Ziad Doughan

Dr. Ziad Doughan

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 March 2022 in IEEE Access
Verified
Journal Article
Open access
ZDZiad Doughan
Ziad Doughan
RKRola Kassem
Rola Kassem
AEAhmad M. El-Hajj
Ahmad M. El-Hajj
AHAli Massoud Haidar
Ali Massoud Haidar
Fooling neural networks is a main concern in the process of Artificial Intelligence optimization. Character perturbation make part of a text unnoticeable for some systems, even for human observers. This research focuses the probe on a novel input preprocessing technique, which applies with the Convolutional Neural Networks for character recognition applications. Using Hand-written, Font, and Arabic Alphabet datasets, we show the CNN competitive results in classification. Then, we apply random deformation by mean of reshaping and relocation on the three testing datasets. As a reaction to change, the CNN demonstrates a confused behavior. Therefore, we design a Weightless Neural Network preprocessor that optimizes the shapes of deformed characters. Finally, we compare the impact of corrected input in improving the CNN performance. In order to measure the success of our preprocessor, we have applied it on well-known CNNs to compare the results between defected and rectified test datasets. Using Keras, we reconstructed LeNet-5, AlexNet and ZFNet for testing. Then, we conduct the test on the three databases to show how the recognition capability of CNNs would decline, then improve after preprocessing.

7 June 2023
Verified
Book Chapter
Closed access
PDF file
HMHadi Al Mubasher
Hadi Al Mubasher
ZDZiad Doughan
Ziad Doughan
LSLayth Sliman
Layth Sliman
AHAli Massoud Haidar
Ali Massoud Haidar

5 May 2023 in IEEE Potentials
Verified
Journal Article
Closed access
ZDZiad Doughan
Ziad Doughan
RKRola Kassem
Rola Kassem
AEAhmad M. El-Hajj
Ahmad M. El-Hajj
AHAli Massoud Haidar
Ali Massoud Haidar
Decades have passed since the world went crazy about the revolution of the machine. Through the years, pioneers in the field tried to imagine how such progress in artificial intelligence (AI) could influence our lives. Were we going to witness sentient machines among us? What would their capabilities, superpowers, and limitations be? Were there ghosts in the machines? Many of these questions confused the greatest minds of our era.

7 May 2025 in IEEE Access
Verified
Journal Article
Open access
ZDZiad Doughan
Ziad Doughan
SISari Itani
Sari Itani
SISamir Itani
Samir Itani
Despite rapid progress in natural language processing (NLP), the development of specialized resources for niche domains—critical for specialized applications like affective computing and emotionally intelligent AI—remains a persistent challenge. While benchmark datasets abound for general tasks, languages like Arabic and fields like multi-dimensional sentiment analysis beyond binary classification as positive or negative suffer from resource scarcity, limiting progress in human-centric applications. To address this gap, we present ArabSis: a novel Arabic corpus for multi-dimensional sentiment analysis across five categorical emotions (Joy, Sadness, Fear, Liking, Hatred). Our work introduces a reproducible framework for creating specialized corpora in low-resource languages, enabling future research in regressive dimensional sentiment analysis and other specialized NLP applications. The ArabSis corpus, developed through systematic data augmentation and human labelling, facilitates advanced analysis using traditional NLP techniques (TF-IDF, Bag of Words) and modern deep learning approaches. It also targets the universal Arabic language whereas previous research focuses on Arabic regardless of the dialect which make small nuances and inconsistencies among dialects unnoticeable and unfixable. We evaluate machine learning (ML) and deep learning (DL) models in one-vs-all classification tasks, demonstrating that ML models (e.g., SVMs, Random Forests) outperform DL counterparts on smaller datasets. An ensemble method combining top-performing models achieves 98.6% accuracy through score averaging and majority voting systems, though revealing inherent biases in ensemble voting mechanisms. The study provides a comprehensive pipeline encompassing data preprocessing, exploratory analysis, and model training, validated through 5-fold cross-validation, establishing a blueprint for developing specialized NLP resources, particularly for under-resourced languages.

30 December 2022 in BAU Journal - Science and Technology
Verified
Journal Article
Open access
PDF file
ZDZiad Doughan
Ziad Doughan
HMHadi Al Mubasher
Hadi Al Mubasher
MBMustafa El Bizri
Mustafa El Bizri
AHAli Massoud Haidar
Ali Massoud Haidar
This paper explores the actual and future impact of the Metaverse as a virtual space. Thus, it focuses the probe on the technical challenges that face this everlasting emerging technology. Today, the Metaverse presents a digital environment to build collective architecture and historical heritage in a virtual space. In this digital world, the modeling and design methodology is based on individual archetypes that can puzzle new elements. Currently, traditional methods require change and adaptation in both the education and work market, especially due to the remote-work integration in the last few years. For example, many components are required to build a Virtual Reality (VR) laboratory or a VR museum. Virtual environments present us with novel opportunities to bring together the real world with a virtual extension or duplication. This technology will remove physical boundaries and design constraints and consequently will open a gate to a metaphysical world. Imagine a world with limitless space where gravity doesn’t exist, and water can float upward. There is no limit for art and architecture but even this magic has its limitation related to computer technology. Therefore, this paper surveys the state-of-the-art computational technologies and the ecosystems of the Metaverse. The paper covers the fields of Computer Vision, Human-Computer Interaction, Artificial Intelligence, Robotics, Internet of Things (IoT), Cloud Computing, and future mobile networks. In application, the Metaverse will allow users to have a fantastic experience being part of worldwide entertainment and socio-economic network.

30 December 2024 in BAU Journal - Science and Technology
Verified
Journal Article
Open access
PDF file
Amira J. ZaylaaAmira J. Zaylaa
Amira J. Zaylaa
IHImane Haidar
Imane Haidar
ZDZiad Doughan
Ziad Doughan
AHAli Massoud Haidar
Ali Massoud Haidar
Limb or organ loss of human remains challenging especially in the world of incessant reliance on touch-based screens and tasks. Thereby, patients can barely withstand and cope with the augmenting restrictions that they encounter due to this loss. Modern means and technologies, such as advanced and artificial parts reduced restrictions on patients with disabilities or a lost limb or organ. Hand prostheses, for instance, provided a powerful tool for improving the functional capabilities of human limbs, thereby improving the quality of life of the user. However, patients using prosthetic arms are still encountering numerous problems, such as, suffering from intact limb and back pain, the high cost of prosthesis systems and hardships related to the performance of the prosthesis, poor control and updating difficulties. Based on the aforementioned problems, the objective was to design a 3D bionic arm which is made of heavy duty plastic with low weight. The aim was to use servo motors instead of stepper motors to maintain less delay and less weight. The aim was also to design an Artificial Intelligence (AI) based-bionic arm program that can be feasibly modified for future purposes, such as adding new gestures and optimizing the system’s control. The new design comprised 3D printing arm, controlling the design, testing motors and EMG sensors, choosing cost-effective parts, simulating and finalizing the real prototype. In conjunction with direct performing motor mechanisms and a full-scale model of a bionic prosthesis, the development was intended for medical rehabilitation of the upper limbs. Experimental results included, the development of a real AI-based system to customize gestures controlled using Neural Network. Results also included maintaining accurate and clean readings of the EMG sensor. Moreover, the new bionic prosthesis arm insured no delay in performance which mimics the normal function of the hand. Results also showed that our design surpassed available designs in its cost-effectiveness, provided that in several other specifications it was comparable. The design was flexible and AI-based controlled. As a future prospect, more algorithms can be tested and more gestures can be tested in the new AI-based design.