Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 23

National Institute of Science & Technology Final Project Presentation

“PREDICTIVE MODELING FOR DIABETES MELLITUS USING


MACHINE LERNING IN BIG DATA”
PRESENTED BY: SANGAM BEHERA
ROLL NO-202250067
BU regd. No. 23020/2015

UNDER THE GUIDANCE OF


DR. MANJUSHREE NAYAK
DEPARTMENT OF COMPUTER SCIENCE
NATIONAL INSTITUTE OF SCIENCE AND TECHNOLOGY

07/08/2024 NAME :- SANGAM BEHERA 202250067 1


National Institute of Science & Technology Final Project Presentation

CONTENTS:-
 OBJECTIVE
 INTRODUCTION
 PROBLEM STATEMENT
 MACHINE LEARNING WORKFLOW
 DATASET DETAILS
 ALGORITHMS
 ACCURACY RESULTS
 CONCLUSION
 FUTURE WORK
 REFERENCES

07/08/2024 NAME :- SANGAM BEHERA 202250067 2


National Institute of Science & Technology Final Project Presentation
OBJECTIVE:-
 Early Detection:
Objective: To identify individuals at high risk of developing diabetes as early as possible.

 Improved Diagnostic Accuracy:


Objective: To enhance the accuracy of diabetes diagnosis using advanced ML algorithms.

 Personalized Healthcare:
Objective: To tailor prevention and treatment strategies to the individual characteristics and
risk profiles of patients.

07/08/2024 NAME :- SANGAM BEHERA 202250067 3


National Institute of Science & Technology Final Project Presentation

INTRODUCTION:

 Diabetes prediction using machine learning involves leveraging advanced algorithms to analyze
patient data and identify individuals at high risk of developing diabetes.

 This approach aims to enable early detection, improve diagnostic accuracy, personalize treatment
plans, and optimize healthcare resources.

 By integrating vast amounts of health data, including genetic, lifestyle, and clinical factors,
machine learning models can provide timely and accurate predictions, facilitating proactive
management and prevention of diabetes.

07/08/2024 NAME :- SANGAM BEHERA 202250067 4


Final Project Presentation
PROBLEM STATEMENT:-
National Institute of Science & Technology

 The rising prevalence of diabetes poses significant public health challenges, with late
diagnoses often missing critical early intervention opportunities.

 Traditional diagnostic methods are insufficient for early detection and precise risk
assessment.

 This project aims to develop a machine learning model to predict an individual's risk of
developing diabetes using their medical history, lifestyle, and other relevant data.

 The goals are to identify high-risk individuals early, improve prediction accuracy,
provide personalized risk assessments, and optimize healthcare resource allocation,
ultimately enhancing patient outcomes and reducing the healthcare burden associated
with diabetes.

07/08/2024 NAME :- SANGAM BEHERA 202250067 5


National Institute of Science & Technology Final Project Presentation
MACHINE LEARNING WORKFLOW:-
we can define the machine learning workflow in 5 stages

 Gathering data

 Data pre-processing

 Researching the model that will be best for the type of data

 Training and testing the model

 Evaluation

07/08/2024 NAME :- SANGAM BEHERA 202250067 6


National Institute of Science & Technology Final Project Presentation

( Machine learning workflow)

07/08/2024 NAME :- SANGAM BEHERA 202250067 7


National Institute of Science & Technology Final Project Presentation

( Training and testing process)

07/08/2024 NAME :- SANGAM BEHERA 202250067 8


National Institute of Science & Technology Final Project Presentation
MACHINE LEARNING:-

 Machine learning (ML) is an area of artificial intelligence (AI) that gives computers the
capacity to automatically learn from data and past experiences to find patterns and
make predictions with little to no human involvement .

 “Learning is any process by which a system improves performance from experience.”


- Herbert Simon

07/08/2024 NAME :- SANGAM BEHERA 202250067 9


National Institute of Science & Technology Final Project Presentation
TYPES OF MACHINE LEARNING:-

07/08/2024 NAME :- SANGAM BEHERA 202250067 10


National Institute of Science & Technology Final Project Presentation

(Supervised machine learning)

07/08/2024 NAME :- SANGAM BEHERA 202250067 11


National Institute of Science & Technology Final Project Presentation

(Working of unsupervised learning)

07/08/2024 NAME :- SANGAM BEHERA 202250067 12


National Institute of Science & Technology Final Project Presentation
DATASET DETAILS:

07/08/2024 NAME :- SANGAM BEHERA 202250067 13


National Institute of Science & Technology Final Project Presentation
ALGORITHM:

Random forest classifier:


 A Random Forest classifier is an
ensemble learning method that improves
prediction accuracy and robustness by
constructing multiple decision trees
during training.

