Search

Feature Selection (FS), a pre-processing step used in the majority of big data processing applications, aims to eliminate irrelevant and redundant features from the data. Its purpose is to select a final set of data characteristics that best represent the data as a whole. To achieve this, it explores every potential solution in order to identify the optimal subset. Meta-heuristic algorithms have been found to be particularly effective in solving FS problems, especially for high-dimensional datasets. This work adopts a recently developed meta-heuristic called the White Shark Optimizer (WSO) due to its simplicity and low computational overhead. However, WSO faces challenges in effectively balancing exploration and exploitation, particularly in complex multi-peak search problems. It tends to converge prematurely and get stuck in local optima, which can lead to poor search performance when dealing with FS problems. To overcome these issues, this paper presents three enhanced binary variants of WSO for well-known FS problems. These variants are as follows: (1) Binary Distribution-based WSO (BDWSO), where the algorithm refines the positions of white sharks by considering both the average and standard deviation of the current shark, the local best shark, and the global best shark. This strategy is designed to alleviate issues of premature convergence and stagnation during iterations , (2) Binary Sine Cosine WSO (BSCWSO), which uses sine and cosine adaptive functions for the social and cognitive components of the position update rule, and (3) Binary Hybrid Sine Cosine WSO (BHSCWSO), which employs sine and cosine acceleration factors to regulate local search and achieve convergence to the global optimal solution. Additionally, the population was initialized using the Opposition-Based Learning (OBL) mechanism, and the sine map was used to modify the inertia weight of WSO. These revised variants of WSO were established to have a better harmony between the exploration and exploitation facets. The proposed methods were extensively compared to the fundamental binary WSO and other well-known algorithms in the field. The experimental findings and comparisons demonstrate that the proposed methods outperform the conventional and most evaluated similar algorithms in terms of robustness and solution quality. In terms of classification accuracy, number of selected features, specificity, sensitivity, and fitness values, the proposed BHSCWSO optimizer performed better than all other proposed peer optimizers, including BSWO, BDWSO, and BSCWSO, in 11, 8, 13, 18, and 10 datasets, respectively. The proposed BHSCWSO optimizer showed performance levels of more than 90% in terms of accuracy, sensitivity, and specificity metric measures on 15, 14, and 11 of the 24 datasets deemed. [ABSTRACT FROM AUTHOR]

Wavelet neural networks (WNN) have been applied in many fields to solve regression as well as classification problems. After the advent of big data, as data gets generated at a brisk pace, it is imperative to analyze it as soon as it is generated owing to the fact that the nature of the data may change dramatically in short time intervals. This is necessitated by the fact that big data is all pervasive and throws computational challenges for data scientists. Therefore, in this paper, we built an efficient Scalable, Parallelized Wavelet Neural Network (SPWNN) which employs the parallel stochastic gradient algorithm (SGD) algorithm. SPWNN is designed and developed under both static and streaming environments in the horizontal parallelization framework. SPWNN is implemented by using Morlet and Gaussian functions as activation functions. This study is conducted on big datasets like gas sensor data which has more than 4 million samples and medical research data which has more than 10,000 features, which are high dimensional in nature. The experimental analysis indicates that in the static environment, SPWNN with Morlet activation function outperformed SPWNN with Gaussian on the classification datasets. However, in the case of regression, there is no clear trend was observed. In contrast, in the streaming environment i.e., Gaussian outperformed Morlet on the classification and Morlet outperformed Gaussian on the regression datasets. Overall, the proposed SPWNN architecture achieved a speedup of 1.22 - 1.78. [ABSTRACT FROM AUTHOR]

Distributed denial of service (DDoS) is an immense threat for Internet based-applications and their resources. It immediately floods the victim system by transmitting a large number of network packets, and due to this, the victim system resources become unavailable for legitimate users. Therefore, this attack is claimed to be a dangerous attack for Internet-based applications and their resources. Several security approaches have been proposed in the literature to protect Internet-based applications from this type of threat. However, the frequency and strength of DDoS attacks are increasing day-by-day. Further, most of the traditional and distributed processing frameworks-based DDoS attack detection systems analyzed network flows in offline batch processing. Hence, they failed to classify network flows in real-time. This paper proposes a novel Spark Streaming and Kafka-based distributed classification system, named by SSK-DDoS, for classifying different types of DDoS attacks and legitimate network flows. This classification approach is implemented using a distributed Spark MLlib machine learning algorithms on a Hadoop cluster and deployed on the Spark streaming platform to classify streams in real-time. The incoming streams consume by Kafka's topic to perform preprocessing tasks such as extracting and formulating features for classifying them into seven groups: Benign, DDoS-DNS, DDoS-LDAP, DDoS-MSSQL, DDoS-NetBIOS, DDoS-UDP, and DDoS-SYN. Further, the SSK-DDoS classification system stores formulated features with their predicted class into the HDFS that will help to retrain the distributed classification approach using a new set of samples. The proposed SSK-DDoS classification system has been validated using the recent CICDDoS2019 dataset. The results show that the proposed SSK-DDoS efficiently classified network flows into seven classes and stored formulated features with the predicted value of each incoming network flow into HDFS. [ABSTRACT FROM AUTHOR]

