Diabetes is a common disease affects one person out of every 11 people. In 2021, 537 million people in the world were suffering from diabetes, and this number is expected to double in the coming years. Diabetes greatly affects the eyes, such as diabetic retinopathy.It is an eye disease that can affect diabetic patients. When this happens, the blood vessels in the retina are damaged due to high blood sugar levels, which can lead to blindness. So early detection can reduce the risk of blindness. Previously, diagnosing diabetic retinopathy was very difficult and time consuming for ophthalmologists. And now, deep learning has made it easier for ophthalmologists to detect and stages diabetic retinopathy.So early detection can reduce the risk of blindness. Previously, diagnosing diabetic retinopathy was very difficult and time consuming for ophthalmologists. And now, deep learning has made it easier for ophthalmologists to detect and stages diabetic retinopathy.In this project, we focus on using deep neural network and an EfficientNetB3 model to detect and classify diabetic retinopathy stages.
23/02/2023
Customers are relying on reviews to make a purchase decision. Fake reviews are untruthful reviews written to mislead customers and can cause financial loss to Electronic Commerce (EC) businesses. Despite the importance of fake review detection, few Arabic studies exist due to the lack of suitable datasets. Therefore, we introduce the first golden-standard Arabic dataset for fake reviews detection with review content in Saudi and Modern Standard Arabic(MSA) dialects for multi-domains, restaurants, hotels, and products. We integrate the FastText word embedding technique with four Deep Neural Networks (DNN) models to study our dataset in single-domain, multi-domain, and cross-domain experiments, which showed a solid performance. Our experimental results showed CNN hybrid models are superior. The highest achieved ac- curacy was 93.71% by CNN+Bi-LSTM in the restaurant domain with a recall of 97.18%. For cross-domain, we tested the impact of zero-learning and Transfer Learning (TL) by introduc- ing ResModel, a CNN+Bi-LSTM model built on the restaurant domain dataset, the ResModel model is used further to build RHModel on the hotel domain dataset and RPModel on the prod- uct domain dataset. We showed the effect of the TL approach in improving model performance in our limited dataset, as RHModel outperformed the four DNN models in the hotel domain with an accuracy of 98.39% and a precision of 100%, we conclude that the limited size of the dataset does not result in poor performance, as the quality and relevance of dataset instances play a major role in the model performance, and to the best of our knowledge, this is the first Arabic paper that utilized a TL approach in fake reviews detection.
23/02/2023
Using a smartphone or web app is an integral part of our automated daily practices, and in order to better understand why technology creates a good relationship with humans we will have to understand the meaning of user experience(UX). The term of UX that appear in early nineties, which is studies the relationships between human and technology. The concept of UX is based on the study and understanding of the user, and through the stages of UX design, it is possible to obtain high-quality products that delight the customer and achieve their goals. In this project we will use UX to evaluate and analyze The Administrative Communications System (ACS)(Enjaz)of Qassim University, which controls internal and external, incoming and outcoming electronic transactions, monitors them,and archives them electronically. In our pa- per, we discussed the concept of UX and its elements, as well as measurements and methods for measuring them. To assess the UX of the Enjaz system, a questionnaire and interviews was chosen. After studying and analyzing the system and identifying the points that cause problems for users, a model will be created that includes solutions to those problems.
23/02/2023
interior design, by its very nature, involves the act of creating. Creative interior design makes our lives comfortable, enjoyable, rich, safe, healthy, efficient, and orderly. Buying prod- ucts for interior design always has a problem in that the products purchased may not satisfy the customers because they cannot put them in place before buying. This project aims to provide a platform using augmented reality technology for design and decoration that will help clients visualize how furniture will look and fit (at a large scale) in their home or business, to reduce time and cost. This project contains a new interior design platform based on the client’s case and innovative features, it creates a virtual room according to the case chosen by the user. We have relied on the filter and resetFulAPI to make it easier and as helpful as possible. The waterfall is used for system development.
