Conference PaperPDF Available

A new approach to body fat percentage prediction using linear regression

Authors:

Abstract and Figures

Increased body fat percentage or obesity is a metabolic disorder that occurs through an increase in energy intake and a decrease in energy expenditure. Different criteria and methods have been proposed to calculate body fat percentage from various sources, but with the astonishing advancements in computer science, particularly artificial intelligence and machine learning systems, these intelligent systems can be utilized in important fields such as medicine. In this study, an intelligent learning model has been designed that can predict individuals' body fat percentage using other parameters. The intelligent model performs prediction using a linear regression algorithm, which is based on 14 other parameters or features of individuals. The proposed model in this study has been trained using a dataset consisting of 252 samples and 15 numerical features. It is worth mentioning that as an innovation in this study, the Principal Component Analysis (PCA) method or algorithm has been used to reduce the dimensions by 0.95 components. By introducing this innovation in the modeling process, the accuracy of the model has significantly improved while reducing the computational load and time complexity of the model. The evaluation of the model is performed using the coefficient of determination or R2 Score, which indicates how well the linear regression model performs the prediction process. The evaluation results for R2 Score on the proposed model in this study are recorded as 0.969161 for training data and 0.979227 for test data.
Content may be subject to copyright.






m.payandeh@qodsiau.ac.ir










Principal Component Analysis PCA

R2 Score













[1]




[4]



























     

 
  
  

      


     



  BIA  
[1]



    
Apriori Matlab    
[2]


Levenberg-MarquardtBayesian Regularization  

Bayesian Regularization
Levenberg-Marquardt
     

L.MB.R
[3]





 


   
[4]




        
  SVM      
[5]
  

 
SKF




]6[


New System Solver
Bayes  
IVNBBMI
IV 
[7]
    
Bias Error Control Term
  

Wilcoxon
[8]

VIKOR (VMFET)
[9]


  
      










Y
X
  



Wr
Foar
Bice
Ank
Kn
Th
Hi
Abd
Ch
Ne
He
We
Ag
BDP
D
18.2
28.6
32.2
23.1
38.5
59.4
99.9
92.5
100.8
37.9
70.1
178.9
44.8
19.15
1.05
mean
0.93
2.02
3.02
1.69
2.41
5.24
7.16
10.78
8.43
2.43
3.66
29.38
12.60
8.36
0.01
std
15.8
21
24.80
19.1
33.00
47.20
85
69.40
79.30
31.10
29.50
118.5
22.00
00.00
0.99
min
17.6
27.3
30.20
22
36.97
56.00
95.5
84.57
94.35
36.40
68.25
159
35.75
12.47
1.04
%25
18.3
28.7
32.05
22.8
38.50
59.00
99.3
90.95
99.65
38
70
176
43.00
19.20
1.05
%50
18.8
30
34.32
24
39.92
62.35
103.5
99.32
105.3
39.42
72.25
197
54.00
25.30
1.07
%75
21.4
34.9
45.00
33.9
49.10
8730
147.7
148.1
136.2
51.20
77.75
363.1
81.00
47.50
1.10
max

0=
min
BDP
 
kg/m³lb
inchcm


      
Heatmap

[-1,+1]






MinMaxScaler
[0,1]
   
YX



Sample
Dataset
252
X.shape
252
Y.shape







Random_State = 0 











Sample
Dataset
176
Train.X
76
Test.X




      
  

       

󰇛 󰇜 󰇛
󰇜󰇛
󰇜
󰇛
󰇜
󰇛
󰇜
󰇛 󰇜

         


[-1,+1]




PCA

Fit





  

󰇛
󰇜 

󰇛
󰇜
󰇛
󰇜

 0.969162Score =
2
R0.979228Score =
2
R
        
     




         

Relu




   


Score
2
Train R
0.899473
NN
0.969162
Proposed Method




     


 
 Regulize     



[10]
CSV

[11]
Python
NumpyPandasScipySeabornScikit-learnMatplotlib

¬



1. 

2. 

3. Levenberg-
MarquardtBayesian Regularization

4. 

5. 
svm
6. Zamri, N.B.A., Bhuvaneswari, T., Aziz, N.A.B.A. and Aziz, N.H.B.A., 2018, July. Feature selection using
simulated Kalman filter (SKF) for prediction of body fat percentage. In Proceedings of the 2018 1st
International Conference on Mathematics and Statistics (pp. 23-27).
7. Mazalan, N.M., 2017. Prediction of body fat status by using naïve bayes technique among university students
(Doctoral dissertation, Universiti Teknologi MARA).
8. Chiong, R., Fan, Z., Hu, Z. and Chiong, F., 2021. Using an improved relative error support vector machine
for body fat prediction. Computer Methods and Programs in Biomedicine, 198, p.105749.
9. Lai, C.M., Chiu, C.C., Shih, Y.C. and Huang, H.P., 2022. A hybrid feature selection algorithm using simplified
swarm optimization for body fat prediction. Computer Methods and Programs in Biomedicine, 226, p.107183.
10. https://www.kaggle.com/datasets/fedesoriano/body-fat-prediction-dataset
11. https://github.com/milad71payandeh/Body-Fat-Prediction

1 Body Mass Index (BMI)
2 Body Fat Percentage
3 Machine Learning
4 Deep Learning
5 Dataset
6 Time Complexity
7 Principal Component Analysis (PCA)


8 Dimension Reduction
9 Linear Regression
10 Data Mining
11 Data Recovery
12 Big Data
13 Database Management
14 Multi Layer Perceptron (MLP)
15 Apriori
16 Support Vector Machine (SVM)
17 Simulated Kalman Filter (SKF)
18 Particle Swarm Optimization (PSO)
19 Metaheuristic
20 Naïve Bayes
21 Density
22 Label
23 Range
24 Train Set
25 Test Set
26 Correlation Rate
27 Pearson Correlation Coefficient
28 R2 Score
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.