Fall prediction is a critical process in ensuring the safety and well-being of individuals, particularly the elderly population. This paper focuses on the development of a fall detection and prediction system using wearable sensors and machine learning algorithms. The system issues an alarm upon predicting the occurrence of falling and sends alerts to a monitoring centre for timely assistance. Wearable sensor devices, including Inertial Measurement Units (IMUs) equipped with accelerometers, gyroscopes, and magnetometers are utilized for data collection. UPFALL, a comprehensive online freely available dataset, had been utilized for training and verifying the proposed system. Several machine learning algorithms, such as K-Nearest Neighbours (KNN), Random Forest, Support Vector Machine (SVM), and Gradient Boosting, are implemented and evaluated. Among these algorithms, KNN demonstrates the highest effectiveness for fall detection having an accuracy of 93.5%. Furthermore, deep learning models, including Gated Recurrent Units (GRU), Long Short-Term Memory (LSTM), and Convolutional Neural Network (CNN), are employed. The GRU model exhibits superior performance among the deep learning approaches by having the least train and test loss of 0.219 and 0.267 respectively. An early fall prediction function is incorporated by establishing a threshold selection process based on logical analysis. The maximum voting concept is employed to determine the optimal threshold.

Developing well-prepared and competent graduates is one of the main goals of all university programs globally. Recently, Computer Science (CS) has achieved tremendous success among all career fields driven by the strong competence in the market and the rapid changes in technology. Our goal is to develop an automated framework that provides efficient management, evaluation and improvement of the CS students education, as well as a profound establishment of a successful study tree for CS university programs. Such a challenging goal comprises major factors that should be inclusively considered. High school (HS) students are expected to join CS university programs with different educational backgrounds and learning capabilities. The strength of association among several performance-related factors including the academic performance of students in HS is evaluated to gain insights and infer indicators in CS programs. The automatic correlational analysis of the prerequisites for each course is also investigated to assess the program structure and dependencies among several CS courses. In this comprehensive study, all these factors are efficiently analyzed in order to investigate the valid causes of low and high performance of both CS university students and programs. Experimental results have concluded several major findings with validated associations that assure and prioritize the importance of evaluation and improvement of CS education.

In this research study, we introduce an innovative approach to managing load distribution in smart grid infrastructures. The primary objective is to regulate power demand during peak hours, utilizing dynamic pricing mechanisms. Unlike traditional centralized systems, our proposed solution operates on a distributed paradigm, deriving its foundation from the rational actions of grid users. To effectively analyze the prevailing scenario, we employ a game-theoretic model to understand the behaviors of users, who, in our model, act primarily out of self-interest. An initial observation indicates that if left to their own devices, these selfish users might make decisions that could be detrimentally arbitrary to the system's efficient functioning.In response to this challenge, we introduce a pricing mechanism designed to enhance the allocation quality influenced by these self-centered users. We begin by establishing the validity of our approach through a proof that the game, even with the existence of selfish users, will reach a Nash equilibrium. This equilibrium ensures that no player can benefit by deviating unilaterally from their chosen strategy after considering an opponent's choice. Following this, we demonstrate that by incorporating our dynamic pricing strategy, the resulting allocation's peak demand is effectively managed. Specifically, the peak of this allocation, even in the worst-case scenario, will not exceed double the value of an ideal or optimal peak. This result underscores the efficiency and efficacy of our proposed mechanism in maintaining a balance between user behavior and systemic demand, ensuring a more stable and sustainable smart grid infrastructure.

In the realm of web security, there is a growing shift towards harnessing machine learning techniques for Cross-Site Scripting (XSS) vulnerability detection. This shift recognizes the potential of automation to streamline identification processes and reduce reliance on manual human analysis. An alternative approach involves security professionals actively executing XSS attacks to precisely pinpoint vulnerable areas within web applications, facilitating targeted remediation. Furthermore, there has been a growing interest in machine learning-based methods for creating XSS payloads in academic and research domains. In this research, we introduce a new model for generating XSS payloads, utilizing a combination of auto-regressive and generative AI models to craft malicious scripts intended to exploit potential vulnerabilities. Our approach to XSS vulnerability detection encompasses both frontend and backend code, providing organizations with a comprehensive means to enhance web application security.

