Nafiz Fahad

Mr. Nafiz Fahad


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
21 January 2026 in Discover Artificial Intelligence
Verified
Journal Article
Open access
PDF file
Rasel AhmedRasel Ahmed
Rasel Ahmed
NFNafiz Fahad
Nafiz Fahad
Md. Saef Ullah MiahMd. Saef Ullah Miah
Md. Saef Ullah Miah
JHJakir Hossen
Jakir Hossen
KBKanta Bhattacharjee
Kanta Bhattacharjee
The emergence of mental health issues among university students has a detrimental impact on their academic and job performance, leading to increased levels of anxiety, dropout mentality, and even suicide. Male students are at higher risk compared to their female counterparts. Early detection of mental health issues is crucial for effective treatment. While previous research has focused primarily on specific departments or universities, few studies have examined data from multiple institutions. This study proposes a novel hypertuned ensemble machine learning model that demonstrates its effectiveness in predicting the mental state of university students based on their past two weeks’ experiences. The dataset was collected through an online survey conducted over a period of one month, involving 50 universities, with a total of 400 responses from undergraduate and graduate students. Of these, 67.1% were male, 31.7% were female, and 1.2% declined to disclose their gender identities. To address the data imbalance issue, a hybrid synthetic minority oversampling method, SMOTE-ENN, is employed. Initially, single machine learning models were utilized, including Decision Tree (DT), Naive Bayes (NB), Ada-Boost, Bagging, Random Forest (RF), and Modified Random Forest. In addition, a grid search was performed for hyperparameter optimization. However, the ensemble model with 5-fold cross-validation achieved the highest accuracy of 0.977, along with outstanding performance metrics of precision, recall, and f1 score, at 0.983, 0.977, and 0.979, respectively. Furthermore, Cohen’s kappa coefficient and Matthews correlation coefficient were utilized to measure reliability. The proposed model demonstrates the ability to predict the mental state of university students.

28 March 2026 in Discover Applied Sciences
Verified
Journal Article
Open access
PDF file
FJFariha Jahan
Fariha Jahan
SSSirajum Munira Shifat
Sirajum Munira Shifat
KMKishor Morol
Kishor Morol
NFNafiz Fahad
Nafiz Fahad
KGKah Ong Michael Goh
Kah Ong Michael Goh
JHJakir Hossen
Jakir Hossen
Generative artificial intelligence (AI) is advancing rapidly, with deepfake technology illustrating the intersection of AI and multimedia forensics. This study examines the classification and evolution of deepfake, exploring their definition, societal impact, advantages, challenges, and future potential. It traces the development of deepfake from early techniques to modern generative models and analyzes the role of large language models (LLMs) in their creation and detection. The study also emphasizes the emerging role of large vision-language models (LVLMs) in enhancing multimodal deepfake detection. By categorizing deepfake advancements and their enabling AI technologies, this research offers a structured framework for researchers and practitioners. Additionally, it highlights critical risks, including misinformation and the erosion of public trust. To address these concerns, the study advocates for a multidisciplinary mitigation strategy, combining technical, ethical, and policy-based approaches, along with a proposed prevention framework.

17 June 2025 in arXiv
Verified
Preprint
Open access
PDF file
Md Abrar JahinMd Abrar Jahin
Md Abrar Jahin
SSShahriar Soudeep
Shahriar Soudeep
AAArian Rahman Aditta
Arian Rahman Aditta
MMM. F. Mridha
M. F. Mridha
NFNafiz Fahad
Nafiz Fahad
JHJakir Hossen
Jakir Hossen
Distinguishing between quark- and gluon-initiated jets is a critical and challenging task in high-energy physics, pivotal for improving new physics searches and precision measurements at the Large Hadron Collider. While deep learning, particularly Convolutional Neural Networks (CNNs), has advanced jet tagging using image-based representations, the potential of Vision Transformer (ViT) architectures, renowned for modeling global contextual information, remains largely underexplored for direct calorimeter image analysis, especially under realistic detector and pileup conditions. This paper presents a systematic evaluation of ViTs and ViT-CNN hybrid models for quark-gluon jet classification using simulated 2012 CMS Open Data. We construct multi-channel jet-view images from detector-level energy deposits (ECAL, HCAL) and reconstructed tracks, enabling an end-to-end learning approach. Our comprehensive benchmarking demonstrates that ViT-based models, notably ViT+MaxViT and ViT+ConvNeXt hybrids, consistently outperform established CNN baselines in F1-score, ROC-AUC, and accuracy, highlighting the advantage of capturing long-range spatial correlations within jet substructure. This work establishes the first systematic framework and robust performance baselines for applying ViT architectures to calorimeter image-based jet classification using public collider data, alongside a structured dataset suitable for further deep learning research in this domain.

2 November 2024 in arXiv
Verified
Preprint
Open access
PDF file
NFNafiz Fahad
Nafiz Fahad
FJFariha Jahan
Fariha Jahan
KMKishor Morol
Kishor Morol
Rasel AhmedRasel Ahmed
Rasel Ahmed
AJAbdullah- Al- Jubair
Abdullah- Al- Jubair
The COVID19 pandemic has had a detrimental impact on the health and welfare of the worlds population. An important strategy in the fight against COVID19 is the effective screening of infected patients, with one of the primary screening methods involving radiological imaging with the use of chest Xrays. This is why this study introduces a customized convolutional neural network (CCNN) for medical image classification. This study used a dataset of 6432 images named Chest Xray (COVID19 and Pneumonia), and images were preprocessed using techniques, including resizing, normalizing, and augmentation, to improve model training and performance. The proposed CCNN was compared with a convolutional neural network (CNN) and other models that used the same dataset. This research found that the Convolutional Neural Network (CCNN) achieved 95.62% validation accuracy and 0.1270 validation loss. This outperformed earlier models and studies using the same dataset. This result indicates that our models learn effectively from training data and adapt efficiently to new, unseen data. In essence, the current CCNN model achieves better medical image classification performance, which is why this CCNN model efficiently classifies medical images. Future research may extend the models application to other medical imaging datasets and develop realtime offline medical image classification websites or apps.
1

12 January 2026 in arXiv
Verified
Preprint
Open access
PDF file
SOShafiul Ajam Opee
Shafiul Ajam Opee
NFNafiz Fahad
Nafiz Fahad
ASAnik Sen
Anik Sen
Rasel AhmedRasel Ahmed
Rasel Ahmed
FJFariha Jahan
Fariha Jahan
RIRashedul Islam
Rashedul Islam
Dementia is a complex syndrome impacting cognitive and emotional functions, with Alzheimer's disease being the most common form. This study focuses on enhancing dementia prediction using machine learning (ML) techniques on patient health data. Supervised learning algorithms are applied in this study, including K-Nearest Neighbors (KNN), Quadratic Discriminant Analysis (QDA), Linear Discriminant Analysis (LDA), and Gaussian Process Classifiers. To address class imbalance and improve model performance, techniques such as Synthetic Minority Over-sampling Technique (SMOTE) and Term Frequency-Inverse Document Frequency (TF-IDF) vectorization were employed. Among the models, LDA achieved the highest testing accuracy of 98%. This study highlights the importance of model interpretability and the correlation of dementia with features such as the presence of the APOE-epsilon4 allele and chronic conditions like diabetes. This research advocates for future ML innovations, particularly in integrating explainable AI approaches, to further improve predictive capabilities in dementia care.