Scribd
Upload
7 views7 pages

BA - Presentation GRP 2

Uploaded by

mostakahamedt
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
7 views7 pages

BA - Presentation GRP 2

Uploaded by

mostakahamedt
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Developing a Predictive model To Predict the Obesity level of Dhaka City Using KNIME Analytics

Platform Using various Obesity Determinants

Prepared by
Prepared for
Group: 02
Mr. Swarup Saha
Md. Shaporan Sujan (Roll: 201-65-060)
Assistant Professor
Mostak Ahamed Tanmoy (Roll: 201-65-071)

Institute of Business Administration (IBA)


University of Dhaka
Objectives

 To use KNIME to create a prediction model that can be used to reliably forecast a customer's level of

obesity based on a variety of dietary, lifestyle, and demographic parameters.

 To evaluate how well the predictive model performs in delivering individualized health advice based on

each person's level of obesity.

 To investigate the viability and consequences of combining the predictive model with focused marketing

initiatives to tackle public health issues and boost business profits.


Scope

 Our long-term strategy involves integrating with healthcare systems to enable early intervention and

conducting study on genetic risk factors as our studies advance.

 The healthcare industry can utilize it for timely intervention, stimulating the creation of preventive

services and focused wellness initiatives.

 The food and beverage business can utilize the model for product creation, whilst the fitness industry

can generate customized programs.

 The findings can be used by public health policy for specific treatments, and the pharmaceutical

industry can employ them for the creation of medications.


Variables Used

High Calorific food Family History of


Gender Monitor calories
consumption Overweight

Vegetable consumption
Age Smoking Physical exercise
Frequency

Daily main meals Daily device usage


Height (m) Alcohol intake
frequency duration
Independent
Variable
Between-meal food Mode of
Weight (Kg) Daily water intake
consumption frequency transportation

Dependent Variable Obesity Level


Feature Selection tool

Decision Tree Random Forest Naïve Bayes

No. of Features Accuracy No. of Features Accuracy No. of Features Accuracy

11 15 15

( Referred as less ( Referred as less ( Referred as less


relevant features: relevant features: relevant features:
I. Age, I. Monitor I. Daily water
II. Height, Calories ) intake)
III. weight, 0.566 0.673 0.629
IV. High calorific
food
consumption,
V. Mode of
transportation)
Predictive Models Used:

Random forest Predictor


Decision Tree
• Combines multiple decision trees
• Appropriate for studying big dataset. and predicts accurately.
• Assumes independence between • Gives confidence level of prediction
predictors. • Less likely to overfit data.
• Resilient to unimportant traits.

Prediction of Obesity
Naïve Bayes Predictor
• Appropriate for studying
big dataset
• Assumes independence
between predictors
• Resilient to unimportant
traits
Result

Methods Used Prediction Accuracy Comment

Decision Tree
47.8%
Almost similar accuracy
Random Forest obtained for all the three
49.4%
models
Naive Bayes
45.8%

You might also like