3830 research outputs found
Sort by
Enhancing Financial Inclusion in Rural India through Fintech Application
No full textThis research investigates the dynamic landscape of financial inclusion in rural India through the lens of Financial Technology (FinTech). Utilizing a comprehensive approach, the study explores the impact of FinTech applications on bridging the financial gap in underserved communities. Through surveys and regression analyses, the research examines demographic factors, attitudes, and challenges influencing the adoption of FinTech services. The findings highlight a positive inclination towards FinTech adoption, with respondents expressing faith in its potential to enhance financial inclusion and increase household income. Language barriers, limited access to technology, and security concerns emerge as significant challenges. The proposed solution introduces a hypothetical multilingual FinTech app to address language obstacles, emphasizing inclusivity. The study contributes novel insights into the intersection of technology, financial inclusion, and socio-economic dynamics in rural India, offering valuable recommendations for policymakers, businesses, and FinTech developers to foster a more inclusive and sustainable financial ecosystem
Traditional Machine Learning Algorithms and Deep Learning for ODI Cricket Prediction
No full textA comparative study between traditional machine learning and a deep neural network approach is presented for predicting winning teams in for One Day International (ODI) cricket games. Data is extracted from the espncricinfo website covering the years 1971 to 2022 for model training. Features include team performance and match conditions. Model performance is evaluated on 2023 match results. Both small (2010–2022) and large datasets (1971-2022) are used for training for comparative purposes. The deep neural ANN achieves an accuracy of 85.4%, outperforming the conventional techniques including ensemble techniques such as random forests and gradient boosting. The deep neural ANN model is shown to outperform in identifying nuances and intricate patterns, demonstrating an ability to use large amounts of historical data to increase accuracy. This study builds upon earlier work to add significant insights to improve ODI cricket result predictions
Exploring determinants of consumer trust in brands' sustainability claims
No full textThe aim of this study is to synthesise research of green marketing practices of sustainable brands, trust as a concept, and theoretical frameworks used to assess the influencing of consumer decision making as a purchasing decision. The research was conducted in a qualitative framework with participants engaging in semi-structured interviews. Reflective Thematic Analysis was used for data analysis. Themes arose related to the idea of both trust builders and trust barriers for a consumer in a brand’s sustainable claims. Brand transparency, brand relationship with consumer, and brand track record emerged as trust building influences, while brand negativity and brand incomplete efforts emerged as negative influences on consumer trust. To positively influence trust in their sustainable claims, brands should approach sustainable initiatives with all facets of the brand, participate in transparent communications about their processes and ensure a positive track record of success in achieving their claim
Mental Note
No full textAmidst a digital era's heightened focus on mental health, the proliferation of mental health apps underscores a growing need for accessible support. However, concerns are raised pertaining to their efficacy and potential for replacing professional intervention. This project introduces Mental Note, a web-based application, as a potential solution to this issue. This application can serve as a repository for mental health professionals' exercises and guidance, but also offers common mental health app features such as mood tracking and journaling. This report outlines the development of this application, including the system requirements, design and scope, as well as the theoretical foundation leading to the development of this application
Malware Detection Using Deep Learning
Full text linkSeveral models were implemented and evaluated in this work on deep learning-based malware detection. These models comprised Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Graph Sample and Aggregated (GraphSAGE), Graph Convolutional Network (GCN), Hyper-parameter GCN, Graph Isomorphism Network, and Graph Attention Network. The Graph Isomorphism Network has the highest accuracy (97.58%), while the other models also had varied degrees of accuracy. The paper promotes cybersecurity by providing a comparative analysis of multiple deep-learning models for malware detection. The outcomes of this study may affect the development of more reliable and accurate malware detection systems. These findings indicate the promise of specialized graph-based neural networks, particularly Graph Isomorphism Networks. This work increases our understanding of deep learning applications in cybersecurity and stresses the need to select appropriate models for certain tasks, paving the way for more powerful malware detection tools
The influence of TikTok on consumer behaviour: An analysis of how TikTok influencers affect Generation Z’ consumption in the wellness sector
Full text linkIn this study, the researchers deeply examine the dynamics of social media’s influence on Generation Z’s consumer behaviour. The research question, “To what extent are Generation Z’s purchase behaviour for wellness products influenced by TikTok digital influencers?” aims to measure the degree to which this influence occurs by applying quantitative research methods.
Two main objectives guided the research: a) To conduct a comprehensive Literature Review, analysing published work on Gen Z, their consumption of wellness products, and the dynamics of influence on the social media platform TikTok among content creators, consumers and brands, and b) To conduct an online survey, exploring how TikTok influencers and their posts promoting wellness products and brands influence Gen Z consumers’ purchasing behaviour towards wellness products.
