RESEARCH

The aim and the scope of this research were to collect data from divorced and married couples via a survey, to uncover the most essential elements that affect divorce in Saudi society. Machine Learning and data mining techniques were used to extract, analyze, and find correlated features, for prediction and classification.

Under Process
Submitted in December 2022

Abstract
When we look at the Saudi community's divorce rates, we can observe that it is growing yearly. This emphasizes the importance of adopting scientifically supported procedures and policies to examine the issue of divorce and avoid it. The scope of this research is to collect data from divorced and married couples via a survey conducted with the assistance of the Al Mawaddah Association for Family Development also, to uncover the essential elements that affect divorce, as well as those that keep a marriage together. Machine Learning and data mining techniques are used to extract, analyze, and find correlated features, for the prediction and classification part, 7 algorithms of Machine learning and two Artificial Neural networks were created, and selected the best hyperparameters through tuning technique, mentioned the number of inputs, hidden, output layers, types of activation function, the optimizer technique to minimize the random noise, and a regularization idea utilized to increase the model's generalization; with justification for which method is better than the others. The performance was measured via evaluation metrics, including the proposed models' accuracy. The results revealed that Random Forest and Logistic Regression got a similar accuracy ratio of 97\% amongst the classifiers for predicting divorce. However, the enhanced architecture of the Neural Network outperforms Machine Learning in terms of performance by 99.52\%. At the end of this research, a comparison between Machine Learning models and Neural Network performance, some solutions, and suggestions are offered to assist the community in reducing divorce cases and maintaining marital relationships.

Read the manuscript 

This paper considers the prediction and analysis of employee attrition using machine learning models. Applying on IBM dataset, five main tests were conducted to predict employee attrition using classifiers. The dataset was tested on random examples for unknown cases. Then, increasing the size of the minority class while decreasing the size of the majority class. In terms of performance evaluation metrics, the proposed approach outperforms other studies, according to the results of the experiment.

Under Process
Submitted in October 2022

Abstract
Machine Learning (ML) has the ability to explore the algorithms that can learn from and make predictions on a given data. Organizations, due to their strategic demands, invest much time and effort in workforce hiring. When employee’s attrition, the companies not only lose a productive staff member but also the resources and funds, in particular, the efforts of HR staff by hiring and selecting certain staff members and training them for their related tasks is invested in them. This paper considers the prediction and analysis of employee attrition using Machine Learning models. Applying on IBM dataset, five main tests were conducted to predict employee attrition using classifiers such as k Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF) classifiers, and Multilayer Perceptron (MLP). The dataset was tested on random examples for unknown cases. Then, increasing the size of the minority class while decreasing the size of the majority class. For the MLP model, providing the details of the Neural Network (NN) model, mentioning the number of inputs, hidden, output layers, types of activation function and the optimizer technique. In terms of performance evaluation metrics, the proposed approach outperforms the aforementioned classifiers, according to the results of the experiment

Read the manuscript 

In this paper, 50 images were collected to detect Saudi car plates. The canny edge method to detect the car edges and different threshold techniques were used to reduce noise. The horizontal projection was applied in the segmentation process to split the plate. After that, a masking technique was utilized to locate and separate the region of interest in the image. Finally, rendering of the results of text on images down the plate regions took place.


Abstract
Automatic license plate recognition has become a significant tool as a result of the development of smart cities. During the experiment studied in the current paper, 50 images were used to detect Saudi car plates. After the preprocessing stage, the canny edge method to detect the car edges and different threshold techniques were used to reduce noise. Horizontal projection was applied in the segmentation process to split the plate. After that, a masking technique was utilized to locate and separate the region of interest in the image. OCR was applied to the processed images to read the characters and numbers in English and Arabic separately. Then, combining the English and Arabic text, after using the re-shaper for the Arabic letters. Finally, rendering of the results of text on images down the plate regions took place. The canny algorithm with projection technique with a proper preprocessing for images produces results with accuracy of 92.4% and 96% for Arabic and English language respectively.

Read the paper

This paper evaluated neural network models for heart attack predictions. Several online learning methods were investigated to predict heart attacks automatically and accurately. An optimizer technique was used to minimize the random noise and the regularization concept was employed to further improve the generalization of the models.

Abstract
Heart Failure represents a critical pathological case that is challenging to predict and discover at an early age, with a notable increase in morbidity and mortality. Machine Learning and Neural Network techniques play a crucial role in predicting heart attacks, diseases and more. These techniques give valuable perspectives for clinicians who may then adjust their diagnosis for each individual patient. This paper evaluated neural network models for heart attacks predictions. Several online learning methods were investigated to automatically and accurately predict heart attacks. The UCI dataset was used in this work to train and evaluate First Order and Second Order Online Learning methods; namely Backpropagation, Delta bar Delta, Levenberg Marquardt and QuickProp learning methods. An optimizer technique was also used to minimize the random noise in the database. A regularization concept was employed to further improve the generalization of the model. Results show that a three layers’ NN model with a Backpropagation algorithm and Nadam optimizer achieved a promising accuracy for the heart attach prediction tasks.

Read the paper

A survey Paper

Abstract
Skin cancer is one of the significant contributors to the cause of death over the world. Melanoma is a well-known kind of skin cancer, which usually is the most malignant lesion compared to other skin lesion types. Classification and diagnosis of skin cancer in medical science field are challenging task due to the amount of data. Skin cancer images or dataset are usually coming in different format. Identification of data involves incredible efforts for preprocessing before the auto-diagnostic work. In this report, Deep Learning (Convolutional Neural Network) is represented to build a model for predicting new cases of skin cancer. In this paper, reviewing several attempts to diagnose skin cancer cases using Deep Learning techniques, such as Convolutional Neural Network CNN for diagnosing melanoma lesions. Basic clinical images were preprocessed to reduce image illumination, then images fed to Convolutional Neural Network models. In the end, the result of each study to differentiate between malignant and benign images in CNN models.


​Read the Survey​​​

R. Antar, S. Alghamdi, J. Alotaibi, and M. Alghamdi, “Automatic number plate recognition of Saudi license car plates,” Engineering, Technology & Applied Science Research, vol. 12, no. 2, pp. 8266–8272, 2022.

---

R. K. Antar, S. T. ALotaibi, and M. AlGhamdi, “Heart attack prediction using neural network and different online learning methods,” International Journal of Computer Science & Network Security, vol. 21, no. 6, pp. 77–88, 2021.