Prediction of physical properties of thermosetting resin by using machine learning and structural formulas of raw materials

MRS Advances ◽  
2020 ◽  
Vol 5 (29-30) ◽  
pp. 1567-1575
Author(s):  
Kokin Nakajin ◽  
Takuya Minami ◽  
Masaaki Kawata ◽  
Toshio Fujita ◽  
Katsumi Murofushi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Physical Properties ◽  
Functional Groups ◽  
Raw Materials ◽  
Functional Materials ◽  
Industrial Applications ◽  
Chemical Information ◽  
Thermosetting Resins ◽  
Mixing Ratios

AbstractThermosetting resins are one of the most widely used functional materials in industrial applications. Although some of the physical properties of thermosetting resins are controlled by changing the functional groups of the raw materials or adjusting their mixing ratios, it was conventionally challenging to construct machine learning (ML) models, which include both mixing ratio and chemical information such as functional groups. To overcome this problem, we propose a machine learning approach based on extended circular fingerprint (ECFP) in this study. First, we predicted the classification of raw materials by the random forest, where ECFP was used as the explanatory variable. Then, we aggregated ECFP for each classification predicted by the random forest. After that, we constructed the prediction model by using the aggregated ECFP, feature quantities of reaction intermediates, and curing conditions of resin as explanatory variables. As a result, the model was able to predict in high accuracy (R^2 = 0.8), for example, the elastic modulus of thermosetting resins. Furthermore, we also show the result of verification of prediction accuracy in first step, such as using the one-hot-encording. Therefore, we confirmed that the properties of thermosetting resins could be predicted using mixed raw materials by the proposed method.

Industrial Applications of Near-Infrared Chemical Imaging Microscopy

2001 ◽  
Vol 7 (S2) ◽  
pp. 162-163
Author(s):  
EN Lewis ◽  
LH Kidder ◽  
KS Haber
Keyword(s):  
Spatial Distribution ◽  
Near Infrared ◽  
Imaging System ◽  
Raw Materials ◽  
Single Point ◽  
Nir Spectroscopy ◽  
Chemical Imaging ◽  
Industrial Applications ◽  
Chemical Information ◽  
Spatially Resolved

Single point near-infrared (NIR) spectroscopy is used extensively for characterizing raw materials and finished products in a wide variety of industries: polymers, paper, film, pharmaceuticals, paintings and coatings, food and beverages, agricultural products. As advanced industrial materials become more complex, their functionality is often determined by the spatial distribution of their discrete sample constituents. However, conventional single point NIR spectroscopy cannot adequately probe the interrelationship between the spatial distribution of sample components with the physical properties of the sample. to fully characterize these samples, it is necessary to probe simultaneously spatial and chemical heterogeneity and correlate these properties with sample characteristics.Recently, we have developed a novel NIR imaging spectrometer that can deliver spatially resolved chemical information very rapidly. in contrast to conventional, single point NIR spectrometers, the imaging system uses an infrared focal-plane array (FPA) to collect up to 76,800 complete spectra, one for each pixel on the array, in approximately one minute.

scholarly journals Analysing the Capability of the Catchment's Spectral Signature for the Regionalization of Hydrological Parameters

2021 ◽  
Author(s):  
Laura Fragoso-Campón ◽  
Pablo Durán-Barroso ◽  
Elia Rosado
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Physical Properties ◽  
Spectral Response ◽  
Spectral Signature ◽  
Spectral Approach ◽  
Hydrological Response ◽  
Random Forest Algorithm ◽  
Hydrological Parameters ◽  
Climatic Environment

