Missing data is common problem faced by researchers and data scientists. Therefore, it is required to handle them appropriately in order to get better and accurate results of data analysis. Objective of this research paper is to provide better understanding of data missingness mechanism, data imputation methods, and to assess performance of the widely used data imputation methods for numeric dataset. It will help practitioners and data scientists to select appropriate method of data imputation for numeric dataset while performing data mining task. In this paper, we comprehensively compare seven data imputation methods namely mean imputation, median imputation, kNN imputation, predictive mean matching, Bayesian Linear Regression (norm), Linear Regression, non-Bayesian (norm.nob), and random sample. We have used five different numeric datasets obtained from UCI machine learning repository for analyzing and comparing performance of the data imputation methods. Performance of the data imputation methods is assessed using Normalized Root Mean Square Error (RMSE) method. The results of analysis show that kNN imputation method outperforms the other methods. It has also been found that performance of the data imputation method is independent of the dataset and percentage of missing values in the dataset.

Many real-world classification problems such as fraud detection, intrusion detection, churn prediction, and anomaly detection suffer from the problem of imbalanced datasets. Therefore, in all such classification tasks, we need to balance the imbalanced datasets before building classifiers for prediction purposes. Several data-balancing techniques (DBT) have been discussed in the literature to address this issue. However, not much work is conducted to assess the performance of DBT. Therefore, in this research paper we empirically assess the performance of the data-preprocessing-level data-balancing techniques, namely: Under Sampling (OS), Over Sampling (OS), Hybrid Sampling (HS), Random Over Sampling Examples (ROSE), Synthetic Minority Over Sampling (SMOTE), and Clustering-Based Under Sampling (CBUS) techniques. We have used six different classifiers and twenty-five different datasets, that have varying levels of imbalance ratio (IR), to assess the performance of DBT. The experimental results indicate that DBT helps to improve the performance of the classifiers. However, no significant difference was observed in the performance of the US, OS, HS, SMOTE, and CBUS. It was also observed that performance of DBT was not consistent across varying levels of IR in the dataset and different classifiers.

Industrial Internet of Things (IIoT) is attempting to integrate the real world into the digital world through smart devices, information technology, and Internet. IIoT is connecting enormous devices in industry 4.0/5.0 which may be heterogeneous in nature. With the evolution of IoT, diverse technologies have been employed to deliver the quality of services to the end users in a seamless manner. Cloud computing has considerably boosted the growth of IIoT by serving the computational and data storage needs of IIoT- and blockchain-based applications in industry 4.0/5.0. However, cloud is providing services to IIoT users, but still, there is a need to improve the latency rate of delivery of services, the transmission rate of end-to-end delivery, and overall throughput of the network channels in industry 4.0 and blockchain-based distributed systems. The cloud servers those are located at remote locations are not capable to offer the quality of services to users who require real-time responses, minimum network latency, and optimum throughput. The advancements in Edge computing are making Edge-Cloud more suitable for end users in blockchain-based transactions and industry 4.0 to serve the requirements of IIoT-based applications. This paper aims at providing the resources of Edge-Cloud to the end users by proposing a soft-computing technique for selecting the most suitable resources from the pool of available resources at Edge-Cloud. This paper is proposing a multicriteria statistical approach for resource selection to exploit the benefits of Edge-Cloud and to suffice the needs of IIoT and blockchain-based applications in industry 4.0. The results obtained from the proposed research assist in enhancing the service providing rate, minimizing the delay in transmission, and optimizing the throughput of Edge-Cloud servers.

Multi criteria decision making (MCDM) methods help decision makers to make preference decision over the available alternatives. Evaluation and selection of the software packages is multi criteria decision making problem. Analytical hierarchy process (AHP) and weighted scoring method (WSM) have widely been used for evaluation and selection of the software packages. Hybrid knowledge based system (HKBS) approach for evaluation and selection of the software packages has been proposed recently. Therefore, there is need to compare HKBS, AHP and WSM. This paper studies and compares these approaches by applying for evaluation and selection of the software components. The comparison shows that HKBS approach for evaluation and selection of the software packages is comparatively better than AHP and WSM with regard to (i) computational efficiency (ii) flexibility in problem solving (iii) knowledge reuse and (iv) consistency and presentation of the evaluation results.