Nowadays, many Certificate Authorities (CAs) that issue certificates may or may not be trusted because not all CAs are reliable and trustworthy. University laboratories and computers of its people (students, lecturer and staff) are thus susceptible to the risk resulting from this mistrust. This study proposes a university owned notary server, which will be managed by the university, to solve the problem using a Certificate Trust List (CTL). Simply put, when students and others use the Web, the notary server checks the certificate to see whether a conflict exists and verifies the signatures and key references in the certificate. If all the information is correct, the notary server sends a response of approval to the client to accept the certificate. Our system enhanced the security in a university by trusting only genuine CAs. Our proposed server is better than regular notary servers because it uses existing infrastructure and online connections, and it does not introduce any overheads or special configurations to the client’s Web browser. Compared with a well known notary server runs over the existing infrastructure, our proposed notary server is 10.8 seconds faster in terms of dealing with untrusted CA and 2.3 seconds faster in terms of dealing with mismatched address of the Web sites.

This paper describes methodology and performance of an experimental research on filtering of web search results. Filtering was performed on the basis of predicted relevance of search results derived from users’ implicit feedback. The feedback was obtained from users’ web browsers and consisted of a set of browsing behavioral metrics, including reading time, clicks on links, mouse pointer and wheel movement patterns, bookmarking, sharing, copying, and whether the search was continued after the page was closed. A multi-layer neural network used to infer from the behaviors how much the user was interested in each filtered document. Neural network, therefore, performed deep learning without human supervision. Predicted relevance measure was compared to the explicit feedback. Obtained results of 89% correct relevance rating prediction suggest that selected set of metrics was successful in terms of correctly predict how relevant the web page was for the user involved in the study. More research is recommended for further advances of information filtering methods.

No abstract

Congestion control and reduction is paramount in enhancing the performance and speed of communication networks. Despite its efficiency in improving the speed and performance of communication networks, it is imperative to mention that the technique exhibit several issues, such as scheduling algorithms. This paper presents a comparison of different results associated with different scheduling algorithms. Multisource was used to simulate and test different scheduling algorithms, including First In First Out (FIFO), Random Packet Drop, and Last In First Out (LIFO). The comparison was made by analyzing different results, including the average delay vs arrival rate, average buffer utilization vs arrival rate, and packet loss ration vs arrival rate. According to the results, it is evident that the average buffer utilization and average packet delays increased as the rate of data transmitted through the network increased. There was no significant change noticed in the average buffer utilization in all the three algorithms. However, the packet loss ratio was high in Random Packet Loss algorithm than in both FIFO and LIFO for slower arrival rates below 1000. Similarly, FIFO exhibited a significantly high average packet delay than both LIFO and Random Packet Drop algorithms.