Universities and students face the challenge of finding the career that best suits the students. Programming is currently one of the highly sought-after careers and is one of the highest paying jobs. However not all students succeed in this career. In fact failure rates in programming field is relatively high globally. In this research, we used a spatial rotation test and Non-Verbal reasoning test as well as gender, at the beginning of the semester to observe the relationship between those tests and students' grades at the end of the semester for 2 groups. Group 1 are Computer Science students studying programming course and group 2 are non-Computer Science students studying programming course. For each group, we created a predictive model using 2 methods. Method 1 uses the conventional aggregate score of cognitive tests and method 2 uses answers to selected questions in each test. In method 2, we applied a data-driven methodology which utilize individual answers to questions as input to the model. To the best of our knowledge we are the first in this field to apply this method. The modeling results show usefulness of those test to predict programming and non-programming course grades using both methods, however method 2 demonstrates much higher superiority over the aggregate method 1.

The smart city concept serves the well-being of the urban population in order to improve quality of life in different life aspects including logistic services. However, the continuous growth of logistics volume combined with information opaqueness and process complexity led to challenging issues related to managing the services and information of logistics. Hence, efficient management of logistics activities with traceability and condition monitoring capabilities is required to ensure quality and safe delivery. It is necessary also to ensure the accuracy and dependability of distribution data. In this context, this paper proposes a smart and privacy-preserving logistics system for high-price goods distribution. An intelligent parcel (iParcel) containing piezoresistive sensors is developed to pack delivered goods during shipping process for violation detection such as severe fall or theft. Moreover, smart contracts based on blockchain is also developed for automatic approval and payment with the consideration of distributing the shipping information between legitimate logistics parties only. A zero-knowledge proof is used to conceal blockchain address and prove the authentication. iParcels are automatically tracked and traced in which upon violation occurrence, the contract is cancelled, and the payment is refunded. The transaction fee per party is reasonable for high-price products in the pay of guarantee successful shipment.

The market needs a deeper and more comprehensive grasp of its insight, where the analytics world and methodologies such as “Sentiment Analysis” come in. These methods can assist people especially “business owners” in gaining live insights into their businesses and determining wheatear customers are satisfied or not. This paper plans to provide indicators by gathering real world Amazon reviews from Egyptian customers. By applying both Bidirectional Encoder Representations from Transformers “Bert” and “Text Blob” sentiment analysis methods. The processes shall determine the overall satisfaction of Egyptian customers in the electronics department - in order to focus on a specific domain. The two methods will be compared for both the Arabic and English languages. The results show that people in Amazon.eg are mostly satisfied with the percentage of 47%. For the performance, BERT outperformed Textblob indicating that word embedding model BERT is more superior than rule-based model Textblob with a difference of 15% - 25%.

The amount and variety of raw data generated in the agriculture sector from numeroussources, including soil sensors and local weather stations, are proliferating. However, these raw data in themselves are meaningless and isolated and, therefore, may offer little value to the farmer. Data usefulness is determined by its context and meaning and by how it is interoperable with data from other sources. Semantic web technology can provide context and meaning to data and its aggregation by providing standard data interchange formats and description languages. In this paper, we introduce the design and overall description of a core ontology that facilitates the process of data interoperability in the agricultural domain.

This paper proposes using entity2rec [1] which utilizes knowledge graph-based embeddings (node2vec) instead of traditional embedding layers in embedding based recommender systems. This opens the door to increasing the accuracy of some of the most implemented recommender systems running in production in many companies by just replacing the traditional embedding layer with node2vec graph embedding without the risk of completely migrating to newer SOTA systems and risking unexpected performance issues. Also, Graph embeddings will be able to incorporate user and item features which can help in solving the well-known Cold start problem in recommender systems. Both embedding methods are compared on the movie-Lens 100-K dataset in an item-item collaborative filtering recommender and we show that the suggested replacement improves the representation learning of the embedding layer by adding a semantic layer that can increase the overall performance of the normal embedding based recommenders. First, normal Recommender systems are introduced, and a brief explanation of both traditional and graph-based embeddings is presented. Then, the proposed approach is presented along with related work. Finally, results are presented along with future work.

Falls are unintentionally events that may occur in all age groups, particularly for elderly. Negative impacts include severe injuries and deaths. Although numerous machine learning models were proposed for fall detection, the formulations of the models are limited to prevent the occurrence of falls. Recently, the emerging research area namely early fall prediction receives an increasing attention. The major challenges of fall prediction are the long period of unseen future data and the nature of uncertainty in the time of occurrence of fall events. To extend the predictability (from 0.5 to 5 s) of the early fall prediction model, we propose a particle swarm optimization-based recurrent neural network and long short-term memory (RNN-LSTM). Results and analysis show that the algorithm yields accuracies of 89.8–98.2%, 88.4–97.1%, and 89.3–97.6% in three benchmark datasets UP Fall dataset, MOBIFALL dataset, and UR Fall dataset, respectively.