Online examinations are of great importance to education. It has become a powerful tool for evaluating students’ knowledge and learning. Adopting modern technology that saves time and ensures security. The researcher developed a Test Bank Management System that can store test items in any subjects. The system is capable of conducting item analysis using the Rasch model scale. Items that undergo analysis based on Rasch scale helped faculty by quantifying each item as “good”, “rejected”, or “revised”. For securing items in the test bank, Data Encryption Standard (DES) algorithm was successfully applied thus ensuring the safety and reliability of the questions in the test bank. Only items that are ready for deployment to the student’s computer during the examinations will be decrypted. In conclusion, the system passed the evaluation process and eliminates redundancy of manual work.

Software inspection is useful to detect the defects in any stage software development methodology especially in early stages. Inspection of software defects can improve the software product quality by decreasing rework cost and time from documents, code, and other deliverables. The objective of this study is to identify existing software inspection techniques which help practitioners and software engineers to improve the software quality and to compare them according to some quality attributes. Rather than proposing new techniques we focus on synthesize the existing approaches. A comparison of some approaches is conducted and analyzed which approach is more feasible for the detection of defects. The results of this study show that there are many formal and informal techniques available in literature related software inspection, it is difficult to say well to one of them, but our analysis focused on finding such techniques which cover maximum quality factors in an inspection. Software inspection is an umbrella activity of whole software development life cycle and we found different approaches and frameworks in software inspection which can full fill our desired parameters to improve software quality.

There is a rapid increase in the demand of healthcare resources in Nigeria mainly healthcare personnel and healthcare facilities. This is due to the prevalence of chronic diseases such as endemic malaria, arthritis, hypertension, diabetes; an upsurge in the rate of avoidable deaths as well as an exponential increase in the population. However, healthcare resources in Nigeria are insufficient. For instance, in recent times, the ratio of doctors to the inhabitants is 1: 4,857. Moreover, more than 60% of the Nigerian populace resides in rural areas where there are extreme shortages of healthcare practitioners and healthcare facilities due to geographical isolation and lack of opportunities. Hence, the low ratio of healthcare practitioners to patients causes a heavy workload on the healthcare practitioners. This however leads to medical errors as healthcare providers work under intense pressure to attend to the medical needs of their patients. This in turns leads to considerable loss of lives. In order to ameliorate this situation, this paper proposes an ontology based framework that will enable healthcare providers in Nigeria to continuously monitor their patients’ health remotely outside the settings of the hospital. This will reduce the workload of the healthcare providers, assist them in decision making process as well as reduce the long waiting hours of the patients within the hospital environment. This framework is also designed to tackle the challenge of semantic interoperability facing healthcare systems around the globe.

We propose a routing algorithm based on AODV approach. We have modified AODV algorithm to obtain better performance. At the same time, we have tried to maintain the features of AODV algorithm by maintaining a backward compatibility in our proposed algorithm. Our algorithm tries to reduce RREQ packets which are broadcasted in original AODV to find routing paths. For this purpose our algorithm uses a location aware approach to find distance to sink node. It also makes use of busyness of nodes while selecting nodes to participate in route discovery mechanism. Also this scaling factor and busyness threshold can be made fixed for each node in the network depending on size and characteristics of the network.

Feature selection is one of the issues that have been raised in the discussion of machine learning and statistical identification model. We have provided definitions for feature selection and definitions needed to understand this issue, we check. Then, different methods for this problem were based on the type of product, as well as how to evaluate candidate subsets of features, we classify the following categories. As in previous studies may not have understood that different methods of assessment data into consideration, We propose a new approach for assessing similarity of data to understand the relationship between diversity and stability of the data is selected. After review and meta-heuristic algorithms to implement the algorithm found that the cluster has better performance compared with other algorithms for feature selection sustained.

Wireless sensor networks (WSNs) are large scale integration in large topology deployed with thousands sensor nodes. Nodes sizes are very small, with low cost, low weight, and limited battery, primary storage, processing power. Sensor nodes have wireless communication capabilities with sensor to monitor physical or environmental conditions. This paper study and evaluate performance for localization and target tracking application with proposed hierarchical localization tracking scheme based on hierarchical binary tree structure. The target detected information is stored at multiple sensor nodes (e.g. node, parent node and grandparent node) which deployed using complete binary tree structure to improve fault tolerance. This drastically reduces number of messaging in the network. Performance of proposed scheme and some existing routing scheme is evaluated using NS2. Simulation result proof increased in network lifetime by 25%, target detection probability by 25%, and reduces error rate by 20%, increased energy efficiency by 20%, fault tolerance, and routing efficiency.

Nowadays, Web services are the leading solution for solving the problem of information systems’ integration. Web services are based on the service oriented architecture (SOA). Preserving privacy of web services is one of the main challenges during their interaction. Therefore, minimizing the number of the disclosed credentials that are required for accessing the web services resources during the interaction is a desirable behavior. Credentials generalization and substitution and credentials encryption can be used for privacy preserving. To the best of our knowledge, in the current privacy preserving approaches for web services, there is no technique that uses the negotiation for credential generalization and substitution between the consumer and the provider in conjunction with the credentials encryption using certificate authority as a third party. In this paper, a proposed approach for web services privacy preserving is proposed. This approach is based on the negotiation, encryption, and certificate authorities as third parties. The proposed approach is implemented and tested. The results show that the number of disclosed sensitive credential is the minimum number of credentials that can be disclosed to guarantee the accessing of the service.

This paper proposes an intuitive approach to facial age classification on child faces – a recursive multi-class binary classification tree – using the texture information obtained from facial images. The face area is divided into small regions from which Local Binary Pattern (LBP) histograms were extracted and concatenated into a single vector efficiently representing a facial image. The classification is based on training a set of binary classifiers using Support Vector Machines (SVMs). Each classifier estimates whether the facial image belongs to a specified age range or not until the last level of the tree is reached where the age is finally determined. Our classification approach also includes an overlapping function that resolves overlaps and conflicts in the outputs of two mutually-exclusive classifiers at each level of the classification tree. Our proposed approach was experimented on a publicly available dataset (FG-NET) and our locally obtained dataset (FAGE) and the results obtained are at par with those of existing works.