Streamlining Big Data Mining with Cutting-Edge Approaches | Chapter 10 | Research Updates in Mathematics and Computer Science Vol. 2

 Big data refers to any assortment of data which are outsized and intricate in nature such that conventional database administration systems and data processing tools cannot process. Feature selection serves primarily to reduce the processing burden of data mining models. To expedite the processing of large volumes of data, parallel processing is implemented using the MapReduce (MR) technique. MapReduce model is applied to big datasets, which is further divided into smaller partition. However, existing algorithms often fall short in enhancing classifier performance significantly. This research advocates for the use of the MR method to conduct feature selection in parallel, thereby improving performance. Additionally, to augment classifier efficacy, this study introduces an innovative approach combining Online Feature Selection (OFS) with an Accelerated Bat Algorithm (ABA) within a framework that pre-processes features ahead of time, without prior knowledge of the feature space. The proposed OFS-ABA method is designed to select relevant and non-redundant features within the MapReduce (MR) framework. Furthermore, an Ensemble Incremental Deep Multiple Layer Perceptron (EIDMLP) classifier is employed to classify dataset samples. The outputs from homogeneous IDMLP classifiers are aggregated using the EIDMPL classifier. The proposed feature selection method and classifier are extensively evaluated across three high-dimensional datasets. The results indicate that the MR-OFS-ABA method outperforms existing feature selection methods such as PSO, APSO, and ASAMO (Accelerated Simulated Annealing and Mutation Operator). Additionally, the performance of the EIDMLP classifier is compared with other existing classifiers including Naïve Bayes (NB), Hoeffding Tree (HT), and Fuzzy Minimal Consistent Class Subset Coverage (FMCCSC)-KNN (K Nearest Neighbor). The methodology is applied to three datasets, and the results are compared across four classifiers and three state-of-the-art feature selection algorithms. Overall, the findings of this research demonstrate improved accuracy and reduced processing time.

Author(s) Details:

Renuka Devi D,
Department of Computer Science Stella Maris College (Autonomous), Chennai, Tamil Nadu, India.

Swetha Margaret TA,
Department of Computer Science Stella Maris College (Autonomous), Chennai, Tamil Nadu, India.

Please see the link here: https://stm.bookpi.org/RUMCS-V2/article/view/13960

Genetic Algorithm and Support Vector Machine for DNS Tunneling Detection: A Hybrid Method Approach | Chapter 10 | Novel Research Aspects in Mathematical and Computer Science Vol. 5

In order to find the best characteristics that may maximise the detection of DNS tunnelling, this research suggests a hybrid technique of genetic algorithm feature selection approach with the support vector machine classifier. Corporations are increasingly spending a lot of money developing web applications as online commerce expands. On the other hand, such dangers can leave companies open to future attacks. One of these dangers is DNS tunnelling, which sends dangerous information through the domain name protocol. As a result, confidential data would be exposed and violated. Machine learning has been the subject of several studies to develop a detecting technique. The strategies used by authors comprised a wide range of characteristics, such as domain length, Bytes, content, DNS traffic volume, hostnames per domain, location, and domain history are all factors. Evidently, there is a critical need for feature selection tasks to be supported in order to discover the finest features. A benchmark dataset for DNS tunnelling was used to assess the proposed method. The new method surpassed the classic SVM by receiving an F-measure of 0.946, proving that it was superior.

Author (s) Details:

Fuqdan A. Al-Ibraheemi,
College of Dentistry, University of Al-Ameed, Iraq.

Sattar Al-Ibraheemi,
Education Ministry, Iraq.

Haleh Amintoosi,
Faculty of Engineering, Ferdowsi University of Mashhad, Iran.

