A material could be harmful to the environment if it does not decay for a long period of time. Regulations such as European REACH are also established to ensure the products from manufacturers have information about biodegradability. However, not all the products have the biodegradation information since testing it all requires time and efforts. QSARs (Quantitative Structure-Activity Relationships) model is encouragingly used to predict the classification of biodegradation for a chemical compound. QSARs model could be extracted from experimental result or by material design simulation. In this study, artificial neural network (ANN) and support vector machine (SVM) model were built to predict ready-biodegradation of a chemical compound. The model was built based on dataset that had been used by Mansouri et al (Kamel Mansouri et al, 2013) on their study. Dataset consisted of 1055 chemical compounds and 41 molecular descriptions for each chemical compound. During building machine learning model, dataset were randomly separated into 791 data for training dataset and 294 data for testing dataset. Scaler was used to standardize the data and since the data consists of multi features, it was reduced by applying 4 principal component analysis (PCA). ANN was built based on many layer perceptrons, several activation functions and solvers were used to find the optimum model. SVM was also built with optimum parameters. The result was the models could classify the biodegradability very well and have quite high accuracy prediction.

1.
Boethling
,
R. S.
“Designing Biodegradable Chemicals. In
Designing Safer Chemicals
DeVito
,
S. C.
,
Garrett
,
R. L.
, Eds.,
American Chemical Society
:
Washington, DC
,
1996
; vol.
640
, pp
156
171
.
2.
European Commission. Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC
.
Off. J. Eur. Union
2006
, L396,
1
849
.
3.
Loonen
,
H.
 et al “
Prediction of biodegradability from chemical structure: Modeling of ready biodegradation test data
.”
Environ. Toxicol. Chem.
1999
,
18
,
1763
1768
.
4.
Tunkel
,
J.
,
Howard
,
P. H.
,
Boethling
,
R. S.
,
Stiteler
,
W.
,
Loonen
,
H.
Predicting ready biodegradability in the Japanese Ministry of International Trade and Industry test
.”
Environ. Toxicol. Chem.
2000
,
19
,
2478
2485
.
5.
Cheng
,
F.
 et al “
In silico assessment of chemical biodegradability
.”
J. Chem. Inf. Model.
2012
,
52
,
655
669
.
6.
Davide
Ballabio
,
Fabrizio
Biganzoli
,
Roberto
Todeschini
&
Viviana
Consonni
. “
Qualitative consensus of QSAR ready biodegradability predictions
”,
2017
.
Toxicological & Environmental Chemistry
,
99
:
7-8
,
1193
1216
.
7.
Kamel
Mansouri
,
Tine
Ringsted
,
Davide
Ballabio
,
Roberto
Todeschini
,
Viviana
Consonni
. “
Quantitative Structure−Activity Relationship Models for Ready Biodegradability of Chemicals
.”
J. Chem. Inf. Model.
2013
,
53
, p.
867
878
.
8.
Dua
,
D.
and
Karra
Taniskidou
, E.
UCI Machine Learning Repository [
http://archive.ics.uci.edu/ml].
2017
Irvine, CA
:
University of California, School of Information and Computer Science
.
9.
Zurada
,
J. M.
Introduction to artificial neural systems
(Vol.
8
).”
1992
,
Minnesota, USA
:
West Publishing Co
.
10.
Ebrahimi
,
E.
,
Monjezi
,
M.
,
Khalesi
,
M. R.
, &
Armaghani
,
D. J.
Bulletin of Engineering Geology and the Environment
”,
2016
,
75
(
1
),
27
.
11.
Cortes
,
C.
;
Vapnik
,
V.
Support-Vector Networks
. In
Machine Learning
1995
,
273
297
12.
Andreas C.
Müller
,
Sarah
Guido
. “Introduction to Machine Learning with Python”.
2017
,
USA
:
O’Reilly Media, Inc
.
13.
Sriparna
Saha
,
Asif
Ekbal
, “
Combining multiple classifiers using vote based classifier ensemble technique for named entity recognition
”.
2013
.
Data & Knowledge Engineering
85
,
15
39
14.
Pedregosa
 et al, “Scikit-learn:Machine Learning in Python”.
JMLR
12
, pp.
2825
2830
,
2011
.
This content is only available via PDF.
You do not currently have access to this content.