22/02/2023
Academic advising plays a pivotal part in the achievement of educational institution pur-poses. It’s an essential element in students’ academic problems and maximizing their satisfac- tion. Given the importance of advising, students are assigned to an academic advisor throughout their studies after entering college. Universities around the world have always tried to amelio- rate academic advising to enhance the student’s experience. In fact, technology has the power to ameliorate the advising process and grease its corresponding tasks and this has historically taken different forms. Accordingly, we built the proposed system to facilitate the learning pro-cess to the student during his university journey, learn about the description of the courses and their nature. The proposed system also aims to accommodate the basic operations required by the academic advisor to communicate with students. These include scheduling an appointment, comparing study plans, view suggestions for places for summer training, view the most important differences between CS and IT departments, create study schedules and offering tips and instructions based on student record and the (GPA)
22/02/2023
With the increase in the number of elderly people around the world, especially in the King- dom of Saudi Arabia, we ask ourselves how to support this critical part of the community. Therefore, we focus on how to utilize the technology advancement to take care of elderly people by providing them with protection, safety, and independent life. This is achieved by designing a wearable device (Auxiliary bracelet) connected to the internet for elderly people that fit on their wrists. The application can monitor the behavior of the elderly through sensors to measure vital signs and in the event of a fall, and then send a report on his vital sign to his assistant through the “Be with them” application linked to the bracelet. A digital hardware model and an application were created to test and evaluate the product concept physically. The sensors (temperature, heartbeat, fall) measured the vital signs then show the results on the application where these information will be helpful for observing the elderly from a family member in case of fall, high/low temperature or high/low heartbeat. the Blynk is an easy-to-use application that enables the person responsible for the elderly to track his condition and read the sensors’ measurements in history over time.
22/02/2023
After a great development in technology around the world, people now prefer to deal with all companies via the Internet, including requesting service providers for the house for cleaning and maintenance, instead of wasting time visiting cleaning companies in reality. At the same time, there are many problems in terms of dealing with cleaning companies and service providers, and companies are having difficulty meeting the needs of consumers. The project aims to identify the problems related to the applications of providing household cleaning services; a section in the paper summarises the most important applications in this field, the advantages and disad- vantages of them, as well as the proposed solutions for these disadvantages. In this project, we provide an online service platform that contains some solutions to some problems facing companies and customers in order to provide faster and more effective service.
21/02/2023
Nowadays, with most of the world operating remotely, online banking is very popular in Saudi Arabia. However, fraudsters often set up fake websites or apps to obtain bank account information, which they use to scam and steal money. They may create fake transactions or manipulate genuine ones to transfer funds to another account they own. This widespread prob- lem requires solutions from banks to reduce the incidence of bank fraud. Our proposed system aims to tackle this issue by analyzing user behavior, identifying unusual behavior, and alerting users to stop the process if necessary. In this research, we applied two different models with two alternative datasets one of the dataset is real while the second simulated dataset. This research evaluated the performance of two different models: first is hybrid neuro-fuzzy model based on combination of deep neural networks and fuzzy logic algorithm (DNF), while the sec- ond is Deep Neural Network (DNN) model. The result of our experiment showed that the DNN model achieved the highest accuracy by reaching 99.95%, while the DNF model is faster which seems to be more acceptable in real-time transactions.
21/02/2023
Navigation applications are very essential to human beings because it allows to go new places easily. Such applications provide important information like road traffic and interesting places nearby. Based on our literature review, we found that most of the existing navigation applications that are used in Saudi Arabia are not real-time. That is the live traffic information given by these applications are inaccurate and may lead to traffic congestion and delay. Thus, this projects aims at developing a community-based navigation application that allows drivers to share real-time information such as traffic, accidents, police traps,and blocked roads. The application is to collect this information and immediately analyzes it in order to show other drivers with the most optimal route to their destination and important information to make their journey safer and easier. The application will be very helpful for crowded cities like Riyadh and Jeddah.
