This page has only limited features, please log in for full access.
Dr. Deepak Sinwar is an Assistant Professor of Department of Computer and Communication Engineering, School of Computing & Information Technology at Manipal University Jaipur, Jaipur, Rajasthan, India. He received his Ph.D and M.Tech degrees in Computer Science and Engineering in 2016 and 2010 respectively; and B.Tech (with honors) in Information Technology in 2008. His research interests include Computational Intelligence, Data Mining, Machine Learning, Reliability Theory, Computer Networks and Pattern Recognition. He is an enthusiastic and motivating technocrat with more than 10 years of research and academic experience at different institutes of higher learning. He has supervised more than 10 students for their M.Tech dissertation work, one for doctoral and currently supervising 4 Ph.D scholars at Manipal University Jaipur. He has authored many research articles in various International Journals, National/ International conferences of repute. He is also working as a reviewer for several International Journals of repute. He is an organizing chair of International conference on “Innovations in Computational Intelligence and Computer Vision” (ICICV-2020) and ICICV-2021 (both proceedings by SpringerNature Singapore). He has received one patent and six copyrights for his innovative contributions. He is a life member of ISTE, and member of ACM, IEEE,
In the modern era, deep learning techniques have emerged as powerful tools in image recognition. Convolutional Neural Networks, one of the deep learning tools, have attained an impressive outcome in this area. Applications such as identifying objects, faces, bones, handwritten digits, and traffic signs signify the importance of Convolutional Neural Networks in the real world. The effectiveness of Convolutional Neural Networks in image recognition motivates the researchers to extend its applications in the field of agriculture for recognition of plant species, yield management, weed detection, soil, and water management, fruit counting, diseases, and pest detection, evaluating the nutrient status of plants, and much more. The availability of voluminous research works in applying deep learning models in agriculture leads to difficulty in selecting a suitable model according to the type of dataset and experimental environment. In this manuscript, the authors present a survey of the existing literature in applying deep Convolutional Neural Networks to predict plant diseases from leaf images. This manuscript presents an exemplary comparison of the pre-processing techniques, Convolutional Neural Network models, frameworks, and optimization techniques applied to detect and classify plant diseases using leaf images as a data set. This manuscript also presents a survey of the datasets and performance metrics used to evaluate the efficacy of models. The manuscript highlights the advantages and disadvantages of different techniques and models proposed in the existing literature. This survey will ease the task of researchers working in the field of applying deep learning techniques for the identification and classification of plant leaf diseases.
Vijaypal Dhaka; Sangeeta Meena; Geeta Rani; Deepak Sinwar; Kavita Kavita; Muhammad Ijaz; Marcin Woźniak. A Survey of Deep Convolutional Neural Networks Applied for Prediction of Plant Leaf Diseases. Sensors 2021, 21, 4749 .
AMA StyleVijaypal Dhaka, Sangeeta Meena, Geeta Rani, Deepak Sinwar, Kavita Kavita, Muhammad Ijaz, Marcin Woźniak. A Survey of Deep Convolutional Neural Networks Applied for Prediction of Plant Leaf Diseases. Sensors. 2021; 21 (14):4749.
Chicago/Turabian StyleVijaypal Dhaka; Sangeeta Meena; Geeta Rani; Deepak Sinwar; Kavita Kavita; Muhammad Ijaz; Marcin Woźniak. 2021. "A Survey of Deep Convolutional Neural Networks Applied for Prediction of Plant Leaf Diseases." Sensors 21, no. 14: 4749.
Predicting the optimum availability of the physical processing unit of sewage treatment plant is defined as a Nondeterministic Polynomial time-hard problem. Recently many researchers have utilized soft computing techniques to handle this issue. However, the existing techniques are far from the optimal solutions as soft computing techniques suffer from various issues such as, poor computational speed, getting stuck in local optima, pre-mature convergence, etc. Therefore, in this work a novel mathematical model is designed and implemented using Markov process and Chapman-Kolmogorov equations derived by assuming arbitrary repair rates and exponentially distributed failure rates. Thereafter, Genetic Algorithm and Particle Swarm Optimization techniques are utilized to optimize the availability and performance of physical processing unit. The needed data has been collected with the help of plant personnel and results are also shared with them. Experimental results reveal that the Particle Swarm Optimization based proposed model outperforms the competitive techniques.