Water resource management in ungauged catchments is complex due to the uncertainties around the hydrological parameters that dominate the streamflow behaviour. These parameters are usually defined by regionalization approaches in which hydrological response patterns are transferred from gauged to ungauged basins. Regression-based methods using physical properties derived from cartographic data sources are widely used. The current remote sensing techniques offer us new standpoints in regionalisation processing since the hydrological response depends on the physical attributes related to the spectral responses of the territory. Moreover, machine learning approaches have not been specifically applied to the regionalization of hydrologic parameters. This work studies the capability of a catchment’s spectral response based on Sentinel-1 and Sentinel-2 data to address a regression-based regionalization of hydrological parameters using a machine learning approach. Hydrological modelling was conducted by the HBV-light model. We tested the random forest algorithm in several regionalization scenarios: the new approach using the catchments’ spectral signature, the traditional method using physical properties and a fusion of them. The calibration results were excellent (median KGE = 0.83), and the regionalized parameters obtained with the random forest algorithm achieved good performance in which the three scenarios showed almost the same goodness of fit (median KGE = 0.45 to 0.50). We found that the effectiveness depends on the climatic environment and that predictions in humid catchments exhibited better performance than those in the driest catchments. The physical approach (median KGE= 0.71) exhibited better performance than the spectral approach (median KGE= 0.64) in humid catchments, whereas spectral regionalization (median KGE= 0.33) outperformed the physical scenario in the driest catchments (median KGE= 0.25). Herein, our results confirm that regionalization is still challenging in Mediterranean climate variants where the new spectral approach showed promising results and time series of satellite data could improve seasonal regionalization methodologies.

High Resolution Electron Microscopy of internal interfaces in layered materials

1990 ◽  
Vol 48 (4) ◽  
pp. 320-321
Author(s):  
Yoichi Ishida ◽  
Hideki Ichinose ◽  
Yutaka Takahashi ◽  
Jin-yeh Wang
Keyword(s):  
Grain Boundaries ◽  
Mirror Symmetry ◽  
Functional Materials ◽  
High Resolution Electron Microscopy ◽  
Layered Materials ◽  
Plane Boundary ◽  
Chemical Information ◽  
Layer Plane ◽  
High Tc ◽  
Cubic Metals

Layered materials draw attention in recent years in response to the world-wide drive to discover new functional materials. High-Tc superconducting oxide is one example. Internal interfaces in such layered materials differ significantly from those of cubic metals. They are often parallel to the layer of the neighboring crystals in sintered samples(layer plane boundary), while periodically ordered interfaces with the two neighboring crystals in mirror symmetry to each other are relatively rare. Consequently, the atomistic features of the interface differ significantly from those of cubic metals. In this paper grain boundaries in sintered high-Tc superconducting oxides, joined interfaces between engineering ceramics with metals, and polytype interfaces in vapor-deposited bicrystal are examined to collect atomic information of the interfaces in layered materials. The analysis proved that they are not neccessarily more complicated than that of simple grain boundaries in cubic metals. The interfaces are majorly layer plane type which is parallel to the compound layer. Secondly, chemical information is often available, which helps the interpretation of the interface atomic structure.

Random Forest Refinement of Pairwise Potentials for Protein-ligand Decoy Detection

2019 ◽  
Author(s):  
Jun Pei ◽  
Zheng Zheng ◽  
Hyunji Kim ◽  
Lin Song ◽  
Sarah Walworth ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Probability Function ◽  
Pair Potential ◽  
Scoring Function ◽  
Stable Structure ◽  
Scoring Functions ◽  
Atom Pair ◽  
Data Set ◽  
Atom Pairs

An accurate scoring function is expected to correctly select the most stable structure from a set of pose candidates. One can hypothesize that a scoring function’s ability to identify the most stable structure might be improved by emphasizing the most relevant atom pairwise interactions. However, it is hard to evaluate the relevant importance for each atom pair using traditional means. With the introduction of machine learning methods, it has become possible to determine the relative importance for each atom pair present in a scoring function. In this work, we use the Random Forest (RF) method to refine a pair potential developed by our laboratory (GARF6) by identifying relevant atom pairs that optimize the performance of the potential on our given task. Our goal is to construct a machine learning (ML) model that can accurately differentiate the native ligand binding pose from candidate poses using a potential refined by RF optimization. We successfully constructed RF models on an unbalanced data set with the ‘comparison’ concept and, the resultant RF models were tested on CASF-2013.5 In a comparison of the performance of our RF models against 29 scoring functions, we found our models outperformed the other scoring functions in predicting the native pose. In addition, we used two artificial designed potential models to address the importance of the GARF potential in the RF models: (1) a scrambled probability function set, which was obtained by mixing up atom pairs and probability functions in GARF, and (2) a uniform probability function set, which share the same peak positions with GARF but have fixed peak heights. The results of accuracy comparison from RF models based on the scrambled, uniform, and original GARF potential clearly showed that the peak positions in the GARF potential are important while the well depths are not. <br>

