scholarly journals Machine learning based energy-free structure predictions of molecules, transition states, and solids

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Dominik Lemm ◽  
Guido Falk von Rudorff ◽  
O. Anatole von Lilienfeld
Keyword(s):  
Machine Learning ◽  
Ab Initio ◽  
Chemical Space ◽  
Transition States ◽  
Computational Prediction ◽  
Training Data ◽  
Data Sets ◽  
Prediction Errors ◽  
Optimization Task ◽  
Out Of Sample

AbstractThe computational prediction of atomistic structure is a long-standing problem in physics, chemistry, materials, and biology. Conventionally, force-fields or ab initio methods determine structure through energy minimization, which is either approximate or computationally demanding. This accuracy/cost trade-off prohibits the generation of synthetic big data sets accounting for chemical space with atomistic detail. Exploiting implicit correlations among relaxed structures in training data sets, our machine learning model Graph-To-Structure (G2S) generalizes across compound space in order to infer interatomic distances for out-of-sample compounds, effectively enabling the direct reconstruction of coordinates, and thereby bypassing the conventional energy optimization task. The numerical evidence collected includes 3D coordinate predictions for organic molecules, transition states, and crystalline solids. G2S improves systematically with training set size, reaching mean absolute interatomic distance prediction errors of less than 0.2 Å for less than eight thousand training structures — on par or better than conventional structure generators. Applicability tests of G2S include successful predictions for systems which typically require manual intervention, improved initial guesses for subsequent conventional ab initio based relaxation, and input generation for subsequent use of structure based quantum machine learning models.

Application of multi-sensor unmanned aerial system for identification of hydrothermal alteration zones

2020 ◽  
Author(s):  
Yosoon Choi ◽  
Jieun Baek ◽  
Jangwon Suh ◽  
Sung-Min Kim
Keyword(s):  
Machine Learning ◽  
Classification Accuracy ◽  
Training Data ◽  
Sensor Data ◽  
Machine Learning Techniques ◽  
Integrated Analysis ◽  
Unmanned Aerial System ◽  
Data Sets ◽  
Learning Techniques ◽  
Hydrothermal Alteration Zones

<p>In this study, we proposed a method to utilize a multi-sensor Unmanned Aerial System (UAS) for exploration of hydrothermal alteration zones. This study selected an area (10m × 20m) composed mainly of the andesite and located on the coast, with wide outcrops and well-developed structural and mineralization elements. Multi-sensor (visible, multispectral, thermal, magnetic) data were acquired in the study area using UAS, and were studied using machine learning techniques. For utilizing the machine learning techniques, we applied the stratified random method to sample 1000 training data in the hydrothermal zone and 1000 training data in the non-hydrothermal zone identified through the field survey. The 2000 training data sets created for supervised learning were first classified into 1500 for training and 500 for testing. Then, 1500 for training were classified into 1200 for training and 300 for validation. The training and validation data for machine learning were generated in five sets to enable cross-validation. Five types of machine learning techniques were applied to the training data sets: k-Nearest Neighbors (k-NN), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Deep Neural Network (DNN). As a result of integrated analysis of multi-sensor data using five types of machine learning techniques, RF and SVM techniques showed high classification accuracy of about 90%. Moreover, performing integrated analysis using multi-sensor data showed relatively higher classification accuracy in all five machine learning techniques than analyzing magnetic sensing data or single optical sensing data only.</p>

scholarly journals Pharmacovigilance from social media: mining adverse drug reaction mentions using sequence labeling with word embedding cluster features

2015 ◽  
Vol 22 (3) ◽  
pp. 671-681 ◽  
Author(s):  
Azadeh Nikfarjam ◽  
Abeed Sarker ◽  
Karen O’Connor ◽  
Rachel Ginn ◽  
Graciela Gonzalez
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Language Processing ◽  
High Performance ◽  
Conditional Random Fields ◽  
Training Data ◽  
Data Sets ◽  
Social Media Mining ◽  
Medical Concepts ◽  
Media Mining

