ARC (Academic Research Collection) (College Dubin)
Not a member yet
359 research outputs found
Sort by
Using Predictive Analytics to identify risk of Heart Disease based on lifestyle factors and health metrics.
Get PDFIn this project, we will report an innovative application, for the healthcare sector usage, which basically is a health tracking and disease prevention application. The application will enable users to log their daily meals, exercise routines, and lifestyle habits, providing a comprehensive overview of the user\u27s health status. By making use of Machine Learning and data analytics, our solution offers a personalised and automated insight and predictive analytics, which empowers users to proactively manage their well-being.
Through a detailed data analysis, users will gain valuable insights of potential diseases development and risk. This report will explore the development process, implementation of machine learning models, data visualisation techniques, and the transformative impact of our application over health management
Supply Chain Optimisation with Machine Learning and Neural Networks: Applications to Demand Planning, Supply Planning, and Inventory Planning.
Get PDFThis thesis explores the impact of machine learning (ML) on supply chain planning, particularly in demand forecasting, supply planning, and inventory optimisation. By analysing literature on supply chain management, data flow, and the intersection of ML and competitive advantage, the author contextualises the research within a globalised market\u27s demands. Case studies, interviews with industry professionals, and raw data collection provide empirical support for evaluating the research objectives and documenting the integration of ML in supply chain processes.
The findings reveal that optimised ML models, particularly those using model stacking (autoregressors, GRUs, and Random Forests), significantly outperform traditional demand forecasting methods, achieving a 70% MAPE improvement over a 45% benchmark. The integration of advanced techniques like XGBoost further optimised supply and inventory planning. The research concludes that leveraging ML not only enhances forecast accuracy but also strengthens supply chain competitiveness through superior planning outputs.
By critically relating empirical data to literature insights, the author demonstrates that ML-driven approaches enhance supply chain management in a Central European wholesale clothing business. This research validates the transformative potential of advanced data analytics for achieving a competitive edge in the supply chain
Using Machine Learning to Identify Hate Speech and Offending Language on Twitter.
Get PDFThis project focuses on applying Machine Learning (ML) techniques to detect hate speech and offensive language on Twitter, addressing ethical concerns like cyberbullying and fostering a safer online environment. The topic is chosen for its societal significance and business relevance, as hostile online behaviour negatively impacts user experiences and platform credibility.
To achieve this, the study implements four distinct ML models to develop an automated system capable of identifying and categorising content as offensive, non-offensive, or neutral. The system aims to contribute to mitigating harmful interactions on social media and improving user safety by effectively classifying potentially problematic content.
The project\u27s approach underlines the importance of integrating technological solutions to address ethical challenges while aligning with business interests in creating more inclusive digital spaces
Recommended Strategies for Tesla Company in Encouraging People to Buy Electric Vehicles (EVs).
Get PDFThis research explores Tesla\u27s challenges in convincing consumers to transition from internal combustion engine vehicles (ICEVs) to electric vehicles (EVs). It employs a mixed methodology, combining primary and secondary research, to evaluate Tesla\u27s strategies in addressing these barriers. Key findings highlight that while Tesla leads in EV innovation, challenges remain in battery technology, charging infrastructure, affordability, and consumer perception. The study recommends further advancements in battery technology, adoption of innovative solutions like Vehicle-to-Grid (V2G) systems, expansion of charging networks, offering additional incentives, and developing cost-effective EV models to enhance affordability and market acceptance
Navigating Opportunities and Challenges: Exploring Immigrant Work Experiences in the Irish Labor Market.
Get PDFThis study compares the experiences of European and non-European immigrants in Ireland\u27s job market. Prior to doing this study, a research knowledge gap was identified. There is limited qualitative study comparing the obstacles and restrictions that European and non-European immigrants face while entering the Irish labor market. This study used a qualitative research design and included a survey, as well as in-depth interviews with four full-time working immigrants in Ireland. The study included two male and two female volunteers between the ages of 26 and 31. Two boys came to Ireland from Italy, while the girls came from Mexico and Venezuela. The study found that immigrants from Latin American encountered higher impediments to work in Ireland compared to European immigrants. Visa limitations hindered career advancement and made non-European immigrants vulnerable to exploitation in the job market
Stock Market Predictions with Machine Learning .
Get PDFThe focus of this project is developing a tool that can be used in conjunction with other methods to help an investor and/or financial analyst in making an informed decision when making an investment choice taking into consideration stock data.
