Conference Presentations by MARTIN MUDUVA
IEOM Society International, 2024
HIV and AIDS remain prominent global health concerns, and antiretroviral medication (ART) plays a... more HIV and AIDS remain prominent global health concerns, and antiretroviral medication (ART) plays a crucial role in treating infected individuals, preventing disease progression, and improving overall health outcomes. However, missed appointments in ART programs pose significant challenges by causing treatment interruptions, unsuppressed viral load, and increased HIV transmission rates. This research employs the CRISP-DM methodology and aims to develop a predictive model that effectively reduces missed appointments among people living with HIV. A comprehensive analysis of patient data, including demographics, clinical information, and appointment history, was conducted to determine the key factors influencing missed appointments. The prediction model was created using machine learning techniques such as decision trees, random forests, and support vector machines. It was determined that random forest produced the best results, having lower square errors and greater R squared. The findings contribute to the advancement of predictive analytics in healthcare, particularly in the context of chronic conditions such as HIV/AIDS.
IEOM Society International, 2024
In drought-prone African countries like Zimbabwe, the uptake of parametric insurance has been low... more In drought-prone African countries like Zimbabwe, the uptake of parametric insurance has been low due to the absence of localized models. Guided by the CRISP-DM model, the present study proposes an AI-based approach to drought prediction in parametric insurance. The study's paramount objectives are establishing trigger thresholds for drought events, assessing their significance, identifying the most effective machine learning models for drought modeling based on the Standardized Precipitation Index (SPI), and forecasting future drought occurrences and their magnitudes. Historical weather data, including temperature and rainfall, are utilized and a range of machine learning models-neural networks, random forest, and support vector machines are employed for drought prediction. The performance of these models is evaluated based on accuracy, reliability, and interpretability, with continuous refinement based on feedback from stakeholders. The significance of this research lies in promoting data-driven decisions, incentivizing preparedness, enabling risk transfer, facilitating rapid insurance payouts, and enhancing financial stability. With accurate drought predictions driving parametric insurance, policyholders can make wellinformed choices, adopt proactive measures, transfer the risk of drought-related losses, receive swift insurance payouts, and improve their financial resilience during drought events.
IEOM Society International, 2024
Diabetes is recognized as one of the world's most prevalent health problems. As diabetic patients... more Diabetes is recognized as one of the world's most prevalent health problems. As diabetic patients grew, so did the percentage of diabetic hospital readmissions. Early readmissions can impact patient well-being, operational efficiency, and financial burden. This study uses machine learning approaches to predict hospital readmissions among diabetes patients. Data was collected from 130 US hospitals. CRISP-DM is used for analysis. Logistic regression (LR) and random forest (RF) classifiers were implemented. The classifier performance was compared. Random Forest outperformed the other model, with an accuracy of 0.89. The model was chosen to enable practical deployments. Researchers used a web-based interface to get data and receive real-time predictions. The results showed that the predictive model used alongside an interface creates a clear and understandable prediction platform. However, the research might involve various datasets and Deep Learning to improve models and findings, in future studies. Furthermore, the model could explore the integration of machine learning interpretability approaches to increase transparency and promote better comprehension of the model's predictions by healthcare practitioners.
IEOM Society International, 2024
Mental health is an important aspect of well-being as it encompasses emotional, psychological and... more Mental health is an important aspect of well-being as it encompasses emotional, psychological and social well-being. The use of patient portals in mental health care has gained attention as a potential tool to improve access to care for individuals with mental illness. Patient portals may be vulnerable to unauthorized access if appropriate security measures are not put in place. This study leverages blockchain technology to create tamper-proof patient records. The proposed solution uses an on-chain database that stores hashes and the actual medical record of a patient as well as an off-chain solution that handles encryption of each user's medical record using their respective keys in a trustless manner before they are uploaded on-chain. A secure smart contract hosted on Ethereum and the Byzantine Fault Tolerance consensus algorithm was used to ensure patient privacy. The research employed the Comparative Analysis Research Methodology as the research methodology and the Kanban methodology as the software development methodology. The research project concludes that the proposed solution addresses the current security issues and data privacy concerns in patient data. The decentralized nature of blockchain ensures security, transparency, and tamper-Proceedings of the International Conference on Industrial Engineering and Operations Management ©IEOM Society International proof storage of information. Further research is needed for future advancements, like integrating blockchain-based patient portals with wearable devices and IoT.