Abstract Objective Social media is becoming increasingly popular as a platform for sharing personal health-related information. This information can be utilized for public health monitoring tasks, particularly for pharmacovigilance, via the use of natural language processing (NLP) techniques. However, the language in social media is highly informal, and user-expressed medical concepts are often nontechnical, descriptive, and challenging to extract. There has been limited progress in addressing these challenges, and thus far, advanced machine learning-based NLP techniques have been underutilized. Our objective is to design a machine learning-based approach to extract mentions of adverse drug reactions (ADRs) from highly informal text in social media. Methods We introduce ADRMine, a machine learning-based concept extraction system that uses conditional random fields (CRFs). ADRMine utilizes a variety of features, including a novel feature for modeling words’ semantic similarities. The similarities are modeled by clustering words based on unsupervised, pretrained word representation vectors (embeddings) generated from unlabeled user posts in social media using a deep learning technique. Results ADRMine outperforms several strong baseline systems in the ADR extraction task by achieving an F-measure of 0.82. Feature analysis demonstrates that the proposed word cluster features significantly improve extraction performance. Conclusion It is possible to extract complex medical concepts, with relatively high performance, from informal, user-generated content. Our approach is particularly scalable, suitable for social media mining, as it relies on large volumes of unlabeled data, thus diminishing the need for large, annotated training data sets.

scholarly journals Predicting Fault Slip via Transfer Learning

2021 ◽  
Author(s):  
Kun Wang ◽  
Christopher Johnson ◽  
Kane Bennett ◽  
Paul Johnson
Keyword(s):  
Machine Learning ◽  
Numerical Simulations ◽  
Transfer Learning ◽  
Laboratory Experiments ◽  
Laboratory Data ◽  
Fault Slip ◽  
Geophysical Data ◽  
Training Data ◽  
Data Sets ◽  
Earthquake Cycle

Abstract Data-driven machine-learning for predicting instantaneous and future fault-slip in laboratory experiments has recently progressed markedly due to large training data sets. In Earth however, earthquake interevent times range from 10's-100's of years and geophysical data typically exist for only a portion of an earthquake cycle. Sparse data presents a serious challenge to training machine learning models. Here we describe a transfer learning approach using numerical simulations to train a convolutional encoder-decoder that predicts fault-slip behavior in laboratory experiments. The model learns a mapping between acoustic emission histories and fault-slip from numerical simulations, and generalizes to produce accurate results using laboratory data. Notably slip-predictions markedly improve using the simulation-data trained-model and training the latent space using a portion of a single laboratory earthquake-cycle. The transfer learning results elucidate the potential of using models trained on numerical simulations and fine-tuned with small geophysical data sets for potential applications to faults in Earth.

scholarly journals Comparison of the Validity and Generalizability of Machine Learning Algorithms for the Prediction of Energy Expenditure: Validation Study

10.2196/23938 ◽  
2021 ◽  
Vol 9 (8) ◽  
pp. e23938
Author(s):  
Ruairi O'Driscoll ◽  
Jake Turicchi ◽  
Mark Hopkins ◽  
Cristiana Duarte ◽  
Graham W Horgan ◽  
...  
Keyword(s):  
Physical Activity ◽  
Machine Learning ◽  
Neural Networks ◽  
Random Forest ◽  
Energy Expenditure ◽  
Superior Performance ◽  
Gradient Boosting ◽  
Support Vector ◽  
Data Sets ◽  
Out Of Sample

Background Accurate solutions for the estimation of physical activity and energy expenditure at scale are needed for a range of medical and health research fields. Machine learning techniques show promise in research-grade accelerometers, and some evidence indicates that these techniques can be applied to more scalable commercial devices. Objective This study aims to test the validity and out-of-sample generalizability of algorithms for the prediction of energy expenditure in several wearables (ie, Fitbit Charge 2, ActiGraph GT3-x, SenseWear Armband Mini, and Polar H7) using two laboratory data sets comprising different activities. Methods Two laboratory studies (study 1: n=59, age 44.4 years, weight 75.7 kg; study 2: n=30, age=31.9 years, weight=70.6 kg), in which adult participants performed a sequential lab-based activity protocol consisting of resting, household, ambulatory, and nonambulatory tasks, were combined in this study. In both studies, accelerometer and physiological data were collected from the wearables alongside energy expenditure using indirect calorimetry. Three regression algorithms were used to predict metabolic equivalents (METs; ie, random forest, gradient boosting, and neural networks), and five classification algorithms (ie, k-nearest neighbor, support vector machine, random forest, gradient boosting, and neural networks) were used for physical activity intensity classification as sedentary, light, or moderate to vigorous. Algorithms were evaluated using leave-one-subject-out cross-validations and out-of-sample validations. Results The root mean square error (RMSE) was lowest for gradient boosting applied to SenseWear and Polar H7 data (0.91 METs), and in the classification task, gradient boost applied to SenseWear and Polar H7 was the most accurate (85.5%). Fitbit models achieved an RMSE of 1.36 METs and 78.2% accuracy for classification. Errors tended to increase in out-of-sample validations with the SenseWear neural network achieving RMSE values of 1.22 METs in the regression tasