Spam Mail Filtering Using Data Mining Approach

This chapter presents an overview of spam email as a serious problem in our internet world and creates a spam filter that reduces the previous weaknesses and provides better identification accuracy with less complexity. Since J48 decision tree is a widely used classification technique due to its simple structure, higher classification accuracy, and lower time complexity, it is used as a spam mail classifier here. Now, with lower complexity, it becomes difficult to get higher accuracy in the case of large number of records. In order to overcome this problem, particle swarm optimization is used here to optimize the spam base dataset, thus optimizing the decision tree model as well as reducing the time complexity. Once the records have been standardized, the decision tree is again used to check the accuracy of the classification. The chapter presents a study on various spam-related issues, various filters used, related work, and potential spam-filtering scope.

A Deep Learning Approach for Detection of Application Layer Attacks in Internet

The recent technological transformation in application deployment, with the enriched availability of applications, induces the attackers to shift the target of the attack to the services provided by the application layer. Application layer DoS or DDoS attacks are launched only after establishing the connection to the server. They are stealthier than network or transport layer attacks. The existing defence mechanisms are unproductive in detecting application layer DoS or DDoS attacks. Hence, this chapter proposes a novel deep learning classification model using an autoencoder to detect application layer DDoS attacks by measuring the deviations in the incoming network traffic. The experimental results show that the proposed deep autoencoder model detects application layer attacks in HTTP traffic more proficiently than existing machine learning models.

Paradigms of Machine Learning and Data Analytics

This chapter conducts a critical review on ML and deep learning tools and techniques in the field of heart disease related to heart disease complexity, prediction, and diagnosis. Only specific papers are selected for the study to extract useful information, which stimulated a new hypothesis to understand further investigation of the heart disease patient.

A Cloud Trusting Mechanism Based on Resource Ranking

The cloud-based computing paradigm helps organizations grow exponentially through means of employing an efficient resource management under the budgetary constraints. As an emerging field, cloud computing has a concept of amalgamation of database techniques, programming, network, and internet. The revolutionary advantages over conventional data computing, storage, and retrieval infrastructures result in an increase in the number of organizational services. Cloud services are feasible in all aspects such as cost, operation, infrastructure (software and hardware) and processing. The efficient resource management with cloud computing has great importance of higher scalability, significant energy saving, and cost reduction. Trustworthiness of the provider significantly influences the possible cloud user in his selection of cloud services. This chapter proposes a cloud service selection model (CSSM) for analyzing any cloud service in detail with multidimensional perspectives.

Improving Scalability in Replicated DRTDBS

The demand for scalability in replicated distributed real-time database systems (RDRTDBS) is still explorative and, despite an increase in real-time applications, many challenges and issues remain in designing a more scalable system. The objective is to improve the scalability of the system during system scale up with new replica sites. Existing research has been mainly conducted in maintaining replica consistency between different replicas via replication protocol. However, very little research has been conducted towards improving scalability and maintaining mutual consistency and timeliness. Consequently, the ultimate aim of this chapter is to improve scalability in RDRTDBS such that performance of the system does not degrade even though new replica sites are added.

Localization of Data Sets in Distributed Database Systems Using Slope-Based Vertical Fragmentation

This chapter explains an algorithm that can perform vertical partitioning of database tables dynamically on distributed database systems. After vertical partitioning, a new algorithm is developed to allocate that fragments to the proper sites. To accomplish this, three major tasks are performed in this chapter. The first task is to develop a partitioning algorithm, which can partition the relation in such a way that it would perform better than most of the existing algorithms. The second task is to allocate the fragments to the appropriate sites where allocating the fragments will incur low communication cost with respect to other sites. The third task is to monitor the change in frequency of queries at different sites as well as same site. If the change in frequency of queries at different sites as well as the same site exceeds the threshold, the re-partitioning and re-allocation are performed.

Causes, Effects, and Consequences of Priority Inversion in Transaction Processing

The problem of priority inversion occurs when a high priority task is required to wait for completion of some other task with low priority as a result of conflict in accessing the shared system resource(s). This problem is discussed by many researchers covering a wide range of research areas. Some of the key research areas are real-time operating systems, real-time systems, real-time databases, and distributed real-time databases. Irrespective of the application area, however, the problem lies with the fact that priority inversion can only be controlled with no method available to eliminate it entirely. In this chapter, the priority inversion-related scheduling issues and research efforts in this direction are discussed. Different approaches and their effectiveness to resolve this problem are analytically compared. Finally, major research accomplishments to date have been summarized and several unanswered research questions have also been listed.

Vehicular Ad Hoc Networks (VANETs)

Over the past few years, there has been significant research interest in field of vehicular ad hoc networks (VANETs). Wireless communication over VANETs supports vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I) communication. Such innovation in wireless communication has improved our daily lives through road safety, comfort driving, traffic efficiency. As special version of MANETs, VANETs bring several new challenges including routing and security challenges in data communication due to characteristics of high mobility, dynamic topology. Therefore, academia and the auto mobile industry are taking interest in several ongoing research projects to establish VANETs. The work presented here focuses on communication in VANETs with their routing and security challenges along with major application of VANETs in several areas.

Term Ordering-Based Query Expansion Technique for Hindi-English CLIR System

The quality of retrieval documents in CLIR is often poor compared to IR system due to (1) query mismatching, (2) multiple representations of query terms, and (3) un-translated query terms. The inappropriate translation may lead to poor quality of results. Hence, automated query translation is performed using the back-translation approach for improvement of query translation. This chapter mainly focuses on query expansion (Q.E) and proposes an algorithm to address the drift query issue for Hindi-English CLIR. The system uses FIRE datasets and a set of 50 queries of Hindi language for evaluation. The purpose of a term ordering-based algorithm is to resolve the drift query issue in Q.E. The result shows that the relevancy of Hindi-English CLIR is improved by performing Q.E. using a term ordering-based algorithm. The outcome achieved 60.18% accuracy of results where Q.E has been performed using a term ordering based algorithm, whereas the result of Q.E without a term ordering-based algorithm stands at 57.46%.

Performance Enhancement of Outlier Removal Using Extreme Value Analysis-Based Mahalonobis Distance

Outlier detection is a part of data analytics that helps users to find discrepancies in working machines by applying outlier detection algorithm on the captured data for every fixed interval. An outlier is a data point that exhibits different properties from other points due to some external or internal forces. These outliers can be detected by clustering the data points. To detect outliers, optimal clustering of data points is important. The problem that arises quite frequently in statistics is identification of groups or clusters of data within a population or sample. The most widely used procedure to identify clusters in a set of observations is k-means using Euclidean distance. Euclidean distance is not so efficient for finding anomaly in multivariate space. This chapter uses k-means algorithm with Mahalanobis distance metric to capture the variance structure of the clusters followed by the application of extreme value analysis (EVA) algorithm to detect the outliers for detecting rare items, events, or observations that raise suspicions from the majority of the data.