Sciend and Information Conference - Springer, 2024
This research explains the perceptions of university students regarding the use of ChatGPT in the... more This research explains the perceptions of university students regarding the use of ChatGPT in their learning and research. The research is based on evidence provided in articles that were considered in a systematic literature review using the PRISMA model. The motivation for this study was informed by a realization that there are limited studies that provide a comprehensive understanding of students' perceptions regarding the use of ChatGPT in learning and research. Many existing studies either focus on educators' perceptions or isolated cases of empirical studies. It is therefore not clear what factors influence students' adoption of ChatGPT. This gap was addressed by this study by conducting a systematic literature review to identify factors influencing the adoption of ChatGPT, the positive impact of ChatGPT on students' education and the concerns raised by students regarding the integration of ChatGPT into academic practice. The research focus was on articles that discuss the students' perceptions of ChatGPT, all of which were drawn from Google Scholar. In total 32 articles were considered for analysis for this study and in this paper, we present five constructs that influence the adoption of ChatGPT in academic activities. Even though the students' perceptions of ChatGPT are widely varied, the findings of this research show that students concur that ChatGPT has a positive influence on their learning. The students further highlighted concerns relating to their use of ChatGPT in learning, which could be improved if the tool's educational effectiveness is to be realized. The contribution of this study is threefold. First, the findings of this research explain the perceptions of students about using ChatGPT in academic practice. Second, the results of this study inform the university management and policymakers about how to effectively integrate ChatGPT into the education system. Third, the study suggests, recommends, and guides educators who wish to incorporate this new, powerful tool (ChatGPT) into their teaching.
Books by MARTIN MUDUVA
AI-Driven Marketing Research and Data Analytics, 2024
The AI voice assistant mobile application was developed to aid drivers in operating their mobile ... more The AI voice assistant mobile application was developed to aid drivers in operating their mobile phones while driving without touching their phones. The literature review examines multiple innovative artificial technologies involved in applications with voice assistants in natural language processing (NLP) techniques. The methodology used involved a qualitative approach, and the design science paradigm was used for the development of the voice assistant for smartphones with NLP techniques. NLP techniques that were applied in the development of the AI voice assistant are smart synthesis, data flow sequence, core and interface accessing, part of speech tagging, named entity recognition, conference resolution, and porter stemming. Some of the operations that are achieved by the application include arithmetic calculations based on voice commands and returning the computer result via voice, searching the internet based on user voice input, and providing a response via voice assistance.
This chapter introduces a methodological approach to implementing supervised machine learning alg... more This chapter introduces a methodological approach to implementing supervised machine learning algorithms and neuromarketing techniques for predicting customer churn. It addresses the challenge of customer attrition faced by businesses and explores how the combination of neuromarketing strategies with machine learning algorithms can enhance churn forecast accuracy. By conducting a comparative study, the chapter assesses the performance of different algorithms when integrated with various neuromarketing approaches, such as biometric, neuroimaging, eye tracking, neurophysiological and facial expression analysis data. It emphasises the importance of understanding the advantages and disadvantages of different algorithms to select the most suitable methods for churn prediction. The chapter provides an overview of customer churn prediction in the context of neuromarketing, highlighting the connection between customer attrition and neuromarketing. The chapter discusses studies that explore customer relationship characteristics, neuroscience methods for understanding consumer behavior and the significance of emotions in churn prediction. Various neuromarketing techniques, including neuroimaging and physiological measurements are examined for their relevance in predicting churn by uncovering emotional and cognitive processes underlying consumer behavior. Through leveraging neuromarketing insights, marketers can develop predictive models using supervised machine learning algorithms that effectively utilise customer data to accurately predict churn and develop targeted retention strategies. The chapter also explores commonly employed machine learning algorithms such as logistic regression, decision trees, random forests, support vector machines, gradient boosting methods and multilayer perceptron designs, in the context of churn prediction using neuromarketing data. Lastly, the chapter emphasises the importance of evaluation metrics in assessing the performance of predictive models in neuromarketing for customer churn prediction.
