Information & Communications Technology
Permanent URI for this collection
Browse
Recent Submissions
Item Predictive analytics of CO2 emission from agri-food activities using aachine learning(Dublin Business School, 2024-08-27) Ragashetti, Sandesh; Dongre, SwatiGlobal warming, Climate change, and Human health are getting impacted due to excessive agri-food emissions. Hence, the predictive analysis of CO2 emissions from agri-food activities is important for policymakers and researchers to develop strategies for sustainable agricultural practices. This study collected and explored secondary historical data on agri-food CO2 emissions in various countries around the world for a time span of 30 years (1990–2020) with machine learning techniques. Since previous research studies left a gap in predicting emissions from the agri-food sector and corresponding temperature rise, this project explores this area by implementing the four predictive models Linear Regression, Decision Trees, Random Forests, and Neural Networks. As a result, exploratory data analysis helps to understand the descriptive statistics, and data visualizations on agri-food activities, emissions, temperature rise, and their relationships. The four predictive models are trained and measured with metrics like MSE, RMSE, MAE, and R-squared. The Linear Regression model emerged as the best model with the highest predictive accuracy, with the lowest RMSE (1.55e-11), MAE (8.37e-12), and highest R2-score (1.00) for CO2 emissions. The study concludes that Linear Regression can serve as a robust tool in predicting CO2 emissions from agri-food activities and helps the policymakers, government bodies, and sustainable environment by providing useful insights and strategies to reduce the environmental impact of agriculture.Item Interpretable machine learning for customer churn prediction(Dublin Business School, 2024) Irmak, Deniz; Sharma, AmitThis study aims to develop and evaluate interpretable machine learning models for predicting customer churn in the telecommunications sector. The dataset, consisting of 7,043 customer records and 21 features, was preprocessed to handle missing values, encode categorical variables, and balance the target class using SMOTE. Five machine learning models were implemented: Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, and Neural Network. The Gradient Boosting model emerged as the most effective, providing a balanced combination of accuracy and interpretability. Partial Dependence Plots (PDPs) and Local Interpretable Model-agnostic Explanations (LIME) were used to explain the model’s predictions, revealing that contract type, monthly charges, and online security services were significant predictors of churn. The results suggest that targeted interventions based on these factors could significantly reduce churn, thereby improving customer retention and business profitability.Item Optimizing logistics routes with advanced algorithms: comprehensive route prediction and efficiency enhancement(2024-09-20) Wu,Zhilei; Magableh, BaselThis study leverages machine learning and data mining techniques to optimise logistics transportation routes, aiming to enhance efficiency in short-distance deliveries. The research aims to develop reliable predictive models, providing optimization solutions for shortdistance logistics routes by comparing the effectiveness of traditional machine learning algorithms and deep neural networks. The model training used historical data from Amazon Logistics in 2018, with key features including route efficiency, package density, and transit time. The study follows the CRISP-DM framework, evaluating the performance of different models in route prediction and optimization. The deep neural network model, combined with graph theory-based algorithms such as the Traveling Salesman Problem (TSP), significantly improves route optimization outcomes. The research concludes that integrating advanced deep learning models with traditional optimization techniques can lead to substantial cost savings and efficiency improvements for logistics companies, while also enhancing customer satisfaction. Future research is recommended to explore larger datasets, real-time data integration, and the impact of economic and social factors.Item Exploring the underrepresentation of women in the fintech industry in Ireland(Dublin Business School, 2024) Abidoye, Ibukun Joan; Smatralova, MonikaThe study explores the underrepresentation of women in the fintech industry in Ireland, focusing on the impact of workplace culture, recruitment practices, and career progression opportunities on gender diversity. Despite global advancements in gender equality, the fintech sector in Ireland continues to exhibit significant disparities, with women underrepresented, particularly in leadership roles. To address this issue, a quantitative survey was conducted with 259 respondents from various fintech companies across Ireland. The survey assessed perceptions of workplaceculture, recruitment processes, and career advancement opportunities concerning gender inclusivity. Data were analysed using descriptive statistics and Pearson correlation coefficient to examine the relationships between these factors and women's representation in the industry. The findings reveal that while there are efforts to promote gender diversity, significant gaps remain. Approximately 65% of respondents believed that their workplace supports gender diversity, yet many reported inconsistencies in policy implementation, recruitment practices, and career advancement opportunities. The hypothesis testing confirmed that positive workplace culture, inclusive recruitment practices, and robust career progression opportunities are significantly correlated with improved representation and retention of women in the industry. This study contributes to the existing literature by highlighting the persistent challenges in achieving gender equality in the Irish fintech industry. It emphasises the need for more effective and consistently applied diversity initiatives to bridge the gender gap. In conclusion, while progress has been made, substantial work remains to ensure that women have equal opportunities in fintech. The study calls for future research to explore these dynamics further and to consider intersectionality and longitudinal impacts on gender diversity.Item Design and implementation of diabetes detection model using machine learning(Dublin Business School, 2024) Johnson Kyanchat, Elisha; Williams, DavidThis work introduces a diabetes prediction machine learning model creation and application process. This work focuses on the performance of machine learning models in predicting the risk of diabetes in individuals and identifying the most relevant factors associated with diabetes using the 2015 Behavioral Risk Factor Surveillance System (BRFSS) dataset, which comprises a strong sample of 253,860 individuals and 21 health-related features. Two primary models—a Support Vector Machine (SVM) and an Artificial Neural Network (ANN)—as well as K-Nearest Neighbors (KNN), Logistic Regression, XGBoost—were also built for comparative study between aforementioned models. Handling missing variables and oversampling to solve class imbalance were part of the steps for training these models. The project sought to evaluate these models' diabetes prediction ability by means of accuracy, precision, recall, F1-score analysis, so establishing their efficacy. Emphasizing the need of variable analysis in improving model accuracy, the results support the continuous study in predictive analytics for treatment of chronic diseases.