The project has three main goals which are to try predicting buy and sell signals with the use of a classification model, and to predict the approximate value for next day’s closing price of a stock of our choice with the use of a regression model.
My last goal with this project is to create models easy to use and understand, and that can provide useful predictions and insights to the end user
Deep Learning Model Compression for Resource-Constrained Environments.
Get PDFThis study examines the effects of three Deep Neural Network compression techniques—Quantisation, Pruning, and Weight Sharing/Clustering—on CNN and ANN models trained for image classification tasks. The models were tested on the CIFAR-10 dataset for multiclass classification and a binary classification task using a dataset derived from COCO. The best validation accuracy achieved was 74.7% with a CNN on CIFAR-10 and 53% with the best ANN. On the COCO dataset, a modified CIFAR-10 CNN model achieved 75%. The models were compressed using the three techniques and benchmarked on a ThinkPad laptop and Raspberry Pi 3B+ based on metrics relevant for resource-constrained Internet of Things (IoT) applications, including accuracy, energy consumption, inference speed, and model file size.
The results revealed that weight sharing increased model file size and reduced throughput by up to 26x but significantly lowered energy consumption by 15.6%. Pruning and Quantisation preserved CNN accuracy while reducing model size by up to 81%, although quantisation increased inference time by an average of 3.5x on the Raspberry Pi. The study highlights how these compression techniques affect model performance and offers insights for deploying deep learning models in resource-limited environments
Evaluation and Implementation of Machine Learning Models to Predict Customer Churn in the Telecommunications Sector.
Get PDFThis research addresses customer churn in the Telecom industry by utilizing Machine Learning (ML) models to predict customers at risk of leaving and provide data-driven retention strategies. The study highlights the effectiveness of ML, particularly in churn prediction, while noting the need for further exploration into the ethical implications of AI, such as potential biases towards vulnerable groups. Using the CRISP-DM framework, the study develops and compares three Supervised Learning (SL) models: Random Forests (RF), LightGBM (LGBM), and XGBoost (XGB), incorporating class resampling techniques to manage data imbalance.
The findings identified five key features as the most significant predictors of churn at Viatel Technology Group (VTG), including customer billing, service retention efforts, and product offerings. Among the models tested, LGBM-SMOTETomek delivered the best performance with a precision of 97.92%, recall of 95.25%, and an F1-score of 96.57%. The research concludes with recommendations to promote automatic payment methods, reward loyal customers, and proactively engage with customers who frequently contact the company
Responsible Natural Language Processing to aid Employee Performance Reviews.
Get PDFThis research explores the use of Natural Language Processing (NLP) techniques in assessing evaluators\u27 written appraisals during Employee Performance Reviews (EPRs), aiming to address biases inherent in traditional methods. By integrating Responsible Artificial Intelligence (AI) and foundational Large Language Models (LLMs), the study seeks to enhance the objectivity, fairness, and ethical transparency of performance evaluations. It highlights the potential of AI systems to ensure comprehensive assessments while promoting trust, ethical standards, and employee retention.
The research also aims to advance the field of AI Ethics in practical Human Resources Management (HRM) applications, particularly through NLP-driven tools. These tools are designed to support HR professionals by creating equitable workplace environments, ensuring fair treatment, and fostering inclusivity. The study underscores its contribution to knowledge on responsible AI implementation in HR practices and inspires further developments in ethical performance evaluation systems
Accelerating the Transition - Understanding and Addressing Barriers to Consumer Adoption of Tesla Electric Vehicles.
Get PDFThis research investigates the key influences on adoption of electric vehicles (EVs): focusing on consumer perceptions, advances in technology and government incentives characteristics through Tesla. The study looks at the challenges to EV acceptance, such as charging infrastructure, cost and range anxiety through both qualitative and quantitative means along with potential motivators for adoption such as environmental benefits, financial incentives, technology advances. Consumer awareness, government policies and brand influence are identified having a High effect on EV adoption whereas work-place incentives have an Intermediate effect with high priority to Next. Key challenges are limited charging infrastructure and high upfront costs, however increasing incentives combined with technological innovation provide opportunities to facilitate rapid market development. Suggestions range from the enlargement of infrastructure and better government-supported incentives to targeted information campaigns. This research suggests that to encourage the new vehicle purchasing decision of electric vehicles, it is necessary not only countering practical but also psychological barriers in a holistic approach supporting potential buyers contributing toward more sustainable transport and environmental targets