Approach to Sustainable Development Scenario using Machine Learning Algorithms
Approach to Sustainable Development Scenario using Machine Learning Algorithms
Volume 9, Issue 12, December – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14613873
Approach to Sustainable Development
Scenario using Machine Learning Algorithms
Tarangini Mukherjee
Graduate, National Institute of Fashion Technology, Kolkata
Abstract:- The world is evolving at a fast pace but the Material Factors
attention to resources has been neglected by mankind for We are defining term called material factors so that this
several decades. It is high time we aim at development can be used as variable in our machine learning
which is sustainable and good for the present and future algorithms.The material factors are Energy Efficiency,
generation. In this paper we will be exploring how machine Biodegradable Index,Recycle factor, Water wastage and
learning algorithms can help us choose the best materials, Carbon Factor . These material factors are inspired from Urban
approach ,style, process, design for Sustainable systems[2]
development.
III. MACHINE LEARNING ALGORITHMS
I. INTRODUCTION
The machine learning algorithms we will be using are
When we see a curtain , pillow cover , cushion cover etc Supervised ,Unsupervised and Semi Supervised learning.
we appreciate the colours , design and pattern. But what if we Supervised learning methods include Decision Trees(DT),
get all this with also keeping the process sustainable?This is Naive Bayesian Classifier (NB), Linear Discriminant
what we will be exploring in this paper using machine learning Analysis(LDA), K-nearest Neighbours(KNN), Support vector
algorithms[1]. Also tuning and modern optimisation machine(SVM) and Neural networks(NN). Unsupervised
techniques will be explored in this paper. learning methods include Gaussian Mixture Model(GMM),
Hidden Markov Model(HMM) and Principal Component
II. MATERIAL SCIENCE Analysis(PCA).Semi Supervised learning consists of
Generative Models(GM), Graph based methods (GB)and Self
Commodities like pillows, cushion cover, curtains all Training methods(ST)
look the best nowadays but what about the world . Factors like
energy efficiency , recycle factor , polymer used, After using these algorithms with our material factors
biodegradability need to be considered while looking the best. that are Energy Efficiency(EE), Biodegradable
Index(BI),Recycle factor(RF), Water wastage(WW) and
Carbon Factor(CF) we found the following impact
factor(<100) of each of the variables
Table 1 The Importance of Material factors for Different Machine Learning Algorithms
Algorithm EE impact BI impact RF impact WW impact CF impact
DT 1.2 77.44 88.22 44.3 33.3
NB 1.3 77.6 75.22 22.321 2.22
LDA 0.2 55.53 83.22 34.43 3.4
KNN 4.4 54.24 54.3 44.22 4.4
SVM 4.4 42.242 23.22 23.2 33.2
NN 2.1 33.3 63.22 33.342 33.3
GMM 7.9 2.3 0.212 41.44 98.11
HMM 17.3 3.3 0.11334 99.33 98.11
PCA 33.32 3.3233 0.1414 22.432 23.31
GM 2.233 2.232 3.44 34.333 23.1
GB 66.64 4.12 4.4373 52.11 33.1
ST 33.3 3.41 2.1224 24.33 12.2
From the above we table it can be observed that While Hidden Markov Model gave high points to
Supervised learning considered (BI) and (RF) to be important (WW)factor, Gaussian Mixture Model considered (CF) factor
for sustainability. While Unsupervised learning considered as the most important. (BI) was considered important when it
(EE) as important factor and Semi Supervised learning came to Decision Trees and the Naive Bayesian Classifier.
considered (WW) as important factor for the process of Neural networks gave the least priority which is to (EE).
sustainability.
IJISRT24DEC2050 www.ijisrt.com 2619
Volume 9, Issue 12, December – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14613873
IV. OPTIMIZATION METHODS [3]. Xinhui Yue; Yuan-Hua Ni;“Accelerated Distributed
Composite Nesterov Gradient Descent
Optimization methods are (in the sense of Machine Algorithm,’’2022 41st Chinese Control Conference
Learning Algorithms) ways that make our algorithm better (CCC),2022,pp. 1-5
which in our case will help in sustainability [4]. Halyna Padalko; Kseniia Bazilevych; Mykola
development.Common methods that suit are use case best are Butkevych; “Heart Failure Development Prediction
Nesterov accelerated gradient[3], using Stochastic Gradient Descent Optimization,”2022
IEEE 9th International Conference on Problems of
Stochastic gradient descent[4], Alternating Direction Infocommunications, Science and Technology (PIC
Method of Multipliers[5] and Newtons’ method[6]. S&T),2022, pp 297-300
[5]. Shiyou Yan; Min Yang; “Alternating direction method
V. CONCLUSION of multipliers with variable stepsize for partially
parallel MR image reconstruction,’’2017 36th Chinese
Thus we have successfully implemented machine Control Conference (CCC),2017, pp 10886 - 10889
learning algorithms for sustainable production and studied [6]. Andrey Pazderin; Sergey Yuferev,“Power flow
how factors such as Carbon factor, Biodegradable Index, calculation by combination of Newton-Raphson
Energy Efficiency, Recycle factor and Water wastage effect method and Newton's method in optimization,’’2009
the process of Sustainable development. Sustainable 35th Annual Conference of IEEE Industrial
production is need of the hour and as humans we must all Electronics,2009, pp 1693 - 1696
contribute in different ways to the cause so that our future
generations can live peacefully.
MACHINE LEARNING OVERVIEW
Machine Learning is the subset of Artificial Intelligence
which allows machines to learn and show improvement from
experience without being programmed.It uses algorithms to
identify different patterns, analysing data and making
decisions
Machine learning algorithms works by feeding them
large amounts of data, use algorithms to learn from the data
and improve performance over time and by using insights they
have learned to make informed decisions
Different Types of Machine Learning are :
Supervised learning : Algorithms build a mathematical
model of data that contains both the inputs and desired
outputs
Unsupervised learning : Algorithms find the structures in
data that has not been labeled
Semi Supervised learning : Falls between supervised and
unsupervised learning.
Weakly Supervised Learning : Training labels are noisy,
limited or imprecise
REFERENCES
[1]. Roopam Gupta; Thomas Rincy N; “A Survey on
Machine Learning Approaches and Its
Techniques,’’2020 IEEE International Students'
Conference on Electrical,Electronics and Computer
Science (SCEECS), 2020, pp. 1-6
[2]. Mikko Koho; Seppo Torvinen; Alexandre Torres
Romiguer; “Objectives, enablers and challenges of
sustainable development and sustainable
manufacturing: Views and opinions of Spanish
companies,’’ 2011 IEEE International Symposium on
Assembly and Manufacturing (ISAM),2011, pp. 1-5
IJISRT24DEC2050 www.ijisrt.com 2620
You might also like
- 100% (6)Download (Ebook) Soft Computing in Materials Development and its Sustainability in the Manufacturing Sector (Edge AI in Future Computing) by Amar Patnaik (editor), Vikas Kukshal (editor), Pankaj Agarwal (editor), Ankush Sharma (editor), Mahavir Choudhary (editor) ISBN 9780367723583, 0367723581 ebook All Chapters PDF70 pages
