A Capstone Project Report
A Capstone Project Report
A Capstone Project Report
Learning”
CAP-769
SUBMITTED BY
Submitted to
Dated:
Declaration by the supervisor
Signature of Supervisor
Acknowledgement
The successful completion of this research paper would not have been possible without the invaluable
support and contributions of several individuals and institutions. We would like to express our sincere
gratitude to all of them.
First and foremost, We are deeply grateful to our esteemed supervisor, Ms. Ranjit Kaur, Assit. Professor
at Lovely Professional University. Her profound knowledge in the field of Machine Learning proved
instrumental in guiding us throughout this journey. Ms. Kaur's insightful suggestions, constructive
criticism, and unwavering encouragement helped us refine our research question, develop a robust
methodology, and navigate the complexities of the research process. Her dedication to our academic
growth and her belief in our abilities instilled in us the confidence to persevere through challenges
We would also like to extend our sincere appreciation to Lovely Professional University for providing the
perfect environment for academic exploration and research. The university's exceptional library resources,
including access to a vast collection of scholarly articles and databases, were invaluable for gathering
relevant information and strengthening the foundation of my research. Additionally, the supportive faculty
and staff were always available to answer my questions and offer guidance whenever needed.
ABSTRACT
Liver diseases represent a significant public health challenge worldwide, with diverse etiologies
and complex pathophysiological mechanisms. Early detection and accurate prediction of liver
disease progression are crucial for guiding clinical decision-making and improving patient
outcomes. In this research paper, we propose the development and evaluation of a comprehensive
liver disease prediction model leveraging machine learning algorithms and multiomic data
integration. The research aims to address existing gaps in predictive modeling for liver diseases,
including limited integration of multiomic data, sparse representation of rare disease subtypes, and
challenges in model interpretability and generalizability. The proposed methodology involves data
collection from publicly available sources, feature selection and engineering, model development
and evaluation using algorithms such as Logistic Regression, Random Forest, K-Nearest
Neighbors (KNN), Decision Tree, and Support Vector Machines (SVM), and validation on
independent cohorts of liver disease patients. Furthermore, the research seeks to enhance model
interpretability and explainability through feature importance analysis and visualization
techniques, and translate the findings into clinical practice through collaboration with healthcare
providers and stakeholders. By bridging the gap between data science and clinical medicine, this
research aims to advance the field of liver disease prediction and contribute to improved diagnosis,
prognosis, and personalized management of liver diseases.
CONTENTS
ABSTRACT i
LIST OF FIGURES 10
LIST OF TABLES 10
CHAPTER – 1: INTRODUCTION 12
1.1 Introduction
List of Tables
Serial Table no Caption Page no
no
1 Table 3.1 Information about Data 19-20
2 Table 4.1 Model Evaluation 33