Deepak Sinwar; Monika Saini; Dilbag Singh; Drishty Goyal; Ashish Kumar. Availability and performance optimization of physical processing unit in sewage treatment plant using genetic algorithm and particle swarm optimization. International Journal of System Assurance Engineering and Management 2021, 1 -12.
AMA StyleDeepak Sinwar, Monika Saini, Dilbag Singh, Drishty Goyal, Ashish Kumar. Availability and performance optimization of physical processing unit in sewage treatment plant using genetic algorithm and particle swarm optimization. International Journal of System Assurance Engineering and Management. 2021; ():1-12.
Chicago/Turabian StyleDeepak Sinwar; Monika Saini; Dilbag Singh; Drishty Goyal; Ashish Kumar. 2021. "Availability and performance optimization of physical processing unit in sewage treatment plant using genetic algorithm and particle swarm optimization." International Journal of System Assurance Engineering and Management , no. : 1-12.
Script recognition has many real-life applications like optical character recognition, document archiving, writer identification, searching within the documents, etc. Automatic script recognition from multilingual documents is a stimulating task, where the system must identify and recognize several types of scripts that can be available on a single page. In offline script recognition, printed or handwritten documents are firstly scanned followed by the process of script recognition, whereas in online script recognition documents are already in soft-copy form. Most of the script recognition techniques presented by researchers so far are based on traditional image processing frameworks. But nowadays, it is observed that Deep Learning-based techniques are more capable of achieving a script recognition task efficiently as well as accurately. This paper provides a comprehensive survey of various techniques available for identification and recognition of multilingual scripts from the last few decades that are mainly focused on Indic scripts. However, some potential non-Indic script identification works are also incorporated for ease of understanding. We hope that this survey can act as a compendium as well as provide future directions to researchers for developing generic OCRs.
Deepak Sinwar; Vijaypal Singh Dhaka; Nitesh Pradhan; Saumya Pandey. Offline script recognition from handwritten and printed multilingual documents: a survey. International Journal on Document Analysis and Recognition (IJDAR) 2021, 24, 97 -121.
AMA StyleDeepak Sinwar, Vijaypal Singh Dhaka, Nitesh Pradhan, Saumya Pandey. Offline script recognition from handwritten and printed multilingual documents: a survey. International Journal on Document Analysis and Recognition (IJDAR). 2021; 24 (1-2):97-121.
Chicago/Turabian StyleDeepak Sinwar; Vijaypal Singh Dhaka; Nitesh Pradhan; Saumya Pandey. 2021. "Offline script recognition from handwritten and printed multilingual documents: a survey." International Journal on Document Analysis and Recognition (IJDAR) 24, no. 1-2: 97-121.
Technology plays a very vital roles in the growth of the economy of any nation. Hence, information communication channel needs to be very strong for timely delivery of information and growth of any country. Mobile technology is the backbone of communication channel in any country who has incorporated it. Since 1980 mobile communication is very popular mode of communication and researches are going on in this area since that time. Starting from the first generation mobile network to fifth generation mobile network, every nation wants to enhance their information communication technology infrastructures in aspect of communication. The 5G mobile technology is subject of debate now a day. Still, most of the countries are in the race for adopting this technology and are ignoring its adverse effects on human health and environments. 5G mobile technology uses millimetre waves and higher frequency band 6 GHz to 100 GHz for communication. Initially, there was appeal made in United Nation Council and later in European Union against the launch of 5G, which was signed by more than three hundred scientists and doctors, stating that the 5G mobile technology is not good for environment. Various research has been conducted regarding the adverse effects of RF-EMF waves, which are generated by cell towers, on human health and environment. 5G uses very dense infrastructure and there is evidence that the RF-EMF radiation level is very strong in fifth generation mobile technology as compare to previous mobile technologies. Hence, the current study is focused on reviewing the impact of 5G mobile technology on flora and fauna Kingdome.
