Journal of Management and Service Science (JMSS)

Utilising Exploratory Data Analysis and Machine Learning Algorithms for Heart Disease Analysis and Prediction

Humra Khan — 2024-04-25

As one of the most common and potentially fatal diseases in the world, heart disease must be detected early for proper treatment. With exploratory data analysis (EDA) and machine learning algorithms for predictive analysis, this research project seeks to thoroughly investigate the different aspects that contribute to heart disease. This will enable prompt diagnosis and risk mitigation. Numerous crucial features affecting the diagnosis of heart disease have been found through in-depth exploratory analysis of data. Among these features, the number of major arteries stained by fluoroscopy, the various forms of chest pain, the maximum heart rate reached, exercise-induced angina, the slope of the peak exercise ST segment, and the ST depression brought on by activity relative to rest stand out as most significant factors. Clinicians can learn a great deal about a patient's risk of developing heart disease by carefully examining these characteristics. In order to put this research's predictive component into practice, machine learning classifiers are built using the UCI heart disease dataset, which contains important variables pertaining to cardiac health. For comparison analysis, six different methods are used: Random Forest (RF), Gradient Boost (GB), K-Nearest Neighbour (KNN), Decision Tree (DT), Support Vector Machine (SVM), and Logistic Regression (LR). After conducting a comprehensive analysis, it has been determined that the Random Forest classifier has the best accuracy rate, attaining a remarkable 85.25%.

Movie Recommender System: Collaborative Filtering Methods

Abhishek Kumar Rai — 2024-04-25

In the current digital era, personalized movie recommendation systems have become essential tools for aiding users in navigating the vast array of available cinematic content. This paper offers an indepth examination of the development and implementation of a Python-based movie recommender system, integrating content-based and collaborative filtering methodologies. Utilizing a comprehensive dataset encompassing diverse movie attributes and audience feedback metrics, our research aims to construct a resilient recommendation engine capable of delivering customized movie suggestions to users. Through a meticulous analysis and evaluation of various similarity metrics, including Pearson Correlation Coefficient, Spearman Rank Correlation Coefficient, Mean-Squared Distance, and Cosine Similarity, we assess the system's effectiveness in generating personalized recommendations. Our findings provide valuable insights into the strengths and weaknesses of each technique, thereby guiding future research and enhancements in movie recommendation systems.

A Hybrid Approach to Movie Recommendation System

Abhay Yadav — 2024-04-25

Recommendation Systems (RS) have become indispensable in today's digital ecosystem, influencing decisions and experiences across several platforms. This paper delves deeply on RS, including its history, functionality, and relevance. It covers many aspects of RS, such as Content-Based and Collaborative Filtering, using examples from a variety of industries, including e-commerce and entertainment. The paper also describes and covers empirical analysis methodologies for comparing RS efficacy and providing a framework. By conducting a qualitative and quantitative analysis, compared these three recommendation systems i.e. content based collaborative and Hybrid. This mixed analysis approach was necessary as Content-Based Filtering systems are not easily quantifiable, and for a movie recommendation system, the qualitative aspect holds significant importance. Through our analysis, it became evident that a hybrid recommendation system consistently outperforms standalone methods in terms of recommendation accuracy and relevance.

Identification and Detection of Driver Drowsiness using Machine Learning Techniques

Mohammad Faisal — 2024-04-25

Many people die or are injured in car accidents. According to the World Health Organization, one million people die from traffic accidents every year worldwide. Drowsy, unrested, or drowsy drivers are drowsy drivers who put themselves and other road users at risk. Studies on car accidents show that major train accidents are caused by driver fatigue. In recent years, it has been revealed that driving while drowsy can cause fatigue. Nowadays, the cause of accidents while climbing is hunger. This situation is a serious problem all over the world and needs to be solved as soon as possible. In recent years, driver hijacking has become one of the leading causes of traffic accidents that can lead to death, serious bodily injury, and accidents. Economic losses and disasters. Driver fatigue can be caused by long hours of driving, drowsiness, fatigue, medications, sleep disorders, and illness. An analysis of various studies shows that there is a need for reliable technology that can detect drowsy driving and warn drivers before an accident occurs. Many studies have been conducted to improve the diagnosis and prediction of drowsy driving using different scales to assess drowsy driving. This study identified several measures categorized by the researchers as physiological, automatic, mental, and behavioral measures. This article deals with the main issues of different sleep detection methods and how to use them to detect drowsiness while driving. To warn drivers before they crash, the analysis focuses on what happens while driving, and as technology advances, it is designed to ideally identify and predict drowsy drivers. This comprehensive review provides a better understanding for researchers conducting fundamental evaluations in a field.

