In this paper, a novel method for EEG(Electroencephalography) based emotion recognition is introd... more In this paper, a novel method for EEG(Electroencephalography) based emotion recognition is introduced. This method uses transfer learning to extract features from multichannel EEG signals, these features are then arranged in an 8×9 map to represent their spatial location on scalp and then we introduce a CNN model which takes in the spatial feature map and extracts spatial relations between EEG channel and finally classify the emotions. First, EEG signals are converted to spectrogram and passed through a pre-trained image classification model to get a feature vector from spectrogram of EEG. Then, feature vectors of different channels are rearranged and are presented as input to a CNN model which extracts spatial features or dependencies of channels as part of training. Finally, CNN outputs are flattened and passed through dense layer to classify between emotion classes. In this study, SEED, SEED-IV and SEED-V EEG emotion data-sets are used for classification and our method achieves best classification accuracy of 97.09% on SEED, 89.81% on SEED-IV and 88.23% on SEED-V data-set with fivefold cross validation.
Digital images play a very important role in different areas in the modern technological scenario... more Digital images play a very important role in different areas in the modern technological scenario. Changing and manipulating the content of the digital image is a very easy task by using powerful image editing tools. In today's technology environment, digital photographs serve a critical function in a variety of fields. Using advanced image editing tools, changing and rearranging the content of a digital image is a simple operation. It is now possible to add, edit, or remove essential aspects from an image despite leaving any perceptible alterations. In addition to determining if the picture is authentic or forged, the metadata of the image may be examined, however, metadata can be altered. In this example, the authors use Error Level Analysis on each picture and matching parameters for error rate analysis to detect images of modifications using Deep Learning on a dataset of a false image and real photos. This experiment shows that by running through 100 epochs, we obtain the be...
Accessing remote services and data is one of the most important applications in Mobile Ad Hoc Net... more Accessing remote services and data is one of the most important applications in Mobile Ad Hoc Networks. Data replication is one of the most popular techniques used to increase data availability by replicating frequently accessed data items locally or nearby. In MANETs, data replication helps to avoid data losses in case of unpredictable network partition and also aids in reducing the number of hops that a data is transmitted from source to destination. In this paper, an extension to DCG (Dynamic Connectivity based Grouping) [5] is proposed called W-DCG (Weighted Dynamic Connectivity based Grouping), which aims at addressing the limitations of DCG and most of the other proposed protocols in literature. The proposed protocol considers node properties such as memory capacity and remaining battery power while replicating data on nodes to achieve better load-balancing. Moreover it uses dynamic access frequencies to replicate data items which are currently in demand. Data sharing is done ...
An efficient analysis of crop yield prediction using Hadoop framework based on random forest approach
In the growth of Information T echnology, Big data come forth as a blazing topic. The main source... more In the growth of Information T echnology, Big data come forth as a blazing topic. The main source of human survival depends on agriculture; where it needs a key contribution in the field of crop data analysis. This paper gives a purpose about how to find experiences from accuracy agriculture information through big data approach. In this way, gathering the valuable data in an effective way drives a framework towards major computational challenges in crop analysis where information is remotely gathered. For the storage purpose of huge data availability in agriculture, we are intending Hadoop framework for our work to store a huge volume of crop data. This work gives a better prediction for the farmers to plant which kind of crops to their farm field based on their soil content to improve the productivity. The random forest algorithm is integrated with the MapReduce programming model in Hadoop framework.
Analysis of feature selection techniques for denial of service (DoS) attacks
2018 4th International Conference on Recent Advances in Information Technology (RAIT), 2018
Network traffic is increasing attributable to the growing use of smart devices and also the Web. ... more Network traffic is increasing attributable to the growing use of smart devices and also the Web. Most intrusion detection research has targeted on feature choice or reduction for building an Intrusion detection system (IDS) that's computationally economical and effective. The proposed system makes use of a heuristic technique which filters the attributes using Information gain, Gain ratio and Correlation that is employed to sight majority category attack i.e., Denial of Service using KDD dataset.