21/02/2023
Even though the number of cars on the road is increasing by the day, traffic congestion has become a serious problem with more vehicles on the road. Emergency vehicles are affected by congestion at a traffic light and are waiting with a large number of vehicles and traditional traffic light systems are currently unable to handle the increased traffic flow efficiently. Therefore, we investigated and came up with a prototype of a multi-module solution called Salek. First, finding the most efficient timing for the light changes over a series of intersections in a straight path with different speed limits. Second, a module that works on identifying emergency vehicles using image classification machine learning techniques. The goal of the first module is to find the most efficient traffic light synchronization to reduce the time of travel and overall gas consumption using graph algorithms. On the other hand, the second module guarantees the right of passage will always go to emergency vehicles when needed, using the YOLOv5 algorithm. Therefore, these two modules work together to find the most efficient scheduling and interruption as needed. Salek is able to reduce traffic overcrowding at signals by 50%, and the number of vehicles in the entire road (more than one intersection) by 42.8%. In addition, Salek can identify emergency vehicles and give the right of passage with an accuracy of 98%, and it can decrease the response time of emergency vehicles compared to a traditional system.
20/02/2023
Cloud Computing has become a global phenomenon in transforming the IT landscape. Its user-friendly characteristics like pay-as-you-go, service on demand, accessibility, etc. Makes it an IT paradigm of recent times. Cloud computing has many architectures based on its numer- ous characteristics. This research will identify, examine, and explain past, present, and future trends. Moreover, it should also present a proposed guideline to improve cloud-based adoption using available cloud architectures. This research is concerned with two areas related to the field of architecture analysis within cloud computing systems, the first is giving an overview of cloud deployment models, the history of its inception, and the techniques used within it, and the second is showing the different architecture of the cloud in terms of Pros, Cons, Strength- ens, and weaknesses in different past/current cloud computing architectures. Currently, a large number of users keep and share a large amount of their data on the cloud, therefore, there is a necessary need to consider the current cloud architecture and find mechanisms to develop it and fill the gaps within it. In this paper, we focus on an emerging and powerful technology area that is proving to exist day by day, which is security within cloud computing systems. We found that several studies discussing architecture in the cloud were previously published and mentioned in our work; However, only a few of them focused on improving security. The main objective of this research is to explore existing risk management frameworks in order to under- stand, review and thus provide suggestions to build a more robust risk management framework that is appropriate for the cloud computing environment. The latest findings in this field were used to suggest a security approach based on effective management of the security risks that occur within cloud computing systems. It recommended that the approach must achieve its desired goal of building an effective security system capable of dealing with potential security risks and recovery mechanisms that can be launched to address various types of threats within the cloud computing environment.
20/02/2023
Face Detection Technique can detect faces in images and achieve high performance on detec- tion rates. Face Detection technologies and applications have recently developed and improved as a result of advancements in computer software and hardware. Nowadays, it is employed in every part of our lives. Face Detection technologies are being employed in a wide range of fields, including access control, law enforcement, entertainment, personal safety, and many more. Based on OpenCV and a machine learning software library that is mainly oriented at com- puter vision. This library consists of more than 2500 algorithms which contain machine learning tools for classification and clustering, image processing, and vision algorithms. OpenCV library has several built-in pre-trained classifiers developed individually for detecting faces, eyes, and smiles. In this project, we choose the Haar cascade Classifier rather than the Local Binary Pattern (LBP) as one of the OpenCV algorithms. LBP features concentrate on detection hit rate and detection speed. As a result, we selected the Haar cascade Classifier because we aimed for accuracy in detecting faces in images rather than the LBP speed rates. This algorithm is the most efficient and reliable for the implementation phase of face detection systems. From our tested, we can see that the HAAR cascade classifier finds all the faces within an average of 0.07 sec, and the accuracy rate is also %100. It has been a success that we have accomplished most of what we had planned at the beginning of the project. Finally, we verified that the system worked as planned by testing it.
20/02/2023