Browsing by Author "Mukund, Y.R."

Filter results by typing the first few letters
Now showing 1 - 6 of 6
  • No Thumbnail Available
    Item
    Improving false alarm rate in intrusion detection systems using Hadoop
    (2016) Mukund, Y.R.; Nayak, S.S.; Chandrasekaran, K.
    Intrusion Detection Systems are a vital part of an organization's security. This paper gives an account of the existing algorithms for Intrusion Detection using Machine Learning, along with certain new ideas for improving the same. The paper mainly talks about employing the Decision Tree mechanism for Intrusion Detection and improve it with the distributed file system, Hadoop. Initially a method that uses a dirty-flags to check the consistency of the Decision Tree, which changes with every wrong classification of the system is employed. The wrong classification is identified by a certain user who informs the system about the same and helps it learn. In the further sections, a new method which does not use a dirty-flag, but rather modifies the Key-Value pair in the results of the reduce() function is tested as an improvement to the previous method. The two methods are compared, with the help of the Hadoop Simulation Tool - YARN. The main aim of the paper is to propose the use of the Distributed File System for Machine Learning along with some improvements to the current Hadoop File System, so that it reduces the total Time Taken, when Machine Learning algorithms are employed along with it. � 2016 IEEE.
  • No Thumbnail Available
    Item
    Improving false alarm rate in intrusion detection systems using Hadoop
    (Institute of Electrical and Electronics Engineers Inc., 2016) Mukund, Y.R.; Nayak, S.S.; Chandrasekaran, K.
    Intrusion Detection Systems are a vital part of an organization's security. This paper gives an account of the existing algorithms for Intrusion Detection using Machine Learning, along with certain new ideas for improving the same. The paper mainly talks about employing the Decision Tree mechanism for Intrusion Detection and improve it with the distributed file system, Hadoop. Initially a method that uses a dirty-flags to check the consistency of the Decision Tree, which changes with every wrong classification of the system is employed. The wrong classification is identified by a certain user who informs the system about the same and helps it learn. In the further sections, a new method which does not use a dirty-flag, but rather modifies the Key-Value pair in the results of the reduce() function is tested as an improvement to the previous method. The two methods are compared, with the help of the Hadoop Simulation Tool - YARN. The main aim of the paper is to propose the use of the Distributed File System for Machine Learning along with some improvements to the current Hadoop File System, so that it reduces the total Time Taken, when Machine Learning algorithms are employed along with it. © 2016 IEEE.
  • No Thumbnail Available
    Item
    Improving RED for reduced UDP packet-drop
    (2015) Mukund, Y.R.; Rohit, C.; Chandavarkar, B.R.
    This paper gives an understanding of how the Random Early Detection(RED) algorithm can be implemented in a network involving UDP sources. The User Datagram Protocol(UDP) protocol is an unreliable protocol and does not have the mechanism to detect the packet drops that are carried out by the RED gateway which results in a bias by the gateway against UDP packets. By manipulating certain parameters of the RED algorithm we can make the algorithm less biased against the UDP packets. Various methods have been simulated and their corresponding results are shown in this paper. The aim is to show that the algorithm can be used in a network which consists of both the TCP and the UDP sources. � 2015 IEEE.
  • No Thumbnail Available
    Item
    Improving RED for reduced UDP packet-drop
    (Institute of Electrical and Electronics Engineers Inc., 2015) Mukund, Y.R.; Rohit, C.; Chandavarkar, B.R.
    This paper gives an understanding of how the Random Early Detection(RED) algorithm can be implemented in a network involving UDP sources. The User Datagram Protocol(UDP) protocol is an unreliable protocol and does not have the mechanism to detect the packet drops that are carried out by the RED gateway which results in a bias by the gateway against UDP packets. By manipulating certain parameters of the RED algorithm we can make the algorithm less biased against the UDP packets. Various methods have been simulated and their corresponding results are shown in this paper. The aim is to show that the algorithm can be used in a network which consists of both the TCP and the UDP sources. © 2015 IEEE.
  • No Thumbnail Available
    Item
    Influence of news on individual confidence bias in stock markets
    (2016) Mukund, Y.R.; Naresh, V.; Patil, S.; Chandrasekaran, K.; Vijaya, Kumar, V.; Gnanamurthy, R.K.
    The Phenomenon of stock markets is a complex one and is something which, has attracted researchers and statisticians for a long time. Complex statistics have long dominated this field where the prediction models are usually stochastic. The advent of machine learning gave us a new way of looking at the problem. Much work has been done in analyzing the stock market to predict the stock index of a particular or-ganization. However, most of the work done is based on the previous stock data and other statistical parameters. Our work, uses data such as the online news articles about a particular company and aims to help a trader conclude the market sentiment towards that company through sentiment analysis. The online raw data is obtained through crawling and is indexed, weighted and subject to sentiment analysis to output the final sentiment of the market. It is found that the Naive-Bayesian Classifier is the more suitable op-tion among the Decision Tree and Random Forests for the task of sentiment analysis. The Final Sentiment Factor ar-rived at, is found to reect the real time market sentiment quite accurately. It is also shown that the sentiment factor can be used as an input to a more complex analysis model. This new model, performs better than the existing models.
  • No Thumbnail Available
    Item
    Influence of news on individual confidence bias in stock markets
    (Association for Computing Machinery acmhelp@acm.org, 2016) Mukund, Y.R.; Naresh, V.; Patil, S.; Chandrasekaran, K.; Vijaya, V.; Gnanamurthy, R.K.
    The Phenomenon of stock markets is a complex one and is something which, has attracted researchers and statisticians for a long time. Complex statistics have long dominated this field where the prediction models are usually stochastic. The advent of machine learning gave us a new way of looking at the problem. Much work has been done in analyzing the stock market to predict the stock index of a particular or-ganization. However, most of the work done is based on the previous stock data and other statistical parameters. Our work, uses data such as the online news articles about a particular company and aims to help a trader conclude the market sentiment towards that company through sentiment analysis. The online raw data is obtained through crawling and is indexed, weighted and subject to sentiment analysis to output the final sentiment of the market. It is found that the Naive-Bayesian Classifier is the more suitable op-tion among the Decision Tree and Random Forests for the task of sentiment analysis. The Final Sentiment Factor ar-rived at, is found to reect the real time market sentiment quite accurately. It is also shown that the sentiment factor can be used as an input to a more complex analysis model. This new model, performs better than the existing models.