As the volume of academic literature continues to burgeon, the necessity for advanced tools to decipher evolving research trends becomes increasingly apparent. This study delves into the utilization of topic modeling techniques—specifically Latent Dirichlet Allocation (LDA), Hierarchical Dirichlet Process (HDP), Non-negative Matrix Factorization (NMF), BERTopic, and Dynamic Topic Modeling (DTM)—applied to a dynamic corpus of research papers. Our research endeavors to confront the challenges posed by capturing temporal dynamics, evolving terminology, and interdisciplinary themes within academic literature. Through a comprehensive comparative investigation of these models, we assess their efficacy in extracting and tracking research topics over time. While DTM exhibited the highest term topic probability, its inclusion of non-meaningful words proved to be a hindrance to its suitability. Conversely, NMF, HDP, LDA, and BERTopic demonstrated comparable performance in topic extraction. Surprisingly, DTM emerged as the most effective model in our research, showcasing its prowess in navigating the intricacies of evolving research trends.

Artificial intelligence is arranged and communicates in a language structure to have a machine learning algorithm that picks up the faster solution in solving humanity’s problems. However, the separation between good and bad all comes down to the human’s intention and determination. Hence, it is crucial to look at the data patterns and come out with the most appropriate machine algorithm. The layering of nodes and information is forming to create a piece of new knowledge. It is a universal language algorithm that speaks with one another to convey the same message. In other words, dimensionality and data simplicity lure in more knowledge and understanding of the problem.

One of the main methods in teaching deaf children effectively is through visualization of learning materials. Previous research has developed augmented reality-based interactive learning media for deaf children. However, these learning media have limitations related to smartphone memory, dependence on certain flashcards, and limited amount of material due to flashcard limitations. Here, researchers innovate to develop SIBI learning media applications based on deep learning technology that material does not depend on a specific flashcard, but the application is able to detect real objects and various images. TensorFlow Lite is considered lighter but still allows the system to run deep learning on mobile devices with low latency. This application is designed to capture a real object or image through the camera, then the object or image will be processed and recognized using TensorFlow Lite. The object recognition process is carried out at the classification layer using switch-case syntax and global class synchronization with the YouTube API. After the image or object is recognized, the system will issue an output in the form of an Indonesian sign language video link from the object.

The management of the financial budget of universities is an extremely complex task due to having numerous different documents and calculation processes. In many developing regions and countries including the Kurdistan Region of Iraq, budget execution and accounting processes are manual. This had deleterious effects on the functioning of their expenditure and income management. This research represents the design and implementation of a responsive web-based system for controlling the financial budget of universities. This system can improve the recording and processing of financial transactions. Moreover, it traces all the stages of the transaction processing from budget releases, does auditing, and accounting of expenditures, incomes, deposits, and funds. Furthermore, it provides financial information on present and past performance. The system is a responsive web-based system, which adjusts the layout of the pages based on the screen size and orientation of the user's device. The system was implemented by using programming languages such as HTML, PHP, JavaScript, jQuery, AJAX, and MySQL. Finally, the provided system needed to be investigated from the performance point of view. Therefore, a questionnaire was used to determine the system’s usability by using the System Usability Scale (SUS) tool. The results revealed that a score of 86.250% of satisfaction has been achieved.

Course adviser in tertiary institution guides students on course enrolment which is part of the registration process for students. It is a phase where a student formally enrolls for requisite courses in a particular semester. Students on gaining admission are required to enroll into courses offered in their chosen programme of study every semester progressively with certain credit limits in each semester. The courses are arranged in an ascending order of complexity such that the criterion for registering for a higher course is to have passed the lower prerequisite course(s). Academic advisers are appointed for students to guide them on course enrolment but due to human factor, a lot of students end up registering for inappropriate courses which leads to inefficiency in career. This research work developed a model that classifies students as either “registrable” Or “not registerable”. Multi-layered Feedforward Neural Networks was used to develop the model that will classify students. The dataset used consists of 150 records, 4 input layers, one hidden layer, and 1output layer. The train/test split of the dataset was in the ration of 80:20. The Networks was trained for 2000 epochs. The accuracy of the model was 0.97. If a student fails more than 15 credit hours of registered courses, such student will be considered “not registerable” and hence redirected to the expert adviser for proper guidance on the course(s) to register.

An unexpected limitation of blood supply to the brain and heart causes the majority of strokes. Stroke severity can be reduced by being aware of the many stroke warning signs in advance. A stroke may result if the flow of blood to a portion of the brain stops suddenly. In this research, we present a strategy for predicting the early start of stroke disease by using Logistic Regression (LR) algorithms. To improve the performance of the model, preprocessing techniques including SMOTE, feature selection and outlier handling were applied to the dataset. This method helped in achieving a balance of class distribution, identifying and removing unimportant features and handling outliers. with the existence of increased blood pressure, body mass, heart conditions, average blood glucose levels, smoking status, prior stroke, and age. Impairment occurs as the brain's neurons gradually die, depending on which area of the brain is affected by the reduced blood supply. Early diagnosis of symptoms can be extremely helpful in predicting stroke and supporting a healthy lifestyle. Furthermore, we performed an experiment using logistic regression (LR) and compared it to a number of other studies that used the same machine learning model, which is logistic regression (LR), and the same dataset. The results showed that our method successfully achieved the highest F1 score and area under curve (AUC) score, which can be a successful tool for stroke disease prediction with an accuracy of 86% compared to the other five studies in the same field. The predictive model for stroke has prospective applications, and as a result, it is still significant for academics and practitioners in the fields of medicine and health sciences.