Explainable AI: Enhancing Interpretability of Machine Learning Models DOI

Duru Kulaklıoğlu

Human computer interaction., Journal Year: 2024, Volume and Issue: 8(1), P. 91 - 91

Published: Dec. 6, 2024

Explainable Artificial Intelligence (XAI) is emerging as a critical field to address the “black box” nature of many machine learning (ML) models. While these models achieve high predictive accuracy, their opacity undermines trust, adoption, and ethical compliance in domains such healthcare, finance, autonomous systems. This research explores methodologies frameworks enhance interpretability ML models, focusing on techniques like feature attribution, surrogate counterfactual explanations. By balancing model complexity transparency, this study highlights strategies bridge gap between performance explainability. The integration XAI into workflows not only fosters trust but also aligns with regulatory requirements, enabling actionable insights for stakeholders. findings reveal roadmap design inherently interpretable tools post-hoc analysis, offering sustainable approach democratize AI.

Language: Английский

An explainable AI for breast cancer classification using vision Transformer (ViT) DOI

Marwa Naas,

Hiba Mzoughi, Ines Njeh

et al.

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 108, P. 108011 - 108011

Published: May 2, 2025

Language: Английский

Citations

0
Enhancing Alzheimer’s Disease Detection: An Explainable Machine Learning Approach with Ensemble Techniques DOI Creative Commons
Eram Mahamud, Md Assaduzzaman,

Jahirul Islam

et al.

Intelligence-Based Medicine, Journal Year: 2025, Volume and Issue: unknown, P. 100240 - 100240

Published: April 1, 2025

Language: Английский

Citations

0
Mapping knowledge landscapes and emerging trends in digital biomarkers for dementia in older adults: A scoping and bibliometric analysis DOI
Azliyana Azizan, Shihua Cao,

Akehsan Dahlan

et al.

Archives of Gerontology and Geriatrics Plus, Journal Year: 2025, Volume and Issue: 2(2), P. 100148 - 100148

Published: April 9, 2025

Language: Английский

Citations

0
Neurodegenerative disorders: A Holistic study of the explainable artificial intelligence applications DOI Creative Commons
Hongyuan Wang, Shiva Toumaj, Arash Heidari

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 153, P. 110752 - 110752

Published: April 18, 2025

Language: Английский

Citations

0
AI-driven early diagnosis of specific mental disorders: a comprehensive study DOI Creative Commons
Firuze Damla Eryılmaz Baran, Meriç Çetin

Cognitive Neurodynamics, Journal Year: 2025, Volume and Issue: 19(1)

Published: May 5, 2025

Abstract One of the areas where artificial intelligence (AI) technologies are used is detection and diagnosis mental disorders. AI approaches, including machine learning deep models, can identify early signs bipolar disorder, schizophrenia, autism spectrum depression, suicidality, dementia by analyzing speech patterns, behaviors, physiological data. These approaches increase diagnostic accuracy enable timely intervention, which crucial for effective treatment. This paper presents a comprehensive literature review applied to disorder using various data sources, such as survey, Electroencephalography (EEG) signal, text image. Applications include predicting anxiety depression levels in online games, detecting schizophrenia from EEG signals, text-based indicators suicidality diagnosing magnetic resonance imaging images. eXtreme Gradient Boosting (XGBoost), light gradient-boosting (LightGBM), random forest (RF), support vector (SVM), K-nearest neighbor were designed convolutional neural networks (CNN), long short-term memory (LSTM) gated recurrent unit (GRU) models suitable dataset models. Data preprocessing techniques wavelet transforms, normalization, clustering optimize model performances, hyperparameter optimization feature extraction performed. While LightGBM technique had highest performance with 96% prediction, optimized SVM stood out 97% accuracy. Autism classification reached 98% XGBoost, RF LightGBM. The LSTM achieved high 83% diagnosis. GRU showed best 93% suicide detection. In dementia, have demonstrated their effectiveness analysis reaching 99% findings study highlight sequential applicability medical or natural language processing. XGBoost noted be highly accurate ML tools clinical diagnoses. addition, advanced pre-processing confirmed significantly improve performance. results obtained this revealed potential decision systems disorders AI, facilitating personalized treatment strategies.

