Dr. N. Deepa
@pondiuni.edu.in
Assistant Professor, Department of Computer Science
Pondicherry University, Karaikal Campus, Karaikal, Puducherry - 609 605
is presently working as Assistant Professor in the Department of Computer Science, School of Engineering and Technology, Pondicherry University, Karaikal Campus, Karaikal, Puducherry Union Territory, India. She completed her Ph.D in the area of Predictive Analytics in September 2018. She is having 18 years of teaching experience. She has consistently published more than 30 research articles in Scopus and SCI indexed journals with high impact factor. She is having more than 600 citations, h-index of 14 and i10-index of 20. She has published three patents in the year 2019. Her research area includes machine learning, artificial intelligence, operation research, predictive analytics and data mining.
RESEARCH INTERESTS
Machine learning, Artificial Intelligence, Predictive analytics, Blockchain, Industry 5.0
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Prabadevi Boopathy, Madhusanka Liyanage, Natarajan Deepa, Mounik Velavali, Shivani Reddy, Praveen Kumar Reddy Maddikunta, Neelu Khare, Thippa Reddy Gadekallu, Won-Joo Hwang, and Quoc-Viet Pham
Elsevier BV
M Baritha Begum, N. Deepa, Mueen Uddin, Rajesh Kaluri, Maha Abdelhaq, and Raed Alsaqour
Elsevier BV
B. Prabadevi, N. Deepa, K. Ganesan, and Gautam Srivastava
Association for Computing Machinery (ACM)
Identification of the best institute for higher education has become one of the most challenging issues in the present education system. It has become more complicated as more institutes exist with extraordinary infrastructural facilities. Therefore, a decision model is required to identify the best institute for higher education based on multiple criteria. This article proposes a Natural Language Processing (NLP) -based decision model for the identification of the best higher education institute using MCDM methods. The existing decision models for the selection of the best higher education institutions consider a limited number of criteria for decision-making. In this proposed model, 17 criteria and 15 institute datasets have been identified for the development of the decision model through extensive research and experts opinion. The NLP-based text analysis approach is applied to extract the relevant information and convert it to a suitable format. As the relative importance of the criteria plays a crucial role in decision-making, CRITIC and Rank centroid methods are applied for the calculation of relative weights of criteria. TOPSIS method is used to generate the ranking grades of alternatives for each criterion. An objective function is defined to calculate the evaluation scores and select the best institute for higher education. It has been observed that the ranks obtained from the developed model match pretty well with the ranks obtained from other MCDM methods and the experts.
N. Deepa, Quoc-Viet Pham, Dinh C. Nguyen, Sweta Bhattacharya, B. Prabadevi, Thippa Reddy Gadekallu, Praveen Kumar Reddy Maddikunta, Fang Fang, and Pubudu N. Pathirana
Elsevier BV
Praveen Kumar Reddy Maddikunta, Quoc-Viet Pham, Prabadevi B, N Deepa, Kapal Dev, Thippa Reddy Gadekallu, Rukhsana Ruby, and Madhusanka Liyanage
Elsevier BV
Thippa Reddy Gadekallu, Quoc-Viet Pham, Dinh C. Nguyen, Praveen Kumar Reddy Maddikunta, N. Deepa, B. Prabadevi, Pubudu N. Pathirana, Jun Zhao, and Won-Joo Hwang
Institute of Electrical and Electronics Engineers (IEEE)
In recent years, blockchain networks have attracted significant attention in many research areas beyond cryptocurrency, one of them being the Edge of Things (EoT) that is enabled by the combination of edge computing and the Internet of Things (IoT). In this context, blockchain networks enabled with unique features, such as decentralization, immutability, and traceability, have the potential to reshape and transform the conventional EoT systems with higher security levels. Particularly, the convergence of blockchain and EoT leads to a new paradigm, called BEoT that has been regarded as a promising enabler for future services and applications. In this article, we present a state-of-the-art review of recent developments in the BEoT technology and discover its great opportunities in many application domains. We start our survey by providing an updated introduction to blockchain and EoT along with their recent advances. Subsequently, we discuss the use of BEoT in a wide range of industrial applications, from smart transportation, smart city, smart healthcare to smart home, and smart grid. Security challenges in the BEoT paradigm are also discussed and analyzed, with some key services, such as access authentication, data privacy preservation, attack detection, and trust management. Finally, some key research challenges and future directions are also highlighted to instigate further research in this promising area.
