مواقع التدريسيينجامعة الكوفة
0Publications
0Lectures

About

- He holds a Bachelor's degree from Al-Furat Al-Awsat University - Engineering Technical College, Najaf - Department of Communication Technologies Engineering.
- He holds a master's degree from the Federal University of the South (Rostov City - Federal Russia) - Department of Information Technology and Communication Systems.
Responsible for network maintenance in the Faculty of Education for Girls from 2014-2019.-
-Responsible for maintaining computers in the Faculty of Education for Girls from 2014-2019.
- Director of the Maintenance and Services Division at the Faculty of Education for Girls in 2019

Publications

3
2026

Detecting and Mitigating Data Leakage in Breast Cancer Prediction Using an Explainable Machine Learning Framework with XGBoost and SHAP

AuthorsKarrar Abdullah Mohammed Habeeban, Aseel Faisal Alshaibanee , Marwah Habiban and Abbas Fadhil Mahdi
JournalInternational Information and Engineering Technology Association
Date2026-08-31
Abstract

Machine learning models have shown considerable potential for clinical prediction; however, their reported performance can be substantially affected by data leakage, leading to unreliable estimates of predictive capability. This study proposes an explainable leakageaware framework for identifying and mitigating data leakage in breast cancer prediction models. The proposed framework integrates data preprocessing, feature auditing, Extreme Gradient Boosting (XGBoost)-based classification, and SHapley Additive exPlanations (SHAP) analysis to examine whether highly influential predictors represent clinically available information or outcome-related leakage. Experiments were conducted using two breast cancer datasets, including the Breast Cancer Wisconsin dataset from Kaggle and a clinical dataset from Mendeley Data. The initial model trained on the Mendeley dataset achieved an accuracy of 0.9710 and receiver operating characteristic-area under the curve (ROC-AUC) of 0.9968; however, SHAP analysis revealed that several dominant predictors were associated with post-outcome information, indicating potential target leakage. After removing leakage-prone variables, the model performance decreased to a more realistic level, achieving an accuracy of 0.8950 and ROC-AUC of 0.9416. Additional experiments using random forest (RF), support vector machine (SVM), and CatBoost confirmed that leakage effects were not limited to a specific algorithm. The results demonstrate that explainability-based auditing can support more reliable evaluation of clinical machine learning models by identifying misleading predictive patterns caused by inappropriate features.

2018

Image Steganography using K-LSB Embedding with Edge Detection and Chaotic Map

AuthorsKarrar Abdalluh Mohammed
JournalInternational Journal of Computer Science and Mobile Computing
Abstract

Steganography is the science of embedding secret data inside data so they can be sent to destination security. Image steganography is the most popular type of carrier to hold information. Many algorithms have been proposed to hide information into digital images; the least significant bit algorithm (LSB) is one of these algorithms that is widely used in steganography.In this paper, we introduced propose steganography method by using edge detection of 6 bits in the image by edge detection and chaotic map. In work, we use 1-LSB or 2-LSB embedding, it depends on the threshold for edge detection image. we use locations not serialized of the pixel to embedding information rely on tent map.

2018

An Enhanced AAM Approach for Face Recognition Application

AuthorsIsraa Mohammed Rahi, Rasha Awad Hussain,Karrar Abdalluh Mohammed
JournalInternational Journal of Computer Science and Mobile Computing
Abstract

Face recognition systems are considered to be the most user-friendly systems in the biometric world. Also, they are one of the active research fields in which the researchers work to improve the efficiency and efficacy of biometric applications. The major challenges reside on these applications are to extract unique and differentiable features that are not disturbed by pose, illuminations and other environmental and physical variations. Active Appearance Model (AAM) is one of the population solutions that extract features by precise modeling of human faces under various physical and environmental circumstances. In this paper, an Enhanced AAM based 1 to N face recognition system is proposed. The experimental results show the performance of the proposed face recognition system.

Lectures

6
View details