Please see the link here: https://stm.bookpi.org/NRAMCS-V5/article/view/7487

Cluster Based MapReduce Technique for Predicting Heart Disease -: A Modelling Approach | Chapter 11 | Recent Advances in Mathematical Research and Computer Science Vol. 6

 The process of classifying data components using class labels is known as classification. It is the technique of guided learning. It's also the process of identifying the model that explains and differentiates data types and their values. MapReduce is a programming paradigm that allows you to execute distributed computation on a computer cluster. The user application spawns a Master controller process and a succession of Map and Reduce jobs in a MapReduce system like Hadoop. The prediction system for heart disease is described in this study, which is based on the MapReduce approach with Relief feature selection and semi-naive Bayes Classification Algorithms.

Author(S) Details

J. Sukanya
Department of Computer Science, M. V. Muthiah Government Arts College for Women, Dindigul, Tamil Nadu, India.

K. Rajiv Gandhi
Department of Computer Science, Alagappa University Model Constituent College of Arts and Science, Paramakudi, Tamil Nadu, India.

V. Palanisamy
Department of Computer Applications, Alagappa University, Karaikudi, Tamil Nadu, India.

View Book:- https://stm.bookpi.org/RAMRCS-V6/article/view/5314

Semi-Automated Text Categorization Using Demonstration and Integration Based Term Set | Chapter 2 | New Approaches in Engineering Research Vol. 1

 Manual Analysis of massive amounts of textual data necessitates a tremendous amount of processing time and effort in interpreting the text and organizing it in the required format. Because of the high dimensionality of feature space, the main issue in the current scenario is text or document categorization. There are numerous methods available today for dealing with text feature selection. This paper aims to present one such semi-automated text categorization feature selection methodology for dealing with massive data sets by employing two phases of David Merrill's First Principles of Instruction (FPI). It employs a pre-defined category group by providing the appropriate training set based on the FPI's demonstration and integration phases Text tokenization, text categorization, and text analysis are all part of the methodology.

Author (S) Details

Dr. (Mrs.) M. Pushpa
Department of Computer Science, Quiad-e-Millath Government Arts College for Women, Chennai, India.

Dr. (Mrs.) K. Nirmala
Department of Computer Science, Quiad-e-Millath Government Arts College for Women, Chennai, India.

Dr. (Mrs.) J. Vijayalakshmi
Department of Information Technology, Sri Sai Ram Engineering College, West Tambaram, Chennai, India.

View Book :- https://stm.bookpi.org/NAER-V1/article/view/1908

Hybridized Swarm Optimization Classifiers with Ensemble Feature Ranking Techniques: Recent Study | Chapter 8 | Emerging Trends in Engineering Research and Technology Vol. 6

Intrusion Detection System (IDS) is a security support mechanism which has become an essential component of security infrastructure to detect attacks, identify and track the intruders. Intrusion Detection Systems are implemented in order to detect malicious activities and it functions behind the firewall, observing for patterns in network traffic that might indicate malicious action. The extreme development of the internet, the high occurrence of the threats over the internet has been the cause in recognizing the need for both IDS and firewall to help in securing a network. Currently many researchers have shown an increasing interest in intrusion detection based on data mining techniques and swarm intelligence techniques. Also, recent research focuses more on the hybridization of techniques to improve the performance of classifiers and it has become commonplace in IDSs which allows researchers to exploit the benefits of individual techniques and approaches. In intrusion detection, the quantity of data is huge that includes thousands of traffic records with number of various features. Selecting a subset of informative features can lead to improved classification accuracy. In this paper ensemble of feature ranking techniques are used to select the most relevant features that can represent the pattern of the network traffic. The efficiency of the presented method is validated on KDDCUP’99 dataset using hybrid swarm based classifier, Simplified Swarm Optimization (SSO) with Ant Colony Optimization (ACO). The performance of the proposed method is compared with the basic classifiers, SSO and hybridization of SSO with Support Vector Machine (SVM). It is shown that the hybridization of SSO with ACO using hybrid feature ranking method outperformed other algorithms and can be efficient in the detection of intrusive behaviour.

Author(s) Details

P. Amudha
Department of CSE, School of Engineering, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore, India

S. Sivakumari
Department of CSE, School of Engineering, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore, India.

View Book :-  http://bp.bookpi.org/index.php/bpi/catalog/book/200