ALUMINUM 1100

Alloy Digest ◽  
1974 ◽  
Vol 23 (2) ◽  
Keyword(s):  
Corrosion Resistance ◽  
Shear Strength ◽  
High Temperature ◽  
Physical Properties ◽  
Tensile Properties ◽  
Pure Aluminum ◽  
Heat Treating ◽  
Industrial Applications ◽  
Temperature Performance ◽  
Commercially Pure

Abstract ALUMINUM 1100 is commercially pure aluminum and is characterized by its excellent ability to be drawn, spun, stamped or forged. It has good weldability, excellent resistance to corrosion and many home, architectural and industrial applications. This datasheet provides information on composition, physical properties, hardness, elasticity, tensile properties, and shear strength as well as fatigue. It also includes information on low and high temperature performance, and corrosion resistance as well as forming, heat treating, machining, and joining. Filing Code: Al-44. Producer or source: Various aluminum companies. Originally published October 1956, revised February 1974.

KUBOTA HK40 and HK50

Alloy Digest ◽  
2007 ◽  
Vol 56 (10) ◽  
Keyword(s):  
High Temperature ◽  
Physical Properties ◽  
Carbon Content ◽  
Tensile Properties ◽  
Oxidation Resistance ◽  
Heat Treating ◽  
Industrial Applications ◽  
High Temperature Strength ◽  
Ni Alloys ◽  
Moderately High Temperature

Abstract Kubota alloys HK40 and HK50 are austenitic Fe-Cr-Ni alloys that have been standard heat-resistant materials for more than four decades. With moderately high temperature strength, oxidation resistance, and carburization resistance the alloys are used in a wide variety of industrial applications. HK 50 has slightly higher carbon content. This datasheet provides information on composition, physical properties, and tensile properties as well as creep. It also includes information on casting, heat treating, machining, and joining. Filing Code: SS-998. Producer or source: Kubota Metal Corporation, Fahramet Division.

BÖHLER W403 VMR

Alloy Digest ◽  
2013 ◽  
Vol 62 (9) ◽  
Keyword(s):  
Heat Treatment ◽  
Chemical Composition ◽  
Physical Properties ◽  
Tool Steel ◽  
Raw Materials ◽  
Heat Treating ◽  
Diffusion Annealing ◽  
Selection Of

Abstract Böhler (or Boehler) W403 VMR is a tool steel with outstanding properties, based not only on a modified chemical composition, but on the selection of highly clean raw materials for melting, remelting under vacuum (VMF), optimized diffusion annealing, and a special heat treatment. This datasheet provides information on composition, physical properties, and elasticity. It also includes information on forming and heat treating. Filing Code: TS-721. Producer or source: Böhler Edelstahl GmbH.

A Study on Host Tropism Determinants of Influenza Virus Using Machine Learning

2020 ◽  
Vol 15 (2) ◽  
pp. 121-134 ◽  
Author(s):  
Eunmi Kwon ◽  
Myeongji Cho ◽  
Hayeon Kim ◽  
Hyeon S. Son
Keyword(s):  
Machine Learning ◽  
Amino Acids ◽  
Influenza Virus ◽  
Random Forest ◽  
Physicochemical Properties ◽  
Protein Sequences ◽  
Influenza Viruses ◽  
Host Tropism ◽  
Post Hoc ◽  
Ha Protein

Background: The host tropism determinants of influenza virus, which cause changes in the host range and increase the likelihood of interaction with specific hosts, are critical for understanding the infection and propagation of the virus in diverse host species. Methods: Six types of protein sequences of influenza viral strains isolated from three classes of hosts (avian, human, and swine) were obtained. Random forest, naïve Bayes classification, and knearest neighbor algorithms were used for host classification. The Java language was used for sequence analysis programming and identifying host-specific position markers. Results: A machine learning technique was explored to derive the physicochemical properties of amino acids used in host classification and prediction. HA protein was found to play the most important role in determining host tropism of the influenza virus, and the random forest method yielded the highest accuracy in host prediction. Conserved amino acids that exhibited host-specific differences were also selected and verified, and they were found to be useful position markers for host classification. Finally, ANOVA analysis and post-hoc testing revealed that the physicochemical properties of amino acids, comprising protein sequences combined with position markers, differed significantly among hosts. Conclusion: The host tropism determinants and position markers described in this study can be used in related research to classify, identify, and predict the hosts of influenza viruses that are currently susceptible or likely to be infected in the future.