Promoting Digital Transformation in Times of Crisis: A Comparative Analysis of Technology Entrepreneurship in the US and Chinese Payment Sectors After COVID-19

Walid Ghodbane — 2024-04-25

The COVID-19 pandemic has significantly impacted the global payments industry, accelerating the shift to alternative digital payment methods. This comparative study examines the effects of the pandemic on technology entrepreneurship and payment innovation in the United States and China. Semi-structured interviews were conducted with 25 payment entrepreneurs across the two countries to gain insights into drivers of innovation, challenges faced, and the role of government policies. A survey of 200 consumers provided perspectives on changing payment preferences and adoption patterns pre- and post-COVID. Results indicate COVID-19 accelerated interest in contactless and mobile payments due to attributes like convenience and security. However, startups faced barriers accessing funding and navigating regulations. While policies promoting innovation, competition and digitization supported the ecosystem, effectiveness varied. Not all businesses, especially small- and medium-sized enterprises, overcame resource and expertise constraints limiting technology adoption. This research enhances understanding of payment industry transformations during crises. It informs policy approaches balancing supportive initiatives and addressing uneven diffusion. Continued efforts fostering innovative entrepreneurship models and accessible digital infrastructure can strengthen pandemic resilience and economic recovery.

Intelligent Chat Bot

Aamir Shahab — 2024-04-25

This research examines how intelligent systems have the potential to transform hu-man-computer interaction, with a special emphasis on the emergence of chat bots. These intelligent systems are an attempt to change many aspects of user involve-ment in response to the rapid advancements in artificial intelligence. The study explores the development and deployment of an intelligent Chatbot, a virtual assistant that can comprehend and reply to user inquiries. Improving user interactions in vari-ous applications is the main goal. The study offers insights into the approach used, difficulties handled, and wider implications for the future of interactive technology by describing the development and implementation of this Chat bot. The study repre-sents a major advancement in the field of AI-driven human-computer interaction and highlights the growing importance of intelligent Chat bots in enhancing user experi-ences.

Empowering Forest Conservation: Harnessing Indigenous Knowledge within Southern Ethiopia’s Forest-Dependent Communities

Mando Choto Doyo — 2024-04-25

This study was carried out in Kercha district, West Guji Zone, Oromia Regional State, Ethiopia, with the overarching goal of investigating Empowering Forest Conservation: Harnessing Indigenous Knowledge within Southern Ethiopia's Forest-Dependent Communities. The study used a qualitative research design with a descriptive technique due to the nature of the stated research problem, which was to explore people's experiences with conserving forests and behaviors. In this study 146 sample households were chosen from 2224 households through multi-stage selection approaches. Both primary and secondary data were utilised. The primary data were gathered through structured questionnaires, key informant interviews, and focus group discussions This study found that the community used distinct indigenous knowledge in various activities. Farmers who illegally cut down forests believed that God cursed them, and Gada elders accused them of isolating from the society. Particularly trees culturally regarded among community like 'Wodessa' and religious respected trees like Oda also certain trees which community used for blessing. The belief and rules governing local communities to replace trees used for other domestic consumption or other purposes, as well as the usage of fire outside of their homes, are governed by societal beliefs and customs. Local communities across different agro-ecological zones have more experience planting trees before taking them down for a specific purpose. Aside from the best indigenous forest conservation practices and the presence of modern religion and culture, the expansion/increased demand of agricultural land market economies or livelihood systems, as well as investment activities, have an impact on indigenous forest conservation practice in studied areas. Based on the study's findings, the researchers advocate that community, government, and policymakers integrate modern forest conservation with traditional forest conservation without compromising the culture and beliefs of the local community for forest conservation.

Cybersecurity: An In-Depth Analytical Review

Kushagra Srivastava — 2024-04-25

This study highlights cybersecurity's core elements and their function in protecting digital systems and data while also providing an overview of the technology's fundamental operation. The research places a focus on the necessity of a proactive and layered approach to defense as well as the significance of cybersecurity in defending organizations from cyber threats. Various security measures are covered, such as network security and access controls. In order to foster a security-conscious culture, the research also looks at the importance of user awareness and training. Organiza-tions can build a strong foundation for risk management, preventing unauthorized access, and guaranteeing the honesty, reliability, and accessibility of their critical in increasingly interconnected and threat-prone digital landscape by understanding the fundamental functionality of cybersecurity.