Rajesh Kumar; Rabira Geleta; Amit Pandey; Deepak Sinwar. Adverse Effects of 5th Generation Mobile Technology on Flora and Fauna: Review Study. IOP Conference Series: Materials Science and Engineering 2021, 1099, 012031 .
AMA StyleRajesh Kumar, Rabira Geleta, Amit Pandey, Deepak Sinwar. Adverse Effects of 5th Generation Mobile Technology on Flora and Fauna: Review Study. IOP Conference Series: Materials Science and Engineering. 2021; 1099 (1):012031.
Chicago/Turabian StyleRajesh Kumar; Rabira Geleta; Amit Pandey; Deepak Sinwar. 2021. "Adverse Effects of 5th Generation Mobile Technology on Flora and Fauna: Review Study." IOP Conference Series: Materials Science and Engineering 1099, no. 1: 012031.
The 2019 novel coronavirus (2019-nCoV), with a beginning stage in Wuhan (China), has spread quickly among people living in different nations and has affected more than 6 million lives. Researchers around the globe are trying to device a solution in the form of the vaccine but unfortunately, till date, there is no full-proof vaccine for 2019-nCoV. Most of the countries are adopting prolonged lockdown and social distancing strategies to counter this pandemic. Flying Ad-hoc Network (FANET) on the other hand, can provide several services such as the delivery of essential items, disinfecting common areas, surveillance, traffic monitoring, communication, temperature monitoring, etc. Availing such services using FANETs during Corona Virus Disease (COVID-19) outbreak is like a boon that can minimize the general problems of mankind to some extent during lockdowns. Nowadays, it is foremost required to deploy solutions that can automate the process of availing several services using FANETs and the same is attracting researchers in great demand. This paper has presented a brief survey of several applications of FANETs (especially UAVs and Drones) that can facilitate mankind in coping with the general problems during COVDI-19 outbreak and subsequent lockdowns. We hope that this survey has covered most of the applications of UAVs and can provide new insights into the research community.
Manisha Devi; Sunil Kumar Maakar; Deepak Sinwar; Mahesh Jangid; Poonam Sangwan. Applications of Flying Ad-hoc Network During COVID-19 Pandemic. IOP Conference Series: Materials Science and Engineering 2021, 1099, 012005 .
AMA StyleManisha Devi, Sunil Kumar Maakar, Deepak Sinwar, Mahesh Jangid, Poonam Sangwan. Applications of Flying Ad-hoc Network During COVID-19 Pandemic. IOP Conference Series: Materials Science and Engineering. 2021; 1099 (1):012005.
Chicago/Turabian StyleManisha Devi; Sunil Kumar Maakar; Deepak Sinwar; Mahesh Jangid; Poonam Sangwan. 2021. "Applications of Flying Ad-hoc Network During COVID-19 Pandemic." IOP Conference Series: Materials Science and Engineering 1099, no. 1: 012005.
The main objective of present study to investigate the availability and profit of power generation systems established in sewage treatment plants. The sewage treatment plant is an industry in which waste sewage water has been treated and waste is used to generate power. For this purpose, a mathematical model has been proposed by considering constant failure and repair rates. The power generating unit in sewage treatment plant comprises six subsystems as sludge digester, gas holding tank, gas burner, gas scrubber, gas engine and power generation. By using appropriate redundancy technique in model development Chapman-Kolmogorov differential equations has been drawn with the help of the Markov birth death process. Availability and profit have been analysed based on making variation in the failure and repair rates. The numerical and graphical results have been depicted for a particular case. The derived results have been discussed with system designers and found useful.