Development of Prediction Models Using Machine Learning Algorithms for Girls with Suspected Central Precocious Puberty: Retrospective Study (Preprint)

2018 ◽  
Author(s):  
Liyan Pan ◽  
Guangjian Liu ◽  
Xiaojian Mao ◽  
Huixian Li ◽  
Jiexin Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Retrospective Study ◽  
Random Forest ◽  
Precocious Puberty ◽  
Prediction Models ◽  
Central Precocious Puberty ◽  
Machine Learning Algorithms ◽  
Stimulation Test ◽  
Gnrh Analogue ◽  
Prediction Probability

BACKGROUND Central precocious puberty (CPP) in girls seriously affects their physical and mental development in childhood. The method of diagnosis—gonadotropin-releasing hormone (GnRH)–stimulation test or GnRH analogue (GnRHa)–stimulation test—is expensive and makes patients uncomfortable due to the need for repeated blood sampling. OBJECTIVE We aimed to combine multiple CPP–related features and construct machine learning models to predict response to the GnRHa-stimulation test. METHODS In this retrospective study, we analyzed clinical and laboratory data of 1757 girls who underwent a GnRHa test in order to develop XGBoost and random forest classifiers for prediction of response to the GnRHa test. The local interpretable model-agnostic explanations (LIME) algorithm was used with the black-box classifiers to increase their interpretability. We measured sensitivity, specificity, and area under receiver operating characteristic (AUC) of the models. RESULTS Both the XGBoost and random forest models achieved good performance in distinguishing between positive and negative responses, with the AUC ranging from 0.88 to 0.90, sensitivity ranging from 77.91% to 77.94%, and specificity ranging from 84.32% to 87.66%. Basal serum luteinizing hormone, follicle-stimulating hormone, and insulin-like growth factor-I levels were found to be the three most important factors. In the interpretable models of LIME, the abovementioned variables made high contributions to the prediction probability. CONCLUSIONS The prediction models we developed can help diagnose CPP and may be used as a prescreening tool before the GnRHa-stimulation test.

Document Preprocessing with TF-IDF to Improve the Polarity Classification Performance of Unstructured Sentiment Analysis

2020 ◽  
pp. 235-242
Author(s):  
Farrikh Alzami ◽  
Erika Devi Udayanti ◽  
Dwi Puji Prabowo ◽  
Rama Aria Megantara
Keyword(s):  
Machine Learning ◽  
Feature Extraction ◽  
Random Forest ◽  
Sentiment Analysis ◽  
Classification Performance ◽  
Document Preparation ◽  
Learning Models ◽  
Polarity Classification ◽  
Negative Sentiment ◽  
Machine Learning Models

Sentiment analysis in terms of polarity classification is very important in everyday life, with the existence of polarity, many people can find out whether the respected document has positive or negative sentiment so that it can help in choosing and making decisions. Sentiment analysis usually done manually. Therefore, an automatic sentiment analysis classification process is needed. However, it is rare to find studies that discuss extraction features and which learning models are suitable for unstructured sentiment analysis types with the Amazon food review case. This research explores some extraction features such as Word Bags, TF-IDF, Word2Vector, as well as a combination of TF-IDF and Word2Vector with several machine learning models such as Random Forest, SVM, KNN and Naïve Bayes to find out a combination of feature extraction and learning models that can help add variety to the analysis of polarity sentiments. By assisting with document preparation such as html tags and punctuation and special characters, using snowball stemming, TF-IDF results obtained with SVM are suitable for obtaining a polarity classification in unstructured sentiment analysis for the case of Amazon food review with a performance result of 87,3 percent.

Sign in / Sign up

Export Citation Format

Share Document