Samarra Journal of Engineering Science and Research

The construction project consists of multiple stages, independent and
overlapping events and activities, and includes various parties,
including decision-makers, consultants, owners, and project
implementers. Completing a construction project is a difficult and
complex process that is hardly devoid of problems or obstacles. The
presence of risks, which may be either beneficial or detrimental, is one
such impediment. Negative risks, uncertainty is the beginning of the
emergence of risks, which have effects and consequences that prevent
the achievement of project objectives. Therefore, the success of the
project is guaranteed if the risks are identified from the beginning of the
project., to complete a project or even a building project. This article
aims to focus on the concept of risk management in general, the extent
of its emphasis, and how to apply it to projects. Specifically, the initial
stage of risk management, called risk identification, was reviewed.
Ninety-one studies were reviewed between 1990 and 2023,
demonstrating the importance of identifying risks, which helps
stakeholders manage the project using modern management methods.

The ear identification problem is the task of recognizing or
authenticating an individual's identity based on the unique
characteristics of their ear. Like fingerprints or face recognition, ear
identification tasks mainly depend on the unique shape, structure,
and geometry of the ear for accurate classification. Moreover,
realizing high classification accuracy is challenging for both security
systems and forensic investigations since it boosts the robustness of
security systems and favors law enforcement efforts. In this work, we
develop a novel deep learning-based multi-class classification
approach for ear images that uses a hybrid-dense CNN architecture
combined with a self-attention mechanism and uniform manifold
approximation and projection (UMAP) for dimensionality reduction.
We conducted our experiments with ear images labeled for multiclass classification, using a Dense CNN network built from XceptionNet and Dense-Net architectures to capture intricate patterns. The
self-attention mechanism enhances feature representation, while
UMAP optimizes feature space by simplifying high-dimensional data.
The model processes low-dimensional features through fully
connected layers and softmax for classification.
The experimental findings of the proposed methodology showed
classification accuracies of 98.24% on the first dataset and 94.66% on
the second. When compared to cutting-edge models like VGG-Net,
Inception Networks, and ResNet-50, our technique outperformed
them in terms of classification accuracy, precision, recall, and F1
score. This framework has great potential for use in a variety of
applications, including biometric identification, medical diagnostics,
forensic science, security systems, and assistive technology.