Monika Saini; Drishty Goyal; Ashish Kumar; Deepak Sinwar. Investigation of performance measures of power generating unit of sewage treatment plant. Journal of Physics: Conference Series 2021, 1714, 012008 .
AMA StyleMonika Saini, Drishty Goyal, Ashish Kumar, Deepak Sinwar. Investigation of performance measures of power generating unit of sewage treatment plant. Journal of Physics: Conference Series. 2021; 1714 (1):012008.
Chicago/Turabian StyleMonika Saini; Drishty Goyal; Ashish Kumar; Deepak Sinwar. 2021. "Investigation of performance measures of power generating unit of sewage treatment plant." Journal of Physics: Conference Series 1714, no. 1: 012008.
Cloud computing technology provides access to the pool of configurable resources including storage space, application, services and on demand network. Involvement of Cloud with the organization minimizes organization efforts toward fulfilling its customer's needs. One of the major advantages of cloud computing is the Single Sign On (SSO) technique that allows the user to access multiple application services using a single user credential. In cloud computing, there are many issues, and challenges to be discussed. However, prevention from security attacks is much more difficult in preserving privacy of user agents. This paper has proposed SSO-based bio-metric authentication architecture for cloud computing services to overcome the security and privacy attacks. Bio-metric authentication is effective for resources controlled by end devices at the time of accessing the cloud services since these devices are computationally inefficient for user information processing during authentication. Accordingly, security attack in the cloud computing gets minimized using the proposed architecture. The proposed architecture also includes a novel approach in which there exist one to one relationship between user agent and the service provider. In this user agents can use their fingerprint while requesting for registration and accessing different cloud application services at cloud. Based on comparative study with several existing architectures, the highlights of the proposed architecture have been presented.
Biniyam Alemu; Rajesh Kumar; Deepak Sinwar; Ghanshyam Raghuwanshi. Fingerprint Based Authentication Architecture for Accessing Multiple Cloud Computing Services using Single User Credential in IOT Environments. Journal of Physics: Conference Series 2021, 1714, 012016 .
AMA StyleBiniyam Alemu, Rajesh Kumar, Deepak Sinwar, Ghanshyam Raghuwanshi. Fingerprint Based Authentication Architecture for Accessing Multiple Cloud Computing Services using Single User Credential in IOT Environments. Journal of Physics: Conference Series. 2021; 1714 (1):012016.
Chicago/Turabian StyleBiniyam Alemu; Rajesh Kumar; Deepak Sinwar; Ghanshyam Raghuwanshi. 2021. "Fingerprint Based Authentication Architecture for Accessing Multiple Cloud Computing Services using Single User Credential in IOT Environments." Journal of Physics: Conference Series 1714, no. 1: 012016.
Corona VIrus Disease 2019 (COVID-19) caused by 2019 novel Coronavirus (2019-nCoV) becomes the global pandemic that has affected the economy of almost every country in the world. Out of 5.6 million cases as on May 26, 2020 it has already consumed around 0.34 million lives. As compared to first case of 2019-nCoV in Wuhan City, China in December 2019, the very first case in Ethiopia was notified on March 12, 2020 and till May 26, 2020 the number of people infected with this disease becomes 655. There are numerous social and geographical factors that are responsible for spreading of the deadly virus. This paper has presented a study of these social and geographical factors, specifically in Ethiopia. To validate the study, several statistical techniques are used to analyze the effect of these factors in calculating transmission rate in Ethiopia. The study shows that the age, gender, immunity, nutritional deficiencies, and deprived household problems are directly correlated with the spread of COVID-19 in Ethiopia.
Rajesh Kumar; Amit Pandey; Rabira Geleta Ibsa; Deepak Sinwar; Vijaypal Singh Dhaka. Study of social and geographical factors affecting the spread of COVID-19 in Ethiopia. Journal of Statistics and Management Systems 2021, 24, 99 -113.
AMA StyleRajesh Kumar, Amit Pandey, Rabira Geleta Ibsa, Deepak Sinwar, Vijaypal Singh Dhaka. Study of social and geographical factors affecting the spread of COVID-19 in Ethiopia. Journal of Statistics and Management Systems. 2021; 24 (1):99-113.