Viable Crop Prediction Scenario in BigData Using a Novel Approach
Agriculture leads a vital role in the human surveillance where it lies as the initial step for th... more Agriculture leads a vital role in the human surveillance where it lies as the initial step for the human civilization. Due to the excessive need for food, the agricultural practices are in large-scale production as a business which is termed as “Agribusiness”. Modern agriculture allows both biological and technological developments such as plant breeding, agrochemicals, genetic breeding, remote sensing, crop monitoring, sensor nodes, and automatic maintenance system. The integration of sensor nodes in the farming field leads to the generation of huge data which can need a learning algorithm for analyses to determine a specific solution. There are various machine learning algorithms in practices which are not suitable for handling large datasets. In this paper, a novel algorithm “Agrifi-prediction algorithm” is created which has the functionality of loading the dataset in hdfs and comparing the previous dataset with the current processing dataset. The experimental process is carried ...
Extending OSGi bundle level resource monitoring for implementation portability
2017 International Conference on Recent Innovations in Signal processing and Embedded Systems (RISE), 2017
OSGi is a specification for modular application development in Java. In a modular application dev... more OSGi is a specification for modular application development in Java. In a modular application development environment, an application is composed of many standalone modules (components). Most of these components are reusable and could be used off the shelf. In such an environment, the application efficiency is determined by the constituent components efficiency. So it is important to track the resource consumption of each component. This paper analyses already proposed solutions for monitoring component level resource consumption. The major drawback of the already existing solutions is that they are dependent on the OSGi implementations based on which they are developed. This paper explains the reason for dependency on OSGi implementation and proposes a solution to avoid implementation dependency.
Node stability based clustering algorithm for mobile ad hoc networks
2008 Fourth International Conference on Wireless Communication and Sensor Networks, 2008
Node mobility is an important factor at the time of clus-tering in ad hoc network because it dire... more Node mobility is an important factor at the time of clus-tering in ad hoc network because it directly affects the stability of cluster. In the proposed work a distributed clustering algorithm which uses node mobility as a main concern with battery power and connectivity in terms of no. of ...
A Comparative Study of Local Outlier Factor Algorithms for Outliers Detection in Data Streams
Outlier detection analyzes data, finds out anomalies, and helps to discover unforeseen activities... more Outlier detection analyzes data, finds out anomalies, and helps to discover unforeseen activities in safety crucial systems. Outlier detection helps in early prediction of various fraudulent activities like credit card theft, fake insurance claim, tax stealing, real-time monitoring, medical systems, online transactions, and many more. Detection of outliers in the data streams is extremely challenging as compared to static data since data streams are continuous, highly changing and unending in nature. Density-based outlier detection using local outlier factor (LOF) is the prominent method for detecting the outliers in the data streams. In this paper, we provide an insight on outlier detection and various challenges involved while detection of outliers in the data streams. We concentrate on density-based outlier detection and review major local outlier factor (LOF) based outlier detection algorithms in details. We also perform a comparative study of existing LOF algorithms to evaluate...
Delay Tolerant Networks are resource constraint network. Selfish behavior of a node causes it to ... more Delay Tolerant Networks are resource constraint network. Selfish behavior of a node causes it to drop legitimate packets of other nodes, leading to low message delivery and wastage of valuable network resources. Here a Minimax Theory Based Scheme is proposed to distribute credits among non selfish nodes .It reduces the message dropping rate, detects selfish node and enhance network performance in terms of latency, aborted message rate and delivery ratio.
A New Master Key Agreement Scheme for Secure Group Communication in Vehicular ad hoc Network
Vehicular ad-hoc networks are obtained by applying important features of mobile ad hoc networks, ... more Vehicular ad-hoc networks are obtained by applying important features of mobile ad hoc networks, in this application, it is not easy to preserve security and make this network private. Various attackers try to exploit VANETs and send forgery messages to deceive other vehicles which may lead harmful for driver/ passenger. To improve traffic safety and comfort of driving and travelling, it is required to obtain integrity of the messages, identifying and defending against the misbehaving vehicles that should be signed and verified before they are trusted while the real identity of vehicles should not be revealed. In this paper, we propose a master key agreement schemes for VANETs, vehicles are divided into separate different groups. In each group, group leader manages its members and distributes public / private key pairs to members for communication. Here group leader contains master key, which may require for encrypting session key and other communication information in the same grou...
