MATRIX Academic International Online Journal Of Engineering And Technology

Machine Learning–Based Recommendation Systems for E-Commerce Platforms: A Comprehensive Review

Shukhrat Makhmudov — Fri, 10 Oct 2025 00:00:00 +0000

Abstract. Recommendation systems have become a key component of
modern e-commerce platforms. Machine learning techniques enable intelligent
product suggestions based on user behavior and purchase history.
This paper presents a comprehensive review of machine learning–based
recommendation systems used in e-commerce platforms. The study explores
collaborative filtering, content-based filtering, hybrid models, and
deep learning approaches. Furthermore, the paper analyzes evaluation
metrics, system architectures, and challenges such as data sparsity and
cold-start problems. The review highlights that machine learning models
significantly improve recommendation accuracy, user engagement, and
online sales performance.

IoT-Based Intelligent Logistics Systems in Smart Cities: A Review of Architectures, Optimization Techniques, and Applications

Ibrokhimkhuja Rikhsikhujaev — Fri, 10 Oct 2025 00:00:00 +0000

Abstract. The rapid growth of urban populations and the expansion
of e-commerce have significantly increased the complexity of logistics
operations in modern cities. Smart city initiatives increasingly rely on
Internet of Things (IoT) technologies to develop intelligent logistics systems
capable of improving delivery efficiency, reducing congestion, and
minimizing environmental impact. This paper presents a comprehensive
review of IoT-based intelligent logistics systems within smart cities. The
study examines key architectural frameworks, communication technologies,
optimization techniques, and practical applications in urban logistics.
Furthermore, the paper explores recent developments in artificial
intelligence, data analytics, and vehicular communication technologies
that enable real-time decision-making in logistics networks. The analysis
highlights the integration of IoT, AI, and autonomous systems for
improving route optimization, fleet management, and last-mile delivery
operations. Finally, research challenges and future directions for intelligent
logistics systems in smart cities are discussed.

The Role of AI in Bridging Genomic Research and Clinical Diagnostic Models: A Comprehensive Review

Mahwish Athar — Fri, 10 Oct 2025 00:00:00 +0000

Abstract. Genomic research and clinical diagnostics will quickly be
reinvented around Artificial Intelligence (AI) to perform scalable and detailed
analysis of complex biological data and enhance the interpretability
of the clinically relevant variation in genomics. The increasing accessibility
of next-generation sequencing, whole-exome sequencing, electronic
health records, medical imaging and other multi-omics modalities
has opened a space in which computational systems capable of linking
molecular knowledge with real-world diagnostic processes can be created.
This review will discuss the utilisation of AI techniques (such as machine
learning, deep learning, generative AI, and multi-modal intelligence) to
assist variant calling, functional annotation, phenotype -genotype mapping,
disease risk prediction, biomarker discovery, and precision medicine.
The article integrates evidence on the topic of genomic diagnostics, AIbased
sequencing interpretation, and clinical decision support, considering
translational readiness, implementation limitation, and ethical issues.
The genomic discovery through the diagnostic deployment bridges by AI
are elaborated through a conceptual architecture and a tabular literature
review to explain the key avenues of interaction. The review concludes
that AI is not an ancillary analytic tool, and is core-central a translational
process of transforming genomic knowledge into clinical action and
patient-centred clinical models.

An Explainable Hybrid Ensemble-Based Analytical Framework for Interpretable and Early Prediction of Coronary Artery Disease

Sachin Upadhyay, Anil Kumar Sagar, Nihar Ranjan Roy — Fri, 10 Oct 2025 00:00:00 +0000

Abstract. Coronary Artery Disease (CAD) is among the leading causes
of mortality worldwide. Early detection systems are essential for improving
preventive cardiology and reducing cardiovascular risk. This research
proposes an explainable hybrid ensemble-based analytical framework designed
for early and interpretable CAD prediction.
The proposed system integrates heterogeneous machine learning classifiers
including Logistic Regression, Random Forest, Gradient Boosting,
and Support Vector Machines through a stacking-based ensemble approach.
An explainability layer based on feature attribution techniques
is incorporated to provide both global and instance-level interpretation
of predictions.
Experimental analysis demonstrates that the proposed framework achieves
improved predictive accuracy and sensitivity compared with traditional
statistical models and standalone machine learning algorithms. Explainability
results highlight clinically significant risk factors such as age,
cholesterol levels, blood pressure, smoking status, and glucose levels. The
framework supports physician-assisted decision-making rather than automated
diagnosis, enabling responsible AI adoption in healthcare.