A5015387958- Advanced Text Analysis Techniques
- Information Retrieval and Search Behavior
- Topic Modeling
- Text and Document Classification Technologies
- Spam and Phishing Detection
- Metaheuristic Optimization Algorithms Research
- Advanced Multi-Objective Optimization Algorithms
- Radiomics and Machine Learning in Medical Imaging
- Recommender Systems and Techniques
- Medical Imaging and Analysis
- Machine Learning and Data Classification
- Digital Media Forensic Detection
- Evolutionary Algorithms and Applications
- Hydrological Forecasting Using AI
- Halal products and consumer behavior
- Algorithms and Data Compression
- Education and Islamic Studies
- Authorship Attribution and Profiling
- Web Data Mining and Analysis
- AI in cancer detection
- Advanced Authentication Protocols Security
- User Authentication and Security Systems
- Advanced Clustering Algorithms Research
- Cryptography and Data Security
Al-Balqa Applied University
2020-2024
Universiti Sains Islam Malaysia
2021
Umm al-Qura University
2021
National University of Malaysia
2013-2015
Abstract Analysis of histopathological images (HIs) is crucial for detecting breast cancer (BR). However, because they vary, it still very difficult to extract well-designed elements. Deep learning (DL) a recent development that used high-level features. DL techniques continue confront several problems, such as the need sufficient training data models, which reduces classification findings. In this study, an ensemble deep transfer convolutional neural network presented address problem. The...
Arabic is one of the Semitic languages in antiquity and six official UN. Also, classification plays a significant essential role modern applications. There big difference between handling English text classification; preprocessing also challenging for text. This paper presents implementation Naïve Bayes classifier with without stemmer. A set four categories 800 documents were used from Text Retrieval Conference (TREC) 2001 dataset. The results showed that light stemmer achieves better than...
The conventional information retrieval (IR) framework consists of four primary phases, namely, pre-processing, indexing, querying and retrieving results. Some phases the current Arabic IR (AIR) have several drawbacks. This research aims to enhance an AIR by improving processes in a framework. We introduce enhanced stop-word list pre-processing level investigate stemmers. In addition, WordNet was utilized corpus query expansion levels. also adopted semantic for Pseudo Relevance Feedback....
The ongoing growth in the vast amount of digital documents and other data Arabic language available online has increased need for classification methods that can deal with complex nature such data. plays a large important role many modern applications interferes sciences, which start from search engines do not end Internet Things. However, addressing Arab errors high performance is largely insufficient to huge quantities reveal documents; while some work was tackled out on text, most...
One of the main issues associated with search engines is query–document vocabulary mismatch problem, a long-standing problem in Information Retrieval (IR). This occurs when user query does not match content stored documents, and it affects most tasks. Automatic expansion (AQE) one common approaches used to address this problem. Various AQE techniques have been proposed; these mainly involve finding synonyms or related words for terms. Word embedding (WE) methods that are currently receiving...
In recent times, many fake faces have been created using deep learning and machine learning. Most made with are referred to as “deepfake photos.” Our study’s primary goal is propose a useful framework for recognizing deep-fake photos transformative techniques. This paper proposed convolutional neural network (CNN) models based on transfer method-ologies in which the designed classifier global average pooling (GAP), dropout, dense layer two neurons that use SoftMax substituted final fully...
Natural language processing represents human in computational technique, which is to achieve the extraction of important words. The verbs and nouns found Arabic are significantly pertinent process differentiating each class label available for purpose machine learning, specifically 'Arabic Clustering'. This paper implemented sourced from Qur'an text clustering further evaluation by using two datasets. limitations conventional clusters were identified, such as k-means on initial centroids....
The adaptation of a Query Expansion (QE) approach for Arabic documents may produce the worst rankings or irrelevant results. Therefore, we have introduced technique, which is to utilise WordNet in corpus and query expansion level. A Point-wise Mutual Information (PMI) corpus-based measure used semantically select synonyms from WordNet. In addition, Automatic (AQE) Pseudo Relevance Feedback (PRF) methods were also explored improve performance information retrieval (AIR) system. experimental...
Authenticated key exchange protocols are a fundamental cryptography building block for establishing secret shared among participating parties over an open network model. Mohamed et al. proposed tightly-secure authenticated without NAXOS approach based on decision linear problem in order to solve designing protocol using trick. al.'s has tight security proof the eCK model under random oracle and its claims verified by games sequence tool. However, we show that cannot withstand compromise...