Search

The emergence of fifth generation networks opens the doors for Internet of Things environment to spread widely. The number of connected devices to fifth generation networks is expected to increase to more than 1.7 billion users by 2025. Each year, millions of modern devices go online at the beginning of the school year and after the holidays, and you can even notice the publicity of Internet of Things devices swinging with the seasons. Nowadays, these devices are considered to be very important to our daily life. That is because they provide power to our homes, organize our work operations and let communications more suitable. As a result of the increasing number of connected devices to fifth generation networks, the necessity to protect these Internet of Things devices against different types of cyber-attacks is also increased. For this reason, many researchers proposed different protocols and schemes to achieve the security of the Internet of Things devices. In this article, we introduce a survey of some protocols proposed by researchers in different domains and make a comparative study between them in terms of their category, authentication process, evaluation methodology, advantages, target, development year and applications within Internet of Things environment. The objective of this survey is to provide researchers with rich information about these protocols and their uses within Internet of Things systems, whether they can be used for cloud radio access networks, Internet of Things general purposes, telecommunications systems, e-healthcare systems or drone delivery service systems. It can also assist them in choosing the proper protocol to be used according to the type of their Internet of Things system.

Social media sites are considered one of the most important sources of data in many fields, such as health, education, and politics. While surveys provide explicit answers to specific questions, posts in social media have the same answers implicitly occurring in the text. This research aims to develop a method for extracting implicit answers from large tweet collections, and to demonstrate this method for an important concern: the problem of heart attacks. The approach is to collect tweets containing &ldquo, heart attack&rdquo, and then select from those the ones with useful information. Informational tweets are those which express real heart attack issues, e.g., &ldquo, Yesterday morning, my grandfather had a heart attack while he was walking around the garden.&rdquo, On the other hand, there are non-informational tweets such as &ldquo, Dropped my iPhone for the first time and almost had a heart attack.&rdquo, The starting point was to manually classify around 7000 tweets as either informational (11%) or non-informational (89%), thus yielding a labeled dataset to use in devising a machine learning classifier that can be applied to our large collection of over 20 million tweets. Tweets were cleaned and converted to a vector representation, suitable to be fed into different machine-learning algorithms: Deep neural networks, support vector machine (SVM), J48 decision tree and na&iuml, ve Bayes. Our experimentation aimed to find the best algorithm to use to build a high-quality classifier. This involved splitting the labeled dataset, with 2/3 used to train the classifier and 1/3 used for evaluation besides cross-validation methods. The deep neural network (DNN) classifier obtained the highest accuracy (95.2%). In addition, it obtained the highest F1-scores with (73.6%) and (97.4%) for informational and non-informational classes, respectively.

The widespread presence of Gestational Diabetes Mellitus (GDM) is increasing in recent years. Women with GDM have to continuously monitor their blood glucose levels to reduce pregnancy risks. Existing practices in Australia comprise monitoring the blood glucose levels by taking readings at least three times a day and then manually sending the readings to the healthcare professionals. This research proposes the use of a Mobile application that can streamline communications with healthcare professionals and provide users with health and diet recommendations to regulate blood glucose levels. The application is intended to be used mainly by women with gestational diabetes. It works by providing healthcare professional access to the glucose’s readings via the cloud. The app also provides diet and physical activities’ recommendations. It also allows users to join community support groups online where they can have access to frequently asked questions and can raise their own queries, which can be answered by healthcare professionals as well.

Forecasting Sepsis length of stay is a challenge for hospitals worldwide. Although there are many attempts to improve sepsis length of stay prediction; however, there is still lack of baselining prediction metrics that can give better results for sepsis length of stay prediction in management hospital systems. This paper introduces a research architecture to predict and benchmark the Length of Stay (LOS) for Sepsis diagnoses from electronic medical records using the machine learning models. The architecture considered the time factor to identify the outperforming algorithms for Sepsis LOS prediction. This work contributes to the field of predictive modelling and information visualization for hospital management systems. Our results showed that the ensemble methods in particular the random forest (RF) outdo other classification models to predict the LOS for Sepsis from electronic medical records for Intensive Care Unit “ICU”-based hospitalizations.

The vulnerability is a flaw in the system’s implementation which may result in severe consequences. The existence of these flaws should be detected and managed. There are several types of research which provide different solutions to detect these flaws through static analysis of the original source codes. Static analysis process has many disadvantages, some of them are; slower than compilation and produce high false positive rate. In this project, we introduce a prediction technique using the output of one of the LLVM passes; “InstCount”. A classifier was built based on the output of this pass on 500 source codes written in C and C++ languages with 88% of accuracy. A comparison between our classifier and Clang static analyzer showed that the classifier super performed to predict the existence of memory leak and Null pointers. The experiment also showed that this classifier could be applied or integrated with static analysis tools for more efficient results.

Content Based Image Retrieval (CBIR) is an approach for retrieving similar images from an image database based on automatically-derived image features. The quality of a retrieval system depends on the features used to describe image content. In this paper, we propose an image clustering system that takes a database of images as input and clusters them using kmeans clustering algorithm taking into consideration color, texture and shape features. Experimental results show that the combination of the three features brings about higher values of accuracy and precision.