Despite the rapid improvement in dental health over the last few decades, a significant portion of our population continue seek dental care every year. Estimates show that 13% of adults seek dental care for dental infection or fatigue within four years. The Social and individual burden of this disease can be reduced by its early detection. However, the symptoms of teeth infection in the early stages are not clear, hence, it would be relatively difficult to predict teeth infections based solely on human skills and experience. Big Data (BD) technologies have a great potential in transforming dental care, as they have revolutionized other industries. In addition to reducing cost, they could save millions of lives and improve patient outcomes. This paper proposes a novel integrated prediction model that extracts hidden knowledge from radiographic datasets containing a large volume of dental X-ray images and utilizes this knowledge to predict dental infections. Initially, preprocessing techniques using morphological skeleton and mean approach is applied to eliminate noise and enhance the images. Next, Multi Scale Segmented Region (MSR) approach, Watershed Approach (WA), Sobel edge Detection (SD), Histogram based Segmentation (HS), Trainable Segmentation (TS), Dual Clustering (DC), and Fuzzy C-Means clustering (FCM) are examined for image segmentation and feature extraction. Among these methods, MSR was selected for feature extraction since it outperformed other methods in terms of accuracy, specificity, precision, recall and F1-score. Then, a set of neural network classifiers are trained to identify patterns in the extracted optimized features and predict dental infections. For this purpose, we have examined Bacterial Optimized Recurrent Neural Networks (BORNN), Deep Learning Neural Networks (DANN), Genetic Optimized Neural Networks (GONN) and Adaptive Neural Networks Algorithm (ADNN). BORNN have shown maximum accuracy and Roc value (98.1% and 0.92 respectively), and minimum error values (MSE = 0.189, MAE = 0.143). The output of the proposed integrated prediction model is fed into a dental robot who proceeds with the treatment process with high accuracy and minimum delay. The proposed prediction model was implemented using a big data analytics tool called Apache SAMOA and experimental results showed its correctness and effectiveness. [ABSTRACT FROM AUTHOR]

Nowadays XML based big bibliographic datasets are common in different domains which provide meta data about articles published in that domain. They have well defined tags which give details of the year, title, authors, abstract, keywords, the type of article, the venue of publishing the article and other such specific details about each article. A lot of statistics can be extracted from this dataset. Most of the time the tag pertaining to domain sub topic information associated with the article will be absent in the dataset as it is not an article attribute. Hence for such statistics articles must be mapped to its associated sub domain. This paper investigates this problem and proposes a fast approach to find trending articles and hot topics from XML based big bibliographic datasets. The proposed framework uses domain ontology to first classify articles into its sub topics. Fast detection of hot topics, trending keywords and articles is achieved using novel Map Reduce algorithms implemented on a hadoop distributed framework. Performance comparison demonstrates that it outperforms its non-Map Reduce counterpart in quickly sorting out the trending keywords and titles in a particular hot topic from XML based bibliographic dataset. [ABSTRACT FROM AUTHOR]

To cope with the rapid evolution of various attacks and the computer networks' increase, an intelligent intrusion detection system is considered as a promising emerging technique for the security of computer networks. Individual classification approaches have not provided complete protection. Indeed, it has been shown that none of them is efficient enough to provide good detection rates and reduce the false alarms rates. In previous works, a comparative study was conducted between the neuro-fuzzy and the genetic-fuzzy approaches. In this study, a hybrid approach is proposed based on the stacking scheme. This approach offers a solution to combine the two basic classifiers in order to take advantage of each one of them. The experimental results have shown the effectiveness of the proposed approach in terms of maximizing the detection rates and reducing the false alarm rates. [ABSTRACT FROM AUTHOR]

Privacy has become an important concern while publishing micro data about a population. The emerging area called privacy preserving data mining (PPDM) focus on individual privacy without compromising data mining results. An adversarial exploitation of published data poses a risk of information disclosure about individuals. On the other hand, imposing privacy constraints on the data results in substantial information loss and compromises the legitimate data analysis. Motivated by the increasing growth of PPDM algorithms, we first investigate the privacy implications and the crosscutting issues between privacy versus utility of data. We present a privacy model that embeds the anonymization procedure in to a learning algorithm and this has mitigated the additional overheads imposed on data mining tasks. Our primary concern about PPDM is that the utility of data should not be compromised by the transformation applied. Different data mining classification workloads are analyzed with the proposed anonymization procedure for any side effects incurred. It is shown empirically that classification accuracy obtained for most of the datasets outperforms the results obtained with original dataset. [ABSTRACT FROM AUTHOR]