Suleman Khan, Saqib Hakak, N. Deepa, B. Prabadevi, Kapal Dev, and Silvia Trelova
Frontiers Media SA
Since its emergence in December 2019, there have been numerous posts and news regarding the COVID-19 pandemic in social media, traditional print, and electronic media. These sources have information from both trusted and non-trusted medical sources. Furthermore, the news from these media are spread rapidly. Spreading a piece of deceptive information may lead to anxiety, unwanted exposure to medical remedies, tricks for digital marketing, and may lead to deadly factors. Therefore, a model for detecting fake news from the news pool is essential. In this work, the dataset which is a fusion of news related to COVID-19 that has been sourced from data from several social media and news sources is used for classification. In the first step, preprocessing is performed on the dataset to remove unwanted text, then tokenization is carried out to extract the tokens from the raw text data collected from various sources. Later, feature selection is performed to avoid the computational overhead incurred in processing all the features in the dataset. The linguistic and sentiment features are extracted for further processing. Finally, several state-of-the-art machine learning algorithms are trained to classify the COVID-19-related dataset. These algorithms are then evaluated using various metrics. The results show that the random forest classifier outperforms the other classifiers with an accuracy of 88.50%.
B. Prabadevi, N. Deepa, L.B. Krithika, Ravi Raj Gulati, and R. Sivakumar
Inderscience Publishers
Ali Kashif Bashir, Suleman Khan, B Prabadevi, N Deepa, Waleed S. Alnumay, Thippa Reddy Gadekallu, and Praveen Kumar Reddy Maddikunta
Wiley
N. Deepa, B. Prabadevi, and Gautam Srivastava
World Scientific Pub Co Pte Lt
Decision making remains a prominent issue in all the problem domains. To make better decisions, multiple factors of the given problem need to be considered and evaluated. Multi-criteria decision-making methods have been used popularly for solving decision-making problems characterized by multiple factors. When multiple factors are considered, it is recommended to categorize the factors into the main criteria and sub-criteria. In this paper, GRAP-an integrated ranking algorithm has been developed by combining Grey Relational Analysis, Rank Sum, and Preference Ranking Organization Method Enrichment Evaluation methods (PROMETHEE) to solve decision-making problems. The weights of the sub-criteria are calculated using the Rank Sum method. Grey Relational Analysis method is used to convert the sub-criteria values into main criteria values in the form of evaluation scores of alternatives. The final ranking scores of the alternatives are obtained using the PROMETHEE method. A decision model is developed using the proposed GRAP algorithm and applied to the Job Profile selection case study. The developed decision model showed much better results compared to other MCDM approaches namely the Simple Additive Weight method, TOPSIS, VIKOR, and Complex Proportional Assessment (COPRAS). Further, a sanity check has been carried out by comparing the results of the decision model with experts’ opinions.
N. Deepa, B. Prabadevi, Praveen Kumar Maddikunta, Thippa Reddy Gadekallu, Thar Baker, M. Ajmal Khan, and Usman Tariq
Springer Science and Business Media LLC
Gautam Srivastava, Thippa Reddy G, N. Deepa, B. Prabadevi, and Praveen Kumar Reddy M
ACM
In recent years, there has been a rapid increase in the applications generating sensitive and personal information based on the Internet of Things (IoT). Due to the sensitive nature of the data there is a huge surge in intruders stealing the data from these applications. Hence a strong intrusion detection systems which can detect the intruders is the need of the hour to build a strong defence systems against the intruders. In this work, a Crow-Search based ensemble classifier is used to classify IoT- based UNSW-NB15 dataset. Firstly, the most significant features are selected from the dataset using Crow-Search algorithm, later these features are fed to the ensemble classifier based on Linear Regression, Random Forest and XGBoost algorithms for training. The performance of the proposed model is then evaluated against the state-of-the-art models to check for its effectiveness. The experimental results prove that the proposed model performs better than the other considered models.
Kathiravan Srinivasan, Lalit Garg, Bor-Yann Chen, Abdulellah A. Alaboudi, N. Z. Jhanjhi, Chang-Tang Chang, B. Prabadevi, and N. Deepa
Computers, Materials and Continua (Tech Science Press)
Expert Systems are interactive and reliable computer-based decisionmaking systems that use both facts and heuristics for solving complex decision-making problems. Generally, the cyclic voltammetry (CV) experiments are executed a random number of times (cycles) to get a stable production of power. However, presently there are not many algorithms or models for predicting the power generation stable criteria in microbial fuel cells. For stability analysis of Medicinal herbs’ CV profiles, an expert system driven by the augmented K-means clustering algorithm is proposed. Our approach requires a dataset that contains voltage-current relationships from CV experiments on the related subjects (plants/herbs). This new approach uses feature engineering and augmented K-means clustering techniques to determine the cycle number beyond which the CV curve stabilizes. We obtain an excellent estimate of the required CV cycles for getting a stable Voltage versus Current curve in this approach. Moreover, this expert system would reduce the time needed and the money spent on running additional and superfluous CV experiments cycles. Thus, it would streamline the process of Bacterial Fuel Cells production using the CV of medicinal herbs.
