[vc_empty_space][vc_empty_space]
Study on building machine learning model to predict biodegradable-ready materials
Putra R.I.D.a, Maulana A.L.a, Saputro A.G.a
a Computational Material Design Laboratory, Engineering Physics Research Group, Faculty of Industrial Technology, Institut Teknologi Bandung, Bandung, Indonesia
[vc_row][vc_column][vc_row_inner][vc_column_inner][vc_separator css=”.vc_custom_1624529070653{padding-top: 30px !important;padding-bottom: 30px !important;}”][/vc_column_inner][/vc_row_inner][vc_row_inner layout=”boxed”][vc_column_inner width=”3/4″ css=”.vc_custom_1624695412187{border-right-width: 1px !important;border-right-color: #dddddd !important;border-right-style: solid !important;border-radius: 1px !important;}”][vc_empty_space][megatron_heading title=”Abstract” size=”size-sm” text_align=”text-left”][vc_column_text]© 2019 American Institute of Physics Inc. All rights reserved.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.[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Author keywords” size=”size-sm” text_align=”text-left”][vc_column_text][/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Indexed keywords” size=”size-sm” text_align=”text-left”][vc_column_text][/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Funding details” size=”size-sm” text_align=”text-left”][vc_column_text]This work is supported by Directorate of Higher Education, Ministry of Research Technology Education, Republic of Indonesia under grant scheme “Penelitian Unggulan Perguruan Tinggi 2018”[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”DOI” size=”size-sm” text_align=”text-left”][vc_column_text]https://doi.org/10.1063/1.5095351[/vc_column_text][/vc_column_inner][vc_column_inner width=”1/4″][vc_column_text]Widget Plumx[/vc_column_text][/vc_column_inner][/vc_row_inner][/vc_column][/vc_row][vc_row][vc_column][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][/vc_column][/vc_row]