or enlarge the image in order to have a clear and efficient view. Zooming operations are mostly performed in our mobile for viewing the images in our gallery and this operation is the most frequently performed operation by the mobile users. By considering all these factors, graphical images are the most widely agent to convey any information to the transmitter in order to know the position, size of an object. With this in concern, we try to focus on graphical image zooming as it is very important process in image processing. In this paper, we have used Fractional Replication for Image Zooming that is processed by copying and replicating the pixels from the base image based on the inexact spatial correspondence between the base image and zoom image. And also, the pixel address of the zoom image is calculated fractionally from the pixel address of the base image. We have implemented all the zooming operations like Zoom out in subject of Y-Axis, Zoom out in subject of X-Axis, Zoom in subject of Y Axis, Zoom in subject of X Axis, Left Side Panning, Right Side Panning, Top Zoom Panning and Bottom Zoom Panning for a graphical image using C programming language. Here we attempt to process the various Zooming operation analysis of the graphical image.
- 30 July 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (2) , 6198-6203
Customer Segment Prognostic System by Machine Learning using Principal Component and Linear Discriminant Analysis
Recently, manufacturing industry faces lots of problem in predicting the customer behavior and group for matching their outcome with the profit. The organizations are finding difficult in identifying the customer behavior for the purpose of predicting the product design so as to increase the profit. The prediction of customer group is a challenging task for all the organization due to the current growing entrepreneurs. This results in using the machine learning algorithms to cluster the customer group for predicting the demand of the customers. This helps in decision making process of manufacturing the products. This paper attempts to predict the customer group for the wine data set extracted from UCI Machine Learning repository. The wine data set is subjected to dimensionality reduction with principal component analysis and linear discriminant analysis. A Performance analysis is done with various classification algorithms and comparative study is done with the performance metric such as accuracy, precision, recall, and f-score. Experimental results shows that after applying dimensionality reduction, the 2 component LDA reduced wine data set with the kernel SVM, Random Forest classifier is found to be effective with the accuracy of 100% compared to other classifiers.
- 30 July 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (2) , 6198-6203
Customer Segment Prognostic System by Machine Learning using Principal Component and Linear Discriminant Analysis
Recently, manufacturing industry faces lots of problem in predicting the customer behavior and group for matching their outcome with the profit. The organizations are finding difficult in identifying the customer behavior for the purpose of predicting the product design so as to increase the profit. The prediction of customer group is a challenging task for all the organization due to the current growing entrepreneurs. This results in using the machine learning algorithms to cluster the customer group for predicting the demand of the customers. This helps in decision making process of manufacturing the products. This paper attempts to predict the customer group for the wine data set extracted from UCI Machine Learning repository. The wine data set is subjected to dimensionality reduction with principal component analysis and linear discriminant analysis. A Performance analysis is done with various classification algorithms and comparative study is done with the performance metric such as accuracy, precision, recall, and f-score. Experimental results shows that after applying dimensionality reduction, the 2 component LDA reduced wine data set with the kernel SVM, Random Forest classifier is found to be effective with the accuracy of 100% compared to other classifiers.
- 30 July 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (2) , 4800-4807
Feature Snatching and Performance Assessment for Connoting the Admittance Likelihood of student using Principal Component Reduction
Recently, engineers are concentrating on designing an effective prediction model for finding the rate of student admission in order to raise the educational growth of the nation. The method to predict the student admission towards the higher education is a challenging task for any educational organization. There is a high visibility of crisis towards admission in the higher education. The admission rate of the student is the major risk to the educational society in the world. The student admission greatly affects the economic, social, academic, profit and cultural growth of the nation. The student admission rate also depends on the admission procedures and policies of the educational institutions. The chance of student admission also depends on the feedback given by all the stake holders of the educational sectors. The forecasting of the student admission is a major task for any educational institution to protect the profit and wealth of the organization. This paper attempts to analyze the performance of the student admission prediction by using machine learning dimensionality reduction algorithms. The Admission Predict dataset from Kaggle machine learning Repository is used for prediction analysis and the features are reduced by feature reduction methods. The prediction of the chance of Admit is achieved in four ways. Firstly, the correlation between each of the dataset attributes are found and depicted as a histogram. Secondly, the top most high correlated features are identified which are directly contributing to the prediction of chance of admit. Thirdly, the Admission Predict dataset is subjected to dimensionality reduction methods like principal component analysis (PCA), Sparse PCA, Incremental PCA , Kernel PCA and Mini Batch Sparse PCA. Fourth, the optimized dimensionality reduced dataset is then executed to analyze and compare the mean squared error, Mean Absolute Error and R2 Score of each method. The implementation is done by python in Anaconda Spyder Navigator Integrated Development Environment. Experimental Result shows that the CGPA, GRE Score and TOEFL Score are highly correlated features in predicting the chance of admit. The execution of performance analysis shows that Incremental PCA have achieved the effective prediction of chance of admit with minimum MSE of 0.09, MAE of 0.24 and reasonable R2 Score of 0.26.