Language: Английский

Citations

0
Unveiling Alzheimer’s Progression: AI-Driven Models for Classifying Stages of Cognitive Impairment Through Medical Imaging DOI
Vaibhav C. Gandhi,

Dhruvi Thakkar,

Mariofanna Milanova

et al.

Lecture notes in computer science, Journal Year: 2025, Volume and Issue: unknown, P. 55 - 87

Published: Jan. 1, 2025

Language: Английский

Citations

0
A Benchmark Dataset for Cricket Sentiment Analysis in Bangla Social Media Text DOI Open Access
Tanjim Mahmud, Rezaul Karim,

Rishita Chakma

et al.

Procedia Computer Science, Journal Year: 2024, Volume and Issue: 238, P. 377 - 384

Published: Jan. 1, 2024

This study introduces a novel benchmark dataset designed for Cricket Sentiment Analysis on Bangla social media posts, emphasizing low-resource setting. The was meticulously curated through manual collection across diverse platforms, ensuring comprehensive representation of user sentiments. Annotations validated quality, achieving remarkable Cohen Kappa score 0.97. Experimentation with machine learning (ML) models revealed challenges, traditional approaches yielding modest RNN accuracy 0.5239. However, deep (DL) showcased significant performance enhancements. LSTM model achieved 0.897 accuracy, while the BiLSTM surpassed expectations at 0.952. These findings highlight DL's efficacy in capturing nuanced sentiments cricket-related contributing high-quality and insights into suitability sentiment analysis linguistic contexts.

Language: Английский

Citations

2
Protein Sequence Classification Through Deep Learning and Encoding Strategies DOI Open Access
Farzana Tasnim, Sultana Umme Habiba, Tanjim Mahmud

et al.

Procedia Computer Science, Journal Year: 2024, Volume and Issue: 238, P. 876 - 881

Published: Jan. 1, 2024

Protein sequence classification is vital for understanding protein functionalities, aiding in the inference of novel functions. Machine learning and deep algorithms have revolutionized this field, offering insights into specific classes This study employs Natural Language Processing (NLP) techniques, including Integer Blosum encoding, efficient classification. SVM with count vectorizer achieves highest accuracy 92%, while encoding CNN surpasses NLP embedding techniques by 4%. The goal to develop an automated system predicting functionality based on classification, contributing advancements proteomics computational biology.

Language: Английский

Citations

2
FiboNeXt: Investigations for Alzheimer’s Disease detection using MRI DOI Creative Commons
Türker Tuncer, Şengül Doğan, Abdülhamit Subaşı

et al.

Biomedical Signal Processing and Control, Journal Year: 2024, Volume and Issue: 103, P. 107422 - 107422

Published: Dec. 25, 2024

Language: Английский

Citations

1
New Era of Intelligent Medicine: Future Scope and Challenges DOI

Ashwani Kumar,

Aanchal Gupta, Utkarsh Raj

et al.

2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), Journal Year: 2024, Volume and Issue: unknown, P. 1 - 6

Published: March 14, 2024

The integration of Artificial Intelligence (AI) into the global healthcare landscape has undergone a remarkable transformation, presenting unprecedented opportunities and challenges. This review explores transformative impact in health care, examining current applications, growth projections, projected compound annual rate (CAGR) for AI market is 37%, reaching $188 billion by 2030. AI's potential to reduce drug development costs prevent medication dosing errors evident. From early models like CASNET contemporary Deep Learning, revolutionized medical diagnostics. envisions future with accessible through chatbots telemedicine, data-driven platforms personalized treatment, data cards. Technological advancements, including increased computational power cloud storage, play pivotal role, challenges managing vast heterogeneous data. concludes addressing dynamic must overcome impact.

Language: Английский

Citations

1