Muslims suffer from not knowing the validity of the Hadiths or verification of its degree, which represented the second source of legislation in Islam after the Quran. Although many sites allow users in general and Muslims, in particular, the possibility of verifying the authenticity of the hadith, but it is through an information system which is connected to the database of the Hadith. So, there is no intelligent system that can distinguish the hadith automatically, therefore, in this study, we propose a model that can recognize and categorize the hadith automatically and conclusion the essential features through Hadith classification into Sahih, Hasan, Da’if, and Maudu, based on machine learning techniques. This study is primarily concerned with classifying the hadith according to the memory and reliability of the Hadith’s narrators. This classification does not depend only on the text of the hadith, as in the rest of the other Arab documents, but depend on also the Sanad of Hadith. Therefore, this study was conducted on three methodologies; these methodologies help us to obtain an accuracy that is more reliable for hadith classification compared to previous researches in this area. In addition to building a model using the Decision tree technique based on the Sanad of hadith for helping us deciding to judge the validity of the hadith, the accuracy of this classifier reached up to 92.59%. Several Learning algorithms have been used in this study, but we reported the best three classifiers (LinearSVC, SGDClassifier, and Logistic regression), which achieved higher accuracy reached up to 93.69%, 93.51, and 92.27% respectively.

Diffuse liver diseases such as fatty liver and cirrhosis, are leading causes of disability and fatality across the world. Early diagnosis of these diseases is extremely important to save lives and improve the effectiveness of treatment. This study proposes a non-invasive method for diagnosing liver diseases using ultrasound images, by classifying liver tissue as normal, steatosis, or cirrhosis, using feature extraction, feature selection, and classification. First, the correlation, homogeneity, variance, entropy, contrast, energy, long run emphasis, run percentage, and standard deviation are determined. Second, the most efficient features are selected based on the Fisher discriminant and manual selection methods. Third, three voting-based sub-classifiers are used, namely, the normal/steatosis, normal/cirrhosis, and steatosis/cirrhosis classifiers. The final liver tissue classification is based on the majority function. Our classification method provides two key contributions: combination of two different feature selection methods, avoiding the limitations of each method while benefiting from their strengths; and classifier categorization into three sub-classifiers, where the overall classification is based on the decision of each individual sub-classifier. We obtained recognition accuracies for the normal/steatosis, normal/cirrhosis, and steatosis/cirrhosis classifiers as 95%, 95.74%, and 94.23%, respectively, and an overall recognition accuracy of 95%, which outperforms other methods.

Chest infections encompass a wide range of conditions, including pneumonia, tuberculosis, bronchitis, and other respiratory illnesses. By harnessing the power of AI and deep learning algorithms, we aim to provide healthcare professionals with a valuable diagnostic tool. The AI model is trained on an extensive dataset of labeled chest x-ray images, enabling it to learn and recognize patterns indicative of different types of infections. By leveraging convolutional neural networks (CNNs) and advanced image recognition techniques, the model automatically identifies abnormalities and subtle indicators of infection, ensuring accurate and efficient analysis. One of the key advantages of AI model is its ability to rapidly process and analyze a large volume of x-ray images. Additionally, the model accommodates variations in image quality and positioning, making it adaptable to different healthcare settings, after conducting the results of the experiments, the accuracy of this model reached up to 92%

There are many classifications of Hadith mainly: according to reliability and memory of hadith' narrators (sahih, hasan, maudu, da’if), and according to the sayings, actions or characteristics of the Prophet (saying, doing, reporting, describing), each Hadith contains three main parts: Sanad, Matn and Taraf. In this study we try to find the relationship between the different classifications of hadith using Association rule mining, part of speech (POS) and Chi-square techniques, based on the text and Sanad of hadith to build a model that is able to distinguish and categorize the hadith categories, using supervised learning. The experimental results showed that the relation (confidence) between saying & sahih is 79.99%, doing & hasan is 64.5%, reporting & da’if is 64.84, describing & maudu is 58.5%, and the best classifier has given high accuracy been Naïve bayes NB; it achieved higher accuracy reached up to 97.5 %, followed by the LinearSVC and K- Neighbors reached up to 96.5%.