By applying a descriptive research design and qualitative survey approach with a sample of 131 participants, of which mostly were Gen Z, the researchers confirmed both hypotheses: 1) TikTok influences Gen Z consumers’ purchasing intentions in the wellness niche and 2) TikTok influencers directly influence Gen Z consumers’ purchasing decision process. From the sample population, 85.3% affirm that they have had the intention to purchase wellness products in the past, based on a TikTok influencer’s recommendations. In contrast, 81.5% of participants said they had already purchased a wellness product based on the recommendation of a TikTok influencer
Unraveling Emotions in Lyrics: A Novel Approach to Enhance Spotify Music Recommendations
Full text linkMusic recommendation systems play a pivotal role in user engagement, satisfaction, and retention on streaming platforms like Spotify. However, traditional methods often fall short of providing diverse and emotionally resonant song suggestions, leading to repetitive playlists and user dissatisfaction. This research therefore delves into the unexplored area of emotion detection within song lyrics to enhance personalized music recommendations on Spotify. The study investigates various machine learning models, including Logistic Regression, Support Vector Machines, Bidirectional LSTM, and DistilBERT, to understand the intricate emotional hints within song lyrics. Through a comparative evaluation of these models, the research identifies Bidirectional LSTM as the most effective, achieving an accuracy of 92%, followed by Random Forest at 82%, and Support Vector Machines and Decision Tree both at 70%. Additionally, the research examines hybrid recommender systems combining case-based k Nearest Neighbours and content-based filtering to offer users nuanced and emotionally connecting song recommendations. The project seeks to optimize music discovery, boost user engagement, foster industry innovation, and ensure a more inclusive representation of artists and genres. Ultimately, the research aspires to introduce a novel perspective to music recommendation systems, one that authentically resonates with user emotions, preferences, and satisfaction
An AI Design of Robot Communication using Natural Language Processing (NLP)
No full textThis study presents a novel approach to robot communication through the lens of Natural Language Processing (NLP), leveraging the capabilities of the BartForConditionalGeneration model from Hugging Face's transformers. The research's core objective is to engineer an AI framework, designated as 'nqa_dbs_model' (Natural Question-Answer Dublin Business School Model), that excels in interpreting and responding to queries using diverse datasets. The model's training involved a dual dataset strategy: firstly, a General Knowledge dataset sourced from Google's Natural Questions, and secondly, a bespoke dataset focused on Dublin Business School (DBS), compiled through meticulous web scraping. In assessing the model's efficacy, a comparative analysis was conducted against OpenAI's gpt-3.5-turbo model. The outcomes reveal a notable proficiency of the 'nqa_dbs_model' in addressing specific organizational queries related to DBS, a feature not mirrored in the gpt-3.5-turbo model. This distinction highlights the 'nqa_dbs_model’s' advanced capability in generating context-specific responses, thereby enhancing the scope of AI in targeted communication scenarios. The conclusion drawn from this research underscores the importance of specialized dataset training in elevating the potential of AI models within the realm of robotic communication, offering a pathway to more nuanced and contextually relevant interactions in AI-assisted environments
Comparative study of traditional vs. transformer machine learning algorithms for Inflammatory Bowel Disease (IBD) Medical Report Classification
No full textInflammatory Bowel Disease (IBD) is a serious chronic condition that affects millions worldwide, yet it remains under-researched, especially in the application of machine learning. This research aims to draw attention to IBD and support the efforts of My Chron's and Colitis Angel (MyCCAngel), a non-profit organization, by developing a medical report classification model to determine whether users of their platform have IBD. The study compares traditional machine learning models, including Multi-layer Perceptron, Support Vector Machines, and Naïve Bayes, with a transformer-based model, Bidirectional Encoder Representations from Transformers (BERT).
MyCCAngel offers a social platform specifically designed for IBD patients, providing them with tools and assistance to manage their daily challenges. Transformer models, such as BERT, represent a recent advancement in machine learning, applying an evolving set of mathematical techniques (Merritt, 2022). This research seeks to evaluate whether these newer models outperform traditional methods in the classification of medical reports.
In this study, a quantitative research approach is employed, relying on data collected from a substantial sample size. This method allows for the identification of patterns and trends within the data, providing a more scalable and objective analysis. By leveraging a dataset, the study aims to draw conclusions that are applicable beyond the immediate sample, enhancing the reliability and applicability of the findings.
In this study, a quantitative research approach is employed, involving the manual collection of 110 medical reports, both IBD and non-IBD, due to the unavailability of an existing dataset. This method allows for the identification of patterns and trends within the data, providing a more scalable and objective analysis. By using dataset, the study aims to draw conclusions that are applicable.
The performance of each model was evaluated based on metrics such as accuracy, precision, recall, and F1-score. The findings indicate that traditional machine learning algorithms, particularly Naïve Bayes, outperform the transformer-based model BERT, achieving an accuracy of 91% compared to BERT's 68%.
This study demonstrates that transformer models are not always superior and that traditional simple models like Naïve Bayes can offer better performance in specific tasks, such as IBD medical report classification. Furthermore, this research is the first to focus on the classification of IBD medical reports, providing valuable insights for future binary classification tasks in the medical field