Chicago/Turabian StyleRajesh Kumar; Amit Pandey; Rabira Geleta Ibsa; Deepak Sinwar; Vijaypal Singh Dhaka. 2021. "Study of social and geographical factors affecting the spread of COVID-19 in Ethiopia." Journal of Statistics and Management Systems 24, no. 1: 99-113.
In the development of the advanced world, information has been created each second in numerous regions like astronomy, social locales, medical fields, transportation, web-based business, logical research, horticulture, video, and sound download. As per an overview, in 60 seconds, 600+ new clients on YouTube and 7 billion queries are executed on Google. In this way, we can say that the immense measure of organized, unstructured, and semi-organized information are produced each second around the cyber world, which should be managed efficiently. Big data conveys properties such as unpredictability, 'V' factor, multivariable information, and it must be put away, recovered, and dispersed. Logical arranged data may work as information in the field of digital world. In the past century, the sources of data as to size were very limited and could be managed using pen and paper. The next generation of data generation tools include Microsoft Excel, Access, and database tools like SQL, MySQL, and DB2.
Vijander Singh; Amit Kumar Bairwa; Deepak Sinwar. An Analysis of Big Data Analytics. Advances in Environmental Engineering and Green Technologies 2021, 203 -230.
AMA StyleVijander Singh, Amit Kumar Bairwa, Deepak Sinwar. An Analysis of Big Data Analytics. Advances in Environmental Engineering and Green Technologies. 2021; ():203-230.
Chicago/Turabian StyleVijander Singh; Amit Kumar Bairwa; Deepak Sinwar. 2021. "An Analysis of Big Data Analytics." Advances in Environmental Engineering and Green Technologies , no. : 203-230.
The novel Corona Virus Disease (COVID-19) has critically influenced millions of human lives and economies globally. Governments around the globe are trying to cope with this pandemic and its adverse effects. Total symptomatic cases of the 2019 novel coronavirus (2019-nCoV) are growing at an exponential rate. There are several key parameters that are responsible for this COVID-19 outbreak which needs to be studied carefully. In this regard, this paper has presented a study of those parameters for finding out their impacts in calculating the total COVID-19 affected cases and deceases. Multiple Linear Regression and Multi-Layer Feed Forward Neural Network are used for this purpose. Analysis based on experimental study shown a strong correlation of age with total number of deaths; and population, diabetic prevalent with total number of cases. It is foremost required to focus on these parameters for minimizing the total affected cases and subsequent deaths due to COVID-19.
Ashish Kumar; Deepak Sinwar; Monika Saini. Study of several key parameters responsible for COVID-19 outbreak using multiple regression analysis and multi-layer feed forward neural network. Journal of Interdisciplinary Mathematics 2020, 24, 53 -75.
AMA StyleAshish Kumar, Deepak Sinwar, Monika Saini. Study of several key parameters responsible for COVID-19 outbreak using multiple regression analysis and multi-layer feed forward neural network. Journal of Interdisciplinary Mathematics. 2020; 24 (1):53-75.
Chicago/Turabian StyleAshish Kumar; Deepak Sinwar; Monika Saini. 2020. "Study of several key parameters responsible for COVID-19 outbreak using multiple regression analysis and multi-layer feed forward neural network." Journal of Interdisciplinary Mathematics 24, no. 1: 53-75.
