E-learning platforms changed the way people learn, they are accessible and flexible; Unfortunately, the trend of online education has also brought an increase in fraud activity too and among them is identity verification. Moreover, the threat of fake face recognition, which allows users to get pass from authenticity procedure by presenting an altered/synthetic facial image, is immense and gives birth this vulnerability into E-learning environments. In this paper, we aggregate the knowledge regarding such capture attacks developed in past years and revisit them together with recent related works. This paper also presents the real-time threat scenario for such attacks on academic integrity which has been incorporated while framing a multi-layer approach to secure faculty students from these kind of deceptions in facial recognition system E-learning platforms. This paper will be useful for researchers who aim to research in the areas of fake face detection in e-learning platforms or in other fields and platforms.

The explosion of medical literature over the past decade has resulted in efficient and accurate techniques for text categorization to handle huge amount of data. This work combines ensemble learning methods with coupled multi-word embedding techniques to improve cancer text classification. The intricate semantic links present in medical tests are frequently outside the scope of traditional word embedding models, resulting in not ideal categorization results. To address this problem, we employ e continuous bag-of-words and Skip-gram approaches that yield more complete word representations capturing multiple linguistic nuances. Subsequently, such embeddings are passed through LGBM, CatBoost, and NGBoost ensemble learning classifiers to enhance classification accuracy. With 99.868% accuracy rate, LGBM and CatBoost were the most successful ensemble approaches examined. These approaches provide solid foundation for future work on the use of ensemble methods with complex word representations and for advancing the field of medical text classification.

Healthcare is considered significant topic in the recent research area. However, one of the most commonly diseases which is heart diseases disease. The possibility of early detection to reduce the number of deaths because it is difficult to predict a heart disorder quickly. Recently, many researchers focused on the implementation of several feature extraction techniques and the help of artificial intelligence algorithms to classify this disease, but classification accuracy remained the only difference between these studies. In this paper, the proposed work for the classification of heart diseases was implemented and tested after selecting methods and techniques for data pre-processing and extracting important features that led to obtaining a competitive classification accuracy that reached higher than 93.5% compared to related studies. This finding encourages us and other field researchers to use methods for feature extraction and other strategies described in this paper to classify other diseases.

There is no doubt that the education and training sector has witnessed a major transformation in today’s fast-paced world supported by modern technology. Perhaps one of the most prominent developments in this field is the emergence of electronic (distance) learning and training technology, which has revolutionized the method of transferring knowledge and developing skills. Electronic learning and training refers (remotely) to the use of electronic devices and the Internet to provide educational and training content that is compatible with the enormous technological developments that have become an inevitable reality, providing trainees with an easy and flexible path to a wide range of learning sources, and consistent with what was mentioned above, this has had a profound impact on the development of And the sustainability of training skills for university graduates. This study investigates the significant impact of e-learning on the development and long-term utility of training skills among recent university graduates. Additionally, it aims to clarify the various ways that e-learning technology plays a crucial part in fostering the growth and sustainability of these skills. In the end, this research aims to give university grads the skills they need to excel in their careers as well as to embrace a lifetime of learning and development.

The way we interact with our houses may change when artificial intelligence (AI) technologies are incorporated into smart home networks. New intelligent systems trend to realize smart homes and provide comfort leaving conditions in the residential sector. The first challenge is how academics and practitioners could stay up with the most recent advancements in the field of artificial intelligence for smart homes due to the subject’s rapid development. Second, it is also challenged to control air condition heating temperature, humidity, and ventilation which are considered as constant power load that significantly affect the related power consumption. In order to apply a smart management framework, in this article, we first discuss research on the artificial intelligence applications which is applied for smart home networks. Secondly, a proposed system is introduced to acquire the best heating-controlling scenario for households. Household load profiles were applied to the proposed system, then the performance of this system was compared to the desired temperature. The algorithm regulates the internal air conditions to keep it close to the desired targets, whether the smart home control system is heating, humidity, or ventilation. Which in turn leads to more household’ comfort and save energy.

Wireless Sensor Networks (WSNs) consist of small sensor devices whose main purpose is to detect phenomena in the target area. WSNs are now being used in a variety of essential applications. Data aggregation is an effective strategy in WSNs for best network performance. Because sensor networks have a high node density, similar data is sensed by numerous nodes, resulting in data redundancy which negatively affects the overall performance of the network. The data aggregation may be used to overcome this problem during packet routing from source nodes to the base station. From the recent literature, researchers are still having trouble finding an efficient and appropriate data aggregation method for WSNs. This paper provides insight and support for researchers by knowing how aggregation technology affects the performance of WSNs in the term of energy efficiency, accuracy, and latency after the applying data aggregation strategy. Also, two algorithms namely SOM and HAC were implemented using the Intel Berkeley Research Lab Dataset to show the impact of data aggregation on the overall performance of the WSNs through the results obtained.