P. M. Durai Raj Vincent, Nivedhitha Mahendran, Jamel Nebhen, N. Deepa, Kathiravan Srinivasan, and Yuh-Chung Hu
Hindawi Limited
Major depressive disorder (MDD) is the most common mental disorder in the present day as all individuals’ lives, irrespective of being employed or unemployed, is going through the depression phase at least once in their lifetime. In simple terms, it is a mood disturbance that can persist for an individual for more than a few weeks to months. In MDD, in most cases, the individuals do not consult a professional, and even if being consulted, the results are not significant as the individuals find it challenging to identify whether they are depressed or not. Depression, most of the time, cooccurs with anxiety and leads to suicide in few cases, among the employees, who are about to handle the pressure at work and home and mostly unnoticing such problems. This is why this work aims to analyze the IT employees who are mostly working with targets. The artificial neural network, which is modeled loosely like the brain, has proved in recent days that it can perform better than most of the classification algorithms. This study has implemented the multilayered neural perceptron and experimented with the backpropagation technique over the data samples collected from IT professionals. This study aims to develop a model that can classify depressed individuals from those who are not depressed effectively with the data collected from them manually and through sensors. The results show that deep-MLP with backpropagation outperforms other machine learning-based models for effective classification.
Praveen Kumar Reddy Maddikunta, Gautam Srivastava, Thippa Reddy Gadekallu, Natarajan Deepa, and Prabadevi Boopathy
Institution of Engineering and Technology (IET)
The internet of things (IoT) is prominently used in the present world. Although it has vast potential in several applications, it has several challenges in the real-world. One of the most important challenges is conservation of battery life in devices used throughout IoT networks. Since many IoT devices are not rechargeable, several steps to conserve the battery life of an IoT network can be taken using the early prediction of battery life. In this study, a machine learning based model implementing a random forest regression algorithm is used to predict the battery life of IoT devices. The proposed model is experimented on ‘Beach Water Quality – Automated Sensors’ data set generated from sensors in an IoT network from the city of Chicago, USA. Several pre-processing techniques like normalisation, transformation and dimensionality reduction are used in this model. The proposed model achieved a 97% predictive accuracy. The results obtained proved that the proposed model performs better than other state-of-art regression algorithms in preserving the battery life of IoT devices.
B. Prabadevi, N. Deepa, L.B. Krithika, and Vani Vinod
IEEE
The most promising of all cancers that are prevailing among and the primary source of women’s deaths worldwide is the cancerous breast cells. Accurate discovery of this type of cancer cells is essential in its early stages, which can be attained via. various data mining and machine learning techniques. Therefore, a comparative analysis among different machine learning techniques such as Random Forest, Support Vector Machine, Naive Bayes, Decision Tree, Neural Networks and Logistic Regression is conducted. It is determined using the WEKA tool. Also, the selected machine learning algorithms are evaluated based on accuracy in prediction results and performance comparison of each classifier with a ROC curve on multiple classifiers is performed.
L.B Krithika, B. Prabadevi, N. Deepa, and Shruthy Bhavanasi
IEEE
E-commerce is the use of the Internet to connect businesses to their suppliers, customers, and other business partners. E-commerce can be implemented with the help of business management software called enterprise resource planning (ERP). It combines the system with key activities such as production, marketing, trading, collaboration with management, resource management, etc. Implementing it correctly and by using the ERP management software, it could make us achieve benefits to the e-commerce in the way to optimize processes required, increase in data accessibility and increased the change in data, and ultimately usage in internal understanding about e-commerce.This influences the path of implementing a business by correctly chosen ERP that similarly matches the companies activities and process. There is a wide range of different ERP systems that are presently available on the market. Except ancient business ERP systems distributed by corporations like SAP, Oracle, and Peoplesoft. These corporations provide various type of open-source variants. Before deciding to adopt an open-source ERP system, a company must consider its pros and cons. Electronic commerce is known as e-commerce, which is an essential traditional business the whole world is dependent on. We have explained in detail, on how a company uses a market strategy in order to attract the customers without its physical presence. This process of implementing e-commerce using various ERP tools is discussed in this paper.
N.Deepa, B.Prabadevi, Krithika L.B, and B.Deepa
IEEE
Managing the source code of the project and other related documents in an organization is a mandatory need, which may ensure clarity in the delivery of the product enhancing the focus of the organization towards its intended product’s quality. In this digital era of computing, we have many software configuration management tools to handle various documents, its revisions, versions and so. In this paper, we analyze the importance of various Version Control Systems (VCS) evolved to assist the software development lifecycle of the project, and compare favourite VCS tools in the market based on their features, measure their performance across chosen attribute. Also, we propose a new tool having some of the best features found in our comparison study as well as a few extra attributes that we believe will raise the quality of this new tool. This tools can combat the issues we face with existing tools in the market.