Due to the dynamic nature of Mobile Ad hoc NETwork (MANET) the designing and development of protocols is very challenging. MANET has a limited number of resources i.e., power, infrastructure, etc. The need for an optimized path for communication amongst node has attracted the use of Swarm Intelligence (SI) based techniques such as Ant-Colony Optimization (ACO), Particle Swarm Optimization (PSO). In sort to broaden the duration of communication availability, the protocol needs to be optimized for finding out the greatest path. SI based techniques are able to solve these numerous routing problems. ACO provides the optimized packet delivery rate, throughput with low power consumption and packet delay. This paper compares the performance of various existing routing protocols viz. AODV, AOMDV as reactive, DSDV as proactive and ACOP using Random Waypoint Mobility Model. The purpose of using this mobility model is generating different scenarios for the same purpose. The analysis of these protocols has been made by implementing irregularities in the scenario using Network Simulator (NS2). Various performance metrics including Packet Delivery Fraction, Throughput and End to end Delay are used for validating the comparative study. Experimental results of the simulation indicate that the performance of the ACOP protocol is found better than other protocols of study.
Deepak Sinwar; Nisha Sharma; Sunil Kumar Maakar; Sudesh Kumar. Analysis and comparison of ant colony optimization algorithm with DSDV, AODV, and AOMDV based on shortest path in MANET. Journal of Information and Optimization Sciences 2020, 41, 621 -632.
AMA StyleDeepak Sinwar, Nisha Sharma, Sunil Kumar Maakar, Sudesh Kumar. Analysis and comparison of ant colony optimization algorithm with DSDV, AODV, and AOMDV based on shortest path in MANET. Journal of Information and Optimization Sciences. 2020; 41 (2):621-632.
Chicago/Turabian StyleDeepak Sinwar; Nisha Sharma; Sunil Kumar Maakar; Sudesh Kumar. 2020. "Analysis and comparison of ant colony optimization algorithm with DSDV, AODV, and AOMDV based on shortest path in MANET." Journal of Information and Optimization Sciences 41, no. 2: 621-632.
This chapter presents different techniques and applications of Artificial Intelligence for yield prediction and smart irrigation. Timely prediction of irrigation requirements and crop yields is necessary for farmer’s welfare and satisfaction. The beforehand prediction significantly contributes to minimizing production cost and maximizing crop yields. The precise prediction of crops’ yields is also useful for government, as it is effective in planning various schemes, transport needs, buying mechanisms, storage infrastructure, and liquid position of the economy before actual selling of crop by farmers to market. This chapter acknowledges the past breakthroughs and emerging Artificial Intelligence-based techniques in precision farming specifically for yield prediction and smart irrigation. Artificial Intelligence-based system provides sufficient information about crop yields at an early stage and its associated smart irrigation management system is effective in the judicious use of essential resources such as water and energy for agriculture.
Deepak Sinwar; Vijaypal Singh Dhaka; Manoj Kumar Sharma; Geeta Rani. AI-Based Yield Prediction and Smart Irrigation. Studies in Big Data 2019, 155 -180.
AMA StyleDeepak Sinwar, Vijaypal Singh Dhaka, Manoj Kumar Sharma, Geeta Rani. AI-Based Yield Prediction and Smart Irrigation. Studies in Big Data. 2019; ():155-180.
Chicago/Turabian StyleDeepak Sinwar; Vijaypal Singh Dhaka; Manoj Kumar Sharma; Geeta Rani. 2019. "AI-Based Yield Prediction and Smart Irrigation." Studies in Big Data , no. : 155-180.
Face recognition systems are models to recognize human faces on the basis of some computation methods. There are lots of applications (Attendance systems, Criminal identifications, Surveillance systems, Human Computer Interfaces etc.) of such systems. This paper aims to build a system which is capable of providing the user with the ability to first create an image database by face detection and feature extraction, and then this database can be used to perform face recognition by comparing the input image (obtained through live video streaming) with the images stored during the training phase. This method used Viola-Jones algorithm for face detection (to create the training set), and Principal Component Analysis (PCA) for face recognition. We have created our own dataset for validating this automatic face recognition system. Experimental work of proposed method under identical physical conditions on our dataset gives 82% accuracy of face recognition.
Kushagra Bisaria; Deepak Sinwar; Manoj Kumar Sharma. A Study on Real Time Face Recognition Using Feature Based Image Processing Frameworks. SSRN Electronic Journal 2019, 1 .