MANET is easily vulnerable because of dynamic topology, infrastructure less nature and also due t... more MANET is easily vulnerable because of dynamic topology, infrastructure less nature and also due to lack of centralized administration. One of the most common attack on such networks is gray hole attack which drop some selective data. It is challenge to keep the network free form this attack. This paper presents a technique for detection of group gray hole attack through destination based scheme when more than one malicious nodes are in
Android based devices are rapidly flourishing day-by-day, due to its ease of use and popularity. ... more Android based devices are rapidly flourishing day-by-day, due to its ease of use and popularity. As a result, the number of malware attacks on Android is also increasing. This paper is based on the Text Mining approach for analyzing Android malware families. The proposed methodology is motivated by the method introduced by Guillermo Suarez-Tangil which aims to automate malware analysis process based on DENDROID. The main issue in this regard is the storage of Family Feature Vectors (FFV) which is stored as sparse matrix. Therefore, this work presents a novel concept of Compressed Row Storage (CRS) to store the statistical features intellectually. By implementing this methodology, the FFV of Malware families are stored in an efficient manner. The experimental result proves that the large reduction (79%) in space needed to store FFV which incorporates only the non-zero elements is observed. This eventually leads to the reduction in the Feature Vector generation time and the Total proc...
On Maximizing Blind Rendezvous Probability in Cognitive Radio Ad Hoc Networks
Establishment of wireless communication links among the cognitive radio (CR) users via rendezvous... more Establishment of wireless communication links among the cognitive radio (CR) users via rendezvous on commonly available channels is the crucial process in cognitive radio ad hoc networks. Extant rendezvous methods utilize the channel hopping (CH) technique (also referred as blind rendezvous) and generate the CH sequence on the top of potential licensed channels and employ random replace operation to contrive the CH sequence for individual CR user. However, incorporation of random replace operation may exhibit poorer performance with the increase in the number of available channels owing to the severe decrease in the probability of replacing an unavailable channel with the same available channel by two or more CR users. In this paper, we present a channel-ranking rendezvous procedure that ranks available channels based on the received SINR and constructs the CH sequence by replacing the unavailable channels with higher ranked available channels. We verify the superiority of channel-r...
In this paper, a novel method for EEG(Electroencephalography) based emotion recognition is introd... more In this paper, a novel method for EEG(Electroencephalography) based emotion recognition is introduced. This method uses transfer learning to extract features from multichannel EEG signals, these features are then arranged in an 8×9 map to represent their spatial location on scalp and then we introduce a CNN model which takes in the spatial feature map and extracts spatial relations between EEG channel and finally classify the emotions. First, EEG signals are converted to spectrogram and passed through a pre-trained image classification model to get a feature vector from spectrogram of EEG. Then, feature vectors of different channels are rearranged and are presented as input to a CNN model which extracts spatial features or dependencies of channels as part of training. Finally, CNN outputs are flattened and passed through dense layer to classify between emotion classes. In this study, SEED, SEED-IV and SEED-V EEG emotion data-sets are used for classification and our method achieves best classification accuracy of 97.09% on SEED, 89.81% on SEED-IV and 88.23% on SEED-V data-set with fivefold cross validation.
Digital images play a very important role in different areas in the modern technological scenario... more Digital images play a very important role in different areas in the modern technological scenario. Changing and manipulating the content of the digital image is a very easy task by using powerful image editing tools. In today's technology environment, digital photographs serve a critical function in a variety of fields. Using advanced image editing tools, changing and rearranging the content of a digital image is a simple operation. It is now possible to add, edit, or remove essential aspects from an image despite leaving any perceptible alterations. In addition to determining if the picture is authentic or forged, the metadata of the image may be examined, however, metadata can be altered. In this example, the authors use Error Level Analysis on each picture and matching parameters for error rate analysis to detect images of modifications using Deep Learning on a dataset of a false image and real photos. This experiment shows that by running through 100 epochs, we obtain the be...
