International Journal of Computing and Digital Systems 

About: International Journal of Computing and Digital Systems is an academic journal published by Deanship of Scientific Research. The journal publishes majorly in the area(s): Computer science & Engineering. It has an ISSN identifier of 2210-142X. It is also open access. Over the lifetime, 752 publications have been published receiving 1923 citations. The journal is also known as: IJCDS.

Hierarchical Reinforcement Learning: A Survey

Abstract: Reinforcement Learning (RL) has been an interesting research area in Machine Learning and AI. Hierarchical Reinforcement Learning (HRL) that decomposes the RL problem into sub-problems where solving each of which will be more powerful than solving the entire problem will be our concern in this paper. A review of the state-of-the-art of HRL has been investigated. Different HRL-based domains have been highlighted. Different problems in such different domains along with some proposed solutions have been addressed. It has been observed that HRL has not yet been surveyed in the current existing research; the reason that motivated us to work on this paper. Concluding remarks are presented. Some ideas have been emerged during the work on this research and have been proposed for pursuing a future research.

Network Based Intrusion Detection Using the UNSW-NB15 Dataset

Abstract: In this work, we apply a two stage anomaly-based network intrusion detection process using the UNSW-NB15 dataset. We use Recursive Feature Elimination and Random Forests among other techniques to select the best dataset features for the purpose of machine learning; then we perform a binary classification in order to identify intrusive traffic from normal one, using a number of data mining techniques, including Logistic Regression, Gradient Boost Machine, and Support Vector Machine. Results of this first stage classification show that the use of Support Vector Machine reports the highest accuracy (82.11%). We then feed the output of Support Vector Machine to a range of multinomial classifiers in order to improve the accuracy of predicting the type of attacks. Specifically, we evaluate the performance of Decision Trees (C5.0), Naïve Bayes and multinomial Support Vector Machine. Applying C5.0 yielded the highest accuracy (74%) and F1 score (86%), and the two-stage hybrid classification improved the accuracy of results by up to 12% (achieving a multi-classification accuracy of 86.04%). Finally, with the support of our results, we present constructive criticism of the UNSW-NB15 dataset.

Analysis of Software Development Methodologies

Abstract: The researcher focuses on the analysis of most common diverse methodologies of software development to choose the best one on the basis of different factors such as project type, size, development environment, and available resources. Software projects provided are positive and negative impacts and provide the stages of software development methodology. Subsequently, the author gives brief details about the common stages of software development in this paper. These stages are mostly used in every software development methodologies (SDMs). The main motive of this research is to provide the details of figures of steps and stages about currently available most common twenty-one (21) SDMs. Software projects are on the functions or stages of the methodology, the project owner's feedbacks in each methodology and suitability of methodology on the small, medium and large size of projects. The Result conducted based on an analysis between them by applying different strategies, development environments, and common practices and based on available resources, which can easily be understood to choose the best methodology, which can be feasible for Small, and Medium Enterprises (SMEs).

Data Aggregation Mechanisms in the Internet of Things: A Study, Qualitative and Quantitative Analysis

Abstract: Internet of Things (IoT), a paradigm added to the ever-growing technological arena in recent times acts like a bridge between the things in the physical world and their representation within the digital world. The basic “things” in the IoT are sensor devices, which gather as well as monitor all types of data on physical machines and human social life. IoT enables data sending and receiving for each “thing” through the communication network. The purpose of Data Aggregation is to decrease the number of communications/transmissions among the objects/things in the Internet of Things framework. The effectiveness of the data aggregation technique employed is a key factor in the success of IoT systems in terms of data freshness and efficiency. Different data aggregation techniques have been proposed in the recent past, which include – Tree-Based, Cluster-Based and Centralized data aggregation techniques. The paper aims at a detailed study and analysis of data aggregation schemes employed in the Internet of Things in terms of working and time complexity. Lowest Common Ancestor (LCA) aided Tree-Based Data Aggregation algorithm is designed. In addition, the Cluster-Based data aggregation algorithm incorporated with the β-dominating set and Centralized Data Aggregation algorithm incorporated with the SUM() aggregation function are proposed. The algorithms are supported by wellformed flowcharts describing the flow and working of the data aggregation mechanisms designed. The results are obtained on a system consisting of 60 nodes with all the three aggregation algorithms being evaluated against each other. The centralized data aggregation algorithm is better when the number of nodes in the network is lesser. However, as the number of nodes increases, the cluster-based and tree-based algorithms produce better results as compared to the centralized data aggregation algorithm.

Certain Study on MPPT Algorithms to track the Global MPP under Partial Shading on Solar PV Module/Array

Abstract: The maximum power point tracking (MPPT) algorithms are used in PV power generation systems to handle the effect due to the partially shaded conditions. This paper confers the algorithms, modeling techniques, and control topologies of photovoltaic (PV) array systems are explored. The problems with conventional MPPT algorithms can be solved by applying various modern optimization algorithms that extract the maximum power from the solar panels. Various reliable techniques are discussed to identify the maximum power point globally. However, each method has advantages and limitations and, this paper presents reviews and findings from the existing optimized methods. The optimized algorithms presented in the various literatures are studied and analyzed. The hybrid MPPT algorithms are also discussed to present its effectiveness in tracking the maximum power point (MPP). The challenges in selecting a proper algorithm for partially shaded PV array are deliberated. Finally, the comparison between the standalone algorithms and the hybrid algorithms are presented for future research.