Preeti Sinha, B. Prabadevi, Sonia Dutta, N Deepa, and Neha Kumari
IEEE
The performance of the multiprocessor system and time-sharing system rely upon CPU scheduling algorithm. Some of the well known efficient CPU scheduling algorithms are Round Robin scheduling (RR), Shortest job first(SJF), Preemptive version of SJF(SRTF), First Come First Serve (FCFS) etc. As of now, the Round Robin scheduling algorithm is considered as the efficient process scheduling algorithm among all the existing CPU scheduling algorithms. However, in RR the shortest one have to wait for a longer time and in SRTF longer process behaves as a suspended process as short tasks keep on executing. This paper proposes a new scheduling process algorithm ESRR (Efficient Shortest Remaining Time Round Robin) that consolidate two of the preemptive version of existing scheduling algorithms namely RR and SRTF. ESRR reduces total waiting time and turn around time in compare to RR and reduces the waiting time of the shorter process and it also provides a longer process to execute faster than SRTF.
N. Deepa, Mohammad Zubair Khan, B. Prabadevi, Durai Raj Vincent P.M., Praveen Kumar Reddy Maddikunta, and Thippa Reddy Gadekallu
Institute of Electrical and Electronics Engineers (IEEE)
Mahalanobis taguchi system (MTS) is a multi-variate statistical method extensively used for feature selection and binary classification problems. The calculation of orthogonal array and signal-to-noise ratio in MTS makes the algorithm complicated when more number of factors are involved in the classification problem. Also the decision is based on the accuracy of normal and abnormal observations of the dataset. In this paper, a multiclass model using Improved Mahalanobis Taguchi System (IMTS) is proposed based on normal observations and Mahalanobis distance for agriculture development. Twenty-six input factors relevant to crop cultivation have been identified and clustered into six main factors for the development of the model. The multiclass model is developed with the consideration of the relative importance of the factors. An objective function is defined for the classification of three crops, namely paddy, sugarcane and groundnut. The classification results are verified against the results obtained from the agriculture experts working in the field. The proposed classifier provides 100% accuracy, recall, precision and 0% error rate when compared with other traditional classifier models.
N. Deepa and B. Prabadevi
Springer International Publishing
Chuan-Yu Chang, Kathiravan Srinivasan, Wei-Chun Wang, Ganapathy Pattukandan Ganapathy, Durai Raj Vincent, and N Deepa
MDPI AG
In recent times, the application of enabling technologies such as digital shearography combined with deep learning approaches in the smart quality assessment of tires, which leads to intelligent tire manufacturing practices with automated defects detection. Digital shearography is a prominent approach that can be employed for identifying the defects in tires, usually not visible to human eyes. In this research, the bubble defects in tire shearography images are detected using a unique ensemble hybrid amalgamation of the convolutional neural networks/ConvNets with high-performance Faster Region-based convolutional neural networks. It can be noticed that the routine of region-proposal generation along with object detection is accomplished using the ConvNets. Primarily, the sliding window based ConvNets are utilized in the proposed model for dividing the input shearography images into regions, in order to identify the bubble defects. Subsequently, this is followed by implementing the Faster Region-based ConvNets for identifying the bubble defects in the tire shearography images and further, it also helps to minimize the false-positive ratio (sometimes referred to as the false alarm ratio). Moreover, it is evident from the experimental results that the proposed hybrid model offers a cent percent detection of bubble defects in the tire shearography images. Also, it can be witnessed that the false-positive ratio gets minimized to 18 percent.
Deepa, Ganesan, Srinivasan, and Chang
MDPI AG
One of the crucial elements in decision-making is the calculation of criteria weights. In this paper, a new Modified Integrated Weighting (MIW) method was proposed to combine the weights obtained using different weight calculation methods into a single set of weights. The weights express the relative significance of the criteria and play an essential role in making correct decisions. The proposed method considered both subjective knowledge of the experts and the objectivity of the problem by combining the subjective and objective weight assignment methods. The proposed weight calculation method was applied to the agriculture dataset for the evaluation of groundnut crop sites. A decision-making model was developed via the proposed MIW method and Complex Proportional Assessment (COPRAS) method to rank the given groundnut crop site dataset. The ranking results of the developed decision model were compared with the ranking results of average yield data and other methods for validation purposes. The developed model exhibited better results for the given dataset and could be used to solve various other decision-making problems, thereby realizing sustainable development.