Accessing remote services and data is one of the most important applications in Mobile Ad Hoc Net... more Accessing remote services and data is one of the most important applications in Mobile Ad Hoc Networks. Data replication is one of the most popular techniques used to increase data availability by replicating frequently accessed data items locally or nearby. In MANETs, data replication helps to avoid data losses in case of unpredictable network partition and also aids in reducing the number of hops that a data is transmitted from source to destination. In this paper, an extension to DCG (Dynamic Connectivity based Grouping) [5] is proposed called W-DCG (Weighted Dynamic Connectivity based Grouping), which aims at addressing the limitations of DCG and most of the other proposed protocols in literature. The proposed protocol considers node properties such as memory capacity and remaining battery power while replicating data on nodes to achieve better load-balancing. Moreover it uses dynamic access frequencies to replicate data items which are currently in demand. Data sharing is done ...
An efficient analysis of crop yield prediction using Hadoop framework based on random forest approach
In the growth of Information T echnology, Big data come forth as a blazing topic. The main source... more In the growth of Information T echnology, Big data come forth as a blazing topic. The main source of human survival depends on agriculture; where it needs a key contribution in the field of crop data analysis. This paper gives a purpose about how to find experiences from accuracy agriculture information through big data approach. In this way, gathering the valuable data in an effective way drives a framework towards major computational challenges in crop analysis where information is remotely gathered. For the storage purpose of huge data availability in agriculture, we are intending Hadoop framework for our work to store a huge volume of crop data. This work gives a better prediction for the farmers to plant which kind of crops to their farm field based on their soil content to improve the productivity. The random forest algorithm is integrated with the MapReduce programming model in Hadoop framework.
Analysis of feature selection techniques for denial of service (DoS) attacks
2018 4th International Conference on Recent Advances in Information Technology (RAIT), 2018
Network traffic is increasing attributable to the growing use of smart devices and also the Web. ... more Network traffic is increasing attributable to the growing use of smart devices and also the Web. Most intrusion detection research has targeted on feature choice or reduction for building an Intrusion detection system (IDS) that's computationally economical and effective. The proposed system makes use of a heuristic technique which filters the attributes using Information gain, Gain ratio and Correlation that is employed to sight majority category attack i.e., Denial of Service using KDD dataset.
Viable Crop Prediction Scenario in BigData Using a Novel Approach
Agriculture leads a vital role in the human surveillance where it lies as the initial step for th... more Agriculture leads a vital role in the human surveillance where it lies as the initial step for the human civilization. Due to the excessive need for food, the agricultural practices are in large-scale production as a business which is termed as “Agribusiness”. Modern agriculture allows both biological and technological developments such as plant breeding, agrochemicals, genetic breeding, remote sensing, crop monitoring, sensor nodes, and automatic maintenance system. The integration of sensor nodes in the farming field leads to the generation of huge data which can need a learning algorithm for analyses to determine a specific solution. There are various machine learning algorithms in practices which are not suitable for handling large datasets. In this paper, a novel algorithm “Agrifi-prediction algorithm” is created which has the functionality of loading the dataset in hdfs and comparing the previous dataset with the current processing dataset. The experimental process is carried ...
Extending OSGi bundle level resource monitoring for implementation portability
2017 International Conference on Recent Innovations in Signal processing and Embedded Systems (RISE), 2017
OSGi is a specification for modular application development in Java. In a modular application dev... more OSGi is a specification for modular application development in Java. In a modular application development environment, an application is composed of many standalone modules (components). Most of these components are reusable and could be used off the shelf. In such an environment, the application efficiency is determined by the constituent components efficiency. So it is important to track the resource consumption of each component. This paper analyses already proposed solutions for monitoring component level resource consumption. The major drawback of the already existing solutions is that they are dependent on the OSGi implementations based on which they are developed. This paper explains the reason for dependency on OSGi implementation and proposes a solution to avoid implementation dependency.
Node stability based clustering algorithm for mobile ad hoc networks
2008 Fourth International Conference on Wireless Communication and Sensor Networks, 2008
Node mobility is an important factor at the time of clus-tering in ad hoc network because it dire... more Node mobility is an important factor at the time of clus-tering in ad hoc network because it directly affects the stability of cluster. In the proposed work a distributed clustering algorithm which uses node mobility as a main concern with battery power and connectivity in terms of no. of ...
