Volume 13, Issue 4, September 2022

Security and Privacy Issues, its Solutions using Block- Chain and Self-healing Technology for Fog Computing IoT Devices
Pages: 1-10 (10) | [Full Text] PDF (634K)
Azhar Abbas, Rizwan Ali Mughal, M. Junaid Arshad
Department of Computer Science, Uuniversity of Engineering and Technology, Lahore-Pakistan

Abstract -
Fog node are applied near to end-users Internet of Things devices, which resultantly limit the low latency rate, location - primarily based totally services, and geographic distribution unfounded capabilities of IOT (Internet of Things) applications. Furthermore, fog computing reduces data offload to the cloud, which reduces the reaction time. Along with these number of benefits, fog computing faces various challenges in term of securities and privacy. The term "Fog Computing" is developed by Cisco to express the extension of Cloud Computing to the system edge. Along with the tremendous developments in IoT, presently the Cloud computing framework have a number of drawbacks, including a lack of flexibility support, area awareness, geo dissemination, delay, and digital hazards. Cloud computing provides data, stockpiling and figure with administrations application to the end-user assistance the fog computing such as offer the kinds of existence like as process, application data and the capacity to the end user. This paper gives a complete understanding of the Fog Internet of Thing devices, security and privacy challenges. We analysis the security and privacy challenges constructed by the Fog computing IoT Devices and their solutions with the help of Block-chain technology technique and self-healing.
 Index Terms - Fog Computing, IoT Devices, Block Chain Technology and Challenges
Performance Evaluation of Classification Algorithms on Academic Performance of Postgraduate Students
Pages: 11-15 (5) | [Full Text] PDF (269K)
Okunlola, O. A. and Ojo, A. K.
Department of Computer Science, University of Ibadan, Nigeria

Abstract -
Educational data mining has contributed to enhancing student academic performance by way of enabling stakeholders in academic institutions to have a pre-knowledge of the risks and dangers ahead and how to mitigate them. Prediction algorithms perform differently on dataset, and so, the need to develop models using different prediction algorithms and evaluating the result of such predictions is very important in order to be sure the best algorithm for a particular dataset is used. This work employed four classifiers: K-Nearest-Neighbour, Neural Network, Naïve Bayes and Decision Tree to model and, evaluated their models to know the performance of each on the target dataset. Their results were evaluated based on the various performance metrics. The results showed that Decision Tree had the highest accuracy on the dataset with test accuracy of 48.25% and therefore is the most suitable out of the four classifiers for performing prediction modelling on the dataset. Naïve Bayes is the second-best prediction model that can be used for predicting academic performance with an accuracy of 36.25%., followed by Neural Network with accuracy of 32.5 % and then K-Nearest Neighbour with accuracy of 32.5% but with lower precision, recall and area under Receiver Operating Characteristic curve.
 Index Terms - Decision Tree, Educational Data Mining, K-Nearest Neighbour, Neural Network and Naïve Bayes
Opinion Mining Using Microblogs of Pakistan Incidents
Pages: 16-19 (4) | [Full Text] PDF (391K)
Muhammad Majid Hussain
CS Department, UET, Lahore

Abstract -
Social media platforms have inevitably grown with the rapid growth of the internet and now play a significant part in almost all of our lives. Twitter is a prominent microblogging platform that has grown in popularity around the world. Nowadays, it's trendy to share all thoughts and feelings one has on such social media platforms. In this research, opinions of the public are captured from microblogs using machine learning techniques on the incidents (Sialkot Incident, Murree Incident, other protest, Johar Town Blast, Anarkali Blast etc) that happened in Pakistan. This research will also include emojis, emoticons, and slang words (TBH - To be honest, OMG - Oh my God) for the opinion of a microblog related to an incident in Pakistan. Tweets data is collected by using 2wttr tool from the v2 Twitter API for academic research through the bearer token of the Twitter developer account. After dataset collection different preprocessing techniques are applied to prepare data for machine learning models and text2emotion library was applied for labeling data and then six machine learning models (Logistic Regression, Naïve Bayes, SVM, Decision Tree Classifier, Random Forest Classifier, KNN) were applied to every incident dataset. Results show that overall SVM performs well than other models and gives highest average accuracy of 95.8%.
 Index Terms - Twitter, Pakistan Incidents, Text2emotion, SVM and Blogs
Detection of Acute Liver Failure from Demographic Data Using Data Mining Algorithms
Pages: 20-28 (8) | [Full Text] PDF (446K)
Ali Akbar, M. Junaid Arshad, Mohsin Yasin
Department of Computer Science and Engineering, University of Engineering and Technology, Lahore, Pakistan

Abstract -
Acute Liver Failure is loss of liver capability that happens quickly — in days or weeks — a no previous liver for the most part in an individual sickness. Its most regularly brought about by a hepatitis infection or medications, like acetaminophen. Intense liver disappointment is more uncommon than constant liver disappointment, which grows all the more leisurely. ALF, also called fulminant hepatic disappointment, can cause serious entanglements, remembering unnecessary draining and expanding tension for the mind. A health related crisis requires hospitalization. Contingent upon the reason, ALF can at times be switched with therapy. Much of the time, however, a liver transfer might be the main fix. Acute liver failure is a complex multisystem sickness that advances rapidly after a disastrous affront to the liver prompting the improvement of encephalopathy. The basic aetiology and the speed of movement emphatically impact the clinical course. The commonest causes are paracetamol, eccentric medication responses, hepatitis B, and seronegative hepatitis. The ideal consideration is multidisciplinary and up to half of the cases get liver transfers, with endurance rates around 75%-90%. Counterfeit liver help gadgets stay dubious in viability in intense liver disappointment.
 Index Terms - Acute Liver Failure, Association Rule Mining, Machine Learning Algorithms and Data Mining