- 30 July 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (2) , 4800-4807
Feature Snatching and Performance Assessment for Connoting the Admittance Likelihood of student using Principal Component Reduction
Recently, engineers are concentrating on designing an effective prediction model for finding the rate of student admission in order to raise the educational growth of the nation. The method to predict the student admission towards the higher education is a challenging task for any educational organization. There is a high visibility of crisis towards admission in the higher education. The admission rate of the student is the major risk to the educational society in the world. The student admission greatly affects the economic, social, academic, profit and cultural growth of the nation. The student admission rate also depends on the admission procedures and policies of the educational institutions. The chance of student admission also depends on the feedback given by all the stake holders of the educational sectors. The forecasting of the student admission is a major task for any educational institution to protect the profit and wealth of the organization. This paper attempts to analyze the performance of the student admission prediction by using machine learning dimensionality reduction algorithms. The Admission Predict dataset from Kaggle machine learning Repository is used for prediction analysis and the features are reduced by feature reduction methods. The prediction of the chance of Admit is achieved in four ways. Firstly, the correlation between each of the dataset attributes are found and depicted as a histogram. Secondly, the top most high correlated features are identified which are directly contributing to the prediction of chance of admit. Thirdly, the Admission Predict dataset is subjected to dimensionality reduction methods like principal component analysis (PCA), Sparse PCA, Incremental PCA , Kernel PCA and Mini Batch Sparse PCA. Fourth, the optimized dimensionality reduced dataset is then executed to analyze and compare the mean squared error, Mean Absolute Error and R2 Score of each method. The implementation is done by python in Anaconda Spyder Navigator Integrated Development Environment. Experimental Result shows that the CGPA, GRE Score and TOEFL Score are highly correlated features in predicting the chance of admit. The execution of performance analysis shows that Incremental PCA have achieved the effective prediction of chance of admit with minimum MSE of 0.09, MAE of 0.24 and reasonable R2 Score of 0.26.
- 18 May 2019
- Published by Springer Nature
- p. 7-19
Design of IoT Blockchain Based Smart Agriculture for Enlightening Safety and Security
- 18 May 2019
- Published by Springer Nature
- p. 7-19
Design of IoT Blockchain Based Smart Agriculture for Enlightening Safety and Security
- 1 February 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Integration of Blockchain and IoT in Satellite Monitoring Process
The Internet of Things is evolving as a complete matured technology to be used in all the Smart applications and it establishes itself in the future generations of internet. As like Internet of Things, Blockchain is the blooming technology in which each node involved in the blockchain contains the distributed ledger which enhances the security and data transparency. Illegal users are not able to perform any fault transaction in the blockchain network due to its ability of performing smart contract and consensus. The Internet of Things can be merged with the blockchain to improve the performance of the application in real time. However, managing the devices connected to the sensors in IoT environment and mining the block chain remains the technical challenge forever. With this background, an attempt is made to survey the core details of blockchain technology and its features. The design architecture by merging IoT and BlockChain for Satellite Monitoring and ended up with some new architectural framework has been proposed in this paper.
- 1 February 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Integration of Blockchain and IoT in Satellite Monitoring Process
The Internet of Things is evolving as a complete matured technology to be used in all the Smart applications and it establishes itself in the future generations of internet. As like Internet of Things, Blockchain is the blooming technology in which each node involved in the blockchain contains the distributed ledger which enhances the security and data transparency. Illegal users are not able to perform any fault transaction in the blockchain network due to its ability of performing smart contract and consensus. The Internet of Things can be merged with the blockchain to improve the performance of the application in real time. However, managing the devices connected to the sensors in IoT environment and mining the block chain remains the technical challenge forever. With this background, an attempt is made to survey the core details of blockchain technology and its features. The design architecture by merging IoT and BlockChain for Satellite Monitoring and ended up with some new architectural framework has been proposed in this paper.