Machine Learning Models for Predicting Employee Attrition: A Data Science Perspective DOI
Anber Abraheem Shlash Mohammad,

Zeyad Alkhazali,

Sulieman Ibraheem Shelash Al-Hawary

и другие.

Data & Metadata, Год журнала: 2025, Номер 4, С. 669 - 669

Опубликована: Фев. 27, 2025

Introduction: Employee attrition poses significant challenges for organizations, impacting productivity and profitability. This study explores patterns using machine learning models, integrating predictive analytics with established human resource theories to identify key drivers of workforce turnover. Methods: The research analysed a dataset comprising demographic, job-related, engagement factors. Logistic Regression was employed as the baseline model interpret linear relationships, while Random Forest Decision Trees captured non-linear interactions. Performance metrics such accuracy, precision, recall, F1-score, AUC-ROC were used evaluate effectiveness, alongside feature importance analysis actionable insights. Results: Results revealed that job satisfaction, tenure, departmental dynamics, levels are critical predictors attrition. emerged most effective model, achieving an accuracy 92% 94%, highlighting its capability capture complex patterns. provided interpretable decision rules, offering practical thresholds HR interventions. complemented these models by insights into direct, relationships between Conclusion: finds improves identifying enabling proactive retention strategies. Predictive strengthens traditional theories, providing structured approach reducing employee Organizations can use enhance stability performance. Future could incorporate qualitative methods longitudinal studies refine strategies assess long-term impacts.

Язык: Английский

Decoding Consumer Behaviour: Leveraging Big Data and Machine Learning for Personalized Digital Marketing DOI Creative Commons
Anber Abraheem Shlash Mohammad, Sulieman Ibraheem Shelash Al-Hawary,

Badrea Al Oraini

и другие.

Data & Metadata, Год журнала: 2025, Номер 4, С. 700 - 700

Опубликована: Фев. 11, 2025

IntroductionBig data analytics and machine learning have transformed digital marketing by enabling data-driven insights for personalization. This study investigates the role of engagement metrics, sentiment analysis, consumer segmentation in enhancing effectiveness. Specifically, it examines how these technologies process interaction to uncover actionable insights, segment audiences, drive purchase conversions.MethodThe employed a mixed-methods approach, integrating big techniques. Descriptive statistics highlighted patterns, while k-means clustering segmented consumers based on behavioural emotional data. Sentiment conducted using Natural Language Processing (NLP), captured emotions as positive, neutral, or negative. Regression analysis evaluated influence social media activity, click-through rates, session duration, scores conversion rates.ResultsDescriptive revealed significant variability sentiment, with 37.5% expressing positive sentiment. Clustering identified three distinct segments, reflecting differences showed that had but statistically insignificant relationship conversions, other such rates exhibited minimal impact. The overall explanatory power regression model was low (R-squared = 0.001), indicating need additional factors understand behaviour.ConclusionThe findings emphasize potential analysis. However, their direct impact is limited without broader variables. A holistic, adaptive framework combining behavioural, emotional, contextual essential maximizing personalization driving outcomes dynamic environments.

Язык: Английский

Процитировано

0
Enhancing Metadata Management And Data-Driven Decision-Making In Sustainable Food Supply Chains Using Blockchain And AI Technologies DOI
Anber Abraheem Shlash Mohammad, Ammar Mohammad Al-Ramadan, Sulieman Ibraheem Shelash Al-Hawary

и другие.

Data & Metadata, Год журнала: 2025, Номер 4, С. 683 - 683

Опубликована: Фев. 14, 2025

Sustainability in food supply chains is a critical global challenge, particularly resource-constrained regions like Jordan, where operational inefficiencies and environmental concerns are prevalent. This study explores the integration of blockchain artificial intelligence (AI) technologies to enhance metadata management, forecast sustainability metrics, support decision-making Jordan’s chains. Blockchain's ability improve accuracy, standardization, traceability, combined with AI’s predictive capabilities, offers powerful solution for addressing challenges.MethodsThe research employed mixed-methods approach, combining real-time data from transaction logs, AI-generated forecasts, stakeholder surveys. Blockchain platforms Hyperledger Fabric Ethereum provided insights into accuracy traceability. AI models were developed using machine learning techniques, such as linear regression, waste reduction, carbon footprint energy efficiency. Multi-Criteria Decision Analysis (MCDA), AHP TOPSIS, was applied evaluate trade-offs among goals.ResultsThe results revealed significant improvements (from 83% 96.66%) reductions traceability time 4.0 2.35 hours) following implementation. demonstrated high explaining 88%, 81%, 76% variance efficiency, respectively. ConclusionThis underscores transformative potential achieving goals. By fostering transparency, insights, data-driven decision-making, these innovations can address key challenges chains, offering actionable strategies stakeholders.

Язык: Английский

Процитировано

0
Machine Learning Models for Predicting Employee Attrition: A Data Science Perspective DOI
Anber Abraheem Shlash Mohammad,

Zeyad Alkhazali,

Sulieman Ibraheem Shelash Al-Hawary

и другие.

Data & Metadata, Год журнала: 2025, Номер 4, С. 669 - 669

Опубликована: Фев. 27, 2025

Introduction: Employee attrition poses significant challenges for organizations, impacting productivity and profitability. This study explores patterns using machine learning models, integrating predictive analytics with established human resource theories to identify key drivers of workforce turnover. Methods: The research analysed a dataset comprising demographic, job-related, engagement factors. Logistic Regression was employed as the baseline model interpret linear relationships, while Random Forest Decision Trees captured non-linear interactions. Performance metrics such accuracy, precision, recall, F1-score, AUC-ROC were used evaluate effectiveness, alongside feature importance analysis actionable insights. Results: Results revealed that job satisfaction, tenure, departmental dynamics, levels are critical predictors attrition. emerged most effective model, achieving an accuracy 92% 94%, highlighting its capability capture complex patterns. provided interpretable decision rules, offering practical thresholds HR interventions. complemented these models by insights into direct, relationships between Conclusion: finds improves identifying enabling proactive retention strategies. Predictive strengthens traditional theories, providing structured approach reducing employee Organizations can use enhance stability performance. Future could incorporate qualitative methods longitudinal studies refine strategies assess long-term impacts.

Язык: Английский

Процитировано

0