Agritourism in Africa, 2023
Agritourism is a dynamic industry poised for growth, with the global market share projected to re... more Agritourism is a dynamic industry poised for growth, with the global market share projected to reach 117.37 billion by 2027. Despite its potential to foster economic growth and human development in Africa, the industry grapples with several obstacles including market inefficiencies, environmental sustainability concerns, and limited access to information. The application of AI in the context of sustainable agritourism remains largely uncharted territory in Africa. Therefore, there is a pressing need to investigate how AI can be effectively harnessed to address these challenges and promote sustainable agritourism, thereby facilitating human development across the African continent. This chapter delves into the potential of artificial intelligence (AI) to facilitate sustainable agritourism in Africa. It begins by providing an overview of the current state of agritourism in Africa, highlighting the challenges it encounters.
Papers by MARTIN MUDUVA
International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2024
Smart cities are comprised of intelligent objects that can collectively and automatically improve... more Smart cities are comprised of intelligent objects that can collectively and automatically improve living standards, preserve lives, and function as a sustainable environment. Drones or Unmanned aerial vehicles (UAV), robotics, cognitive computing, and the Internet of Things (IoT) are mandatory to enhance the intellectual ability of smart cities by enhancing connectivity, energy efficiency, and Quality of Signal (QoS). Consequently, the integration of drones with IoT plays a crucial part in enabling a vast array of smart-city applications. Drones are undeniably the technology of the future. They glide in the air, keeping an eye on things in their metropolis. They do not need human control or operation. They capture data based on visuals and sounds by employing a variety of sensors, webcams, and mics, and then transfer it to a gateway for processing and retrieving information. Drones will let us gather vast volumes of data for processing, while also enhancing the intelligence of smart cities. The drone's signal is vulnerable to absorption, refraction, diffraction, and attenuation. Therefore, it is essential to predict the signal from the drone. This study presents an intelligent method using Deep Learning to predict the signal strength for enhancing network connectivity and delivering the desired QoS of IoTs and drone integration. This enables effective data transmission, boosts QoS for end-users/devices, and reduces data transmission power consumption.
International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2023
Android Social Media Applications have become a yardstick in facilitating a platform for human so... more Android Social Media Applications have become a yardstick in facilitating a platform for human socialization on cyber space. They are an inevitable alternative, which is fast replacing most traditional ways that lacks full multimedia interaction adored by many. These applications are of forensic value as they account for most activities helpful in either incriminating or exonerating suspects in cases of adverse events. By default, most social applications store activity data in specific directories they create at the background of the hosting Android devices. Through expertise, this data can be extracted and analyzed to come up with meaningful insights useful in an inquiry of digital evidence interest. This study focused on forensics of Twitter and Clubhouse android based social media applications. The approach taken was to install these applications on emerging Android devices using the Samsung Galaxy S20+ (SMGS20+) and Samsung Galaxy Tab A7 (SMGTA7), populate known test data, perform data acquisition, execute data analysis noting results and then do a comparative analysis of tools and techniques utilized towards provisioning alternative solutions.
International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2024
Ransomware is a malicious code designed to encrypt and lock personal data such as documents and p... more Ransomware is a malicious code designed to encrypt and lock personal data such as documents and photos, in order to create opportunity for extorting money from the victims. Android operating systems are particularly targeted due to their large market share. Previous studies have primarily relied on signature-based detection methods, which require sufficient data samples and labelled signatures. However, modern ransomware utilizes obfuscation techniques that make it challenging to analyse using static methods. This project proposes a hybrid analysis approach for Android ransomware, employing the SVM algorithm for detection. The novelty lies in the limited exploration of SVM algorithms for ransomware analysis. The dataset used in the study was obtained from CICA and Mal 2017. Static features, including permissions, intents, encoding methods, and API calls were used, along with dynamic features such as network activities and system calls. The SVM model achieved good performance, with 81% accuracy and 90% precision using static features, and 100% accuracy with dynamic features.
Uploads
Conference Presentations by MARTIN MUDUVA
Books by MARTIN MUDUVA
Papers by MARTIN MUDUVA