A Comparative Study of Local Outlier Factor Algorithms for Outliers Detection in Data Streams
Outlier detection analyzes data, finds out anomalies, and helps to discover unforeseen activities... more Outlier detection analyzes data, finds out anomalies, and helps to discover unforeseen activities in safety crucial systems. Outlier detection helps in early prediction of various fraudulent activities like credit card theft, fake insurance claim, tax stealing, real-time monitoring, medical systems, online transactions, and many more. Detection of outliers in the data streams is extremely challenging as compared to static data since data streams are continuous, highly changing and unending in nature. Density-based outlier detection using local outlier factor (LOF) is the prominent method for detecting the outliers in the data streams. In this paper, we provide an insight on outlier detection and various challenges involved while detection of outliers in the data streams. We concentrate on density-based outlier detection and review major local outlier factor (LOF) based outlier detection algorithms in details. We also perform a comparative study of existing LOF algorithms to evaluate...
Delay Tolerant Networks are resource constraint network. Selfish behavior of a node causes it to ... more Delay Tolerant Networks are resource constraint network. Selfish behavior of a node causes it to drop legitimate packets of other nodes, leading to low message delivery and wastage of valuable network resources. Here a Minimax Theory Based Scheme is proposed to distribute credits among non selfish nodes .It reduces the message dropping rate, detects selfish node and enhance network performance in terms of latency, aborted message rate and delivery ratio.
A New Master Key Agreement Scheme for Secure Group Communication in Vehicular ad hoc Network
Vehicular ad-hoc networks are obtained by applying important features of mobile ad hoc networks, ... more Vehicular ad-hoc networks are obtained by applying important features of mobile ad hoc networks, in this application, it is not easy to preserve security and make this network private. Various attackers try to exploit VANETs and send forgery messages to deceive other vehicles which may lead harmful for driver/ passenger. To improve traffic safety and comfort of driving and travelling, it is required to obtain integrity of the messages, identifying and defending against the misbehaving vehicles that should be signed and verified before they are trusted while the real identity of vehicles should not be revealed. In this paper, we propose a master key agreement schemes for VANETs, vehicles are divided into separate different groups. In each group, group leader manages its members and distributes public / private key pairs to members for communication. Here group leader contains master key, which may require for encrypting session key and other communication information in the same grou...
MANET is easily vulnerable because of dynamic topology, infrastructure less nature and also due t... more MANET is easily vulnerable because of dynamic topology, infrastructure less nature and also due to lack of centralized administration. One of the most common attack on such networks is gray hole attack which drop some selective data. It is challenge to keep the network free form this attack. This paper presents a technique for detection of group gray hole attack through destination based scheme when more than one malicious nodes are in
Android based devices are rapidly flourishing day-by-day, due to its ease of use and popularity. ... more Android based devices are rapidly flourishing day-by-day, due to its ease of use and popularity. As a result, the number of malware attacks on Android is also increasing. This paper is based on the Text Mining approach for analyzing Android malware families. The proposed methodology is motivated by the method introduced by Guillermo Suarez-Tangil which aims to automate malware analysis process based on DENDROID. The main issue in this regard is the storage of Family Feature Vectors (FFV) which is stored as sparse matrix. Therefore, this work presents a novel concept of Compressed Row Storage (CRS) to store the statistical features intellectually. By implementing this methodology, the FFV of Malware families are stored in an efficient manner. The experimental result proves that the large reduction (79%) in space needed to store FFV which incorporates only the non-zero elements is observed. This eventually leads to the reduction in the Feature Vector generation time and the Total proc...
On Maximizing Blind Rendezvous Probability in Cognitive Radio Ad Hoc Networks
Establishment of wireless communication links among the cognitive radio (CR) users via rendezvous... more Establishment of wireless communication links among the cognitive radio (CR) users via rendezvous on commonly available channels is the crucial process in cognitive radio ad hoc networks. Extant rendezvous methods utilize the channel hopping (CH) technique (also referred as blind rendezvous) and generate the CH sequence on the top of potential licensed channels and employ random replace operation to contrive the CH sequence for individual CR user. However, incorporation of random replace operation may exhibit poorer performance with the increase in the number of available channels owing to the severe decrease in the probability of replacing an unavailable channel with the same available channel by two or more CR users. In this paper, we present a channel-ranking rendezvous procedure that ranks available channels based on the received SINR and constructs the CH sequence by replacing the unavailable channels with higher ranked available channels. We verify the superiority of channel-r...
Uploads
Papers by Meenu Chawla