 Each tree is built from a random subset


of the data and features, enhancing
diversity among the trees.

 The final prediction is made by


aggregating the outputs of all the trees,
typically through majority voting for
classification or averaging for
regression.
07/08/2024 NAME :- SANGAM BEHERA 202250067 14
National Institute of Science & Technology Final Project Presentation
DECISION TREE :-

 A decision tree is a versatile and


intuitive machine learning algorithm
used for both classification and
regression tasks.
 It operates by recursively splitting the
dataset into subsets based on the value
of input features, forming a tree-like
structure where each internal node
represents a feature, each branch
represents a decision rule, and each
leaf node represents an outcome.
 Decision trees are easy to visualize and
interpret, making them useful for
understanding complex data
relationships.
07/08/2024 NAME :- SANGAM BEHERA 202250067 15
National Institute of Science & Technology Final Project Presentation

LOGISTIC REGRESSION :-

 In machine learning, logistic


regression stands as a fundamental
yet powerful tool for binary
classification tasks. Unlike its name
suggests, logistic regression is a
statistical method used to model the
probability of a binary outcome
based on one or more predictor
variables.

 This relationship is then


transformed using the logistic
function, also known as the sigmoid
function, to yield probabilities
between 0 and 1.
07/08/2024 NAME :- SANGAM BEHERA 202250067 16
National Institute of Science & Technology Final Project Presentation

ACCURACY RESULTS:

07/08/2024 NAME :- SANGAM BEHERA 202250067 17


National Institute of Science & Technology Final Project Presentation

MACHINE LEARNING BASED DIABETES CLASSIFICATION

07/08/2024 NAME :- SANGAM BEHERA 202250067 18


National Institute of Science & Technology Final Project Presentation

CONCLUSION:-
 The application of machine learning techniques for diabetes prediction holds immense promise in
revolutionizing healthcare practices.

 By leveraging advanced algorithms and vast datasets encompassing genetic, lifestyle, and clinical
factors, these models can provide early detection, improve diagnostic accuracy, and offer
personalized risk assessments for individuals at risk of developing diabetes.

 Through proactive intervention and tailored treatment plans, the burden of diabetes-related
complications can be significantly reduced, leading to improved patient outcomes and optimized
allocation of healthcare resources.

07/08/2024 NAME :- SANGAM BEHERA 202250067 19


National Institute of Science & Technology Final Project Presentation

FUTURE SCOPE:-
 Integration with Wearable Devices and IoT
 Genomic and Multi-Omics Data Integration
 Explainable AI (XAI) for Enhanced Interpretability
 Population-Level Risk Stratification and Public Health Interventions

07/08/2024 NAME :- SANGAM BEHERA 202250067 20


National Institute of Science & Technology Final Project Presentation

ACKNOWEDGEMENT:

I would like to thanks:

 NIST
 Dr. Bishnukar Nayak(Msc principal)
 Mr,Sumanta Kumar Pattnaik (Course co-ordinator)
 Dr. Brojo Kishore Mishra(H.O.D)
 Dr. Manjushree Nayak(Guide & project Co-Ordinator)
 All respected sir & madam
 Friends

07/08/2024 NAME :- SANGAM BEHERA 202250067 21


National Institute of Science & Technology Final Project Presentation

REFERENCES

 Solanki P, Baldaniya D, Jogani D, Chaudhary B, Shah M, Kshirsagar A (2021) Artificial


intelligence: new age of transformation in petroleum upstream. Pet Res (in press).

 Ali, R.; Hardie, R.C.; Narayanan, B.N.; Kebede, T.M. IMNets: Deep Learning Using an
Incremental Modular Network Synthesis Approach for Medical Imaging Applications. Appl.
Sci. 2022, 12, 5500.

 Jena, P.K.; Khuntia, B.; Palai, C.; Nayak, M.; Mishra, T.K.; Mohanty, S.N. A Novel
Approach for Diabetic Retinopathy Screening Using Asymmetric Deep Learning Features.
Big Data Cogn. Comput. 2023, 7, 25.

07/08/2024 NAME :- SANGAM BEHERA 202250067 22


National Institute of Science & Technology Final Project Presentation

07/08/2024 NAME :- SANGAM BEHERA 202250067 23

You might also like