AMA StyleKushagra Bisaria, Deepak Sinwar, Manoj Kumar Sharma. A Study on Real Time Face Recognition Using Feature Based Image Processing Frameworks. SSRN Electronic Journal. 2019; ():1.
Chicago/Turabian StyleKushagra Bisaria; Deepak Sinwar; Manoj Kumar Sharma. 2019. "A Study on Real Time Face Recognition Using Feature Based Image Processing Frameworks." SSRN Electronic Journal , no. : 1.
Mining of frequent patterns have attracted great attention in the last few decades due to their vast variety of applications in real life. Recently the problem of detecting outliers from transactional datasets has been considered as the process of mining infrequent patterns. As we know that frequent patterns are those patterns in transactional datasets which are preferred by most of the customers; whereas infrequent patterns are not. Here the concept of outliers is being considered as those patterns which are infrequent at all but their weight count is higher from threshold values. This paper has presented an approach WFPG to detect infrequent weighted patterns from transactional datasets. Experimental work and theoretical analysis show that the WFPG has detected weighted infrequent patterns/ item sets with lower support count as compared to their weight value.
Deepak Sinwar; Vijaypal Singh Dhaka; Manoj Kumar Sharma. Detecting Infrequent Weighted Patterns as Outliers using WFPG. 2018 3rd International Innovative Applications of Computational Intelligence on Power, Energy and Controls with their Impact on Humanity (CIPECH) 2018, 1 -5.
AMA StyleDeepak Sinwar, Vijaypal Singh Dhaka, Manoj Kumar Sharma. Detecting Infrequent Weighted Patterns as Outliers using WFPG. 2018 3rd International Innovative Applications of Computational Intelligence on Power, Energy and Controls with their Impact on Humanity (CIPECH). 2018; ():1-5.
Chicago/Turabian StyleDeepak Sinwar; Vijaypal Singh Dhaka; Manoj Kumar Sharma. 2018. "Detecting Infrequent Weighted Patterns as Outliers using WFPG." 2018 3rd International Innovative Applications of Computational Intelligence on Power, Energy and Controls with their Impact on Humanity (CIPECH) , no. : 1-5.
As we know that Outlier detection is one of the important aspects of Data Mining, which generally aims to identify potential outliers from datasets. Outliers may sometimes plays important role while taking effective business decisions. This work provides a study of various outlier detection techniques and compares their effectiveness in terms of number of outlier detection, kappa statistic and mean absolute error. Seven algorithms of different categories were tested on three real world datasets to validate the study. We have used pattern based detection of outliers using Multilayer Perceptron, Radial Basis Function Networks, Naïve Bayes Classifiers and Pattern Clustering techniques viz. K-Means, EM and the Agglomerative Hierarchical Clustering. Experimental results show that the Hierarchical Clustering outperforms all other algorithms in terms of number of outlier detection, whereas Multilayer Perceptron and J48 Decision Tree have the highest Kappa Statistic measure. Performance of EM clustering was worst amongst all the algorithms because it was unable to classify all the instances; whereas the performance of RBF Networks and Naïve Bayes Classifiers was almost same and not so satisfactory in terms of outlier detection percentage, Kappa Statistic and Mean Absolute Error.
Deepak Sinwar; V. S. Dhaka. Outlier detection from multidimensional space using multilayer perceptron, RBF networks and pattern clustering techniques. 2015 International Conference on Advances in Computer Engineering and Applications 2015, 573 -579.
AMA StyleDeepak Sinwar, V. S. Dhaka. Outlier detection from multidimensional space using multilayer perceptron, RBF networks and pattern clustering techniques. 2015 International Conference on Advances in Computer Engineering and Applications. 2015; ():573-579.
Chicago/Turabian StyleDeepak Sinwar; V. S. Dhaka. 2015. "Outlier detection from multidimensional space using multilayer perceptron, RBF networks and pattern clustering techniques." 2015 International Conference on Advances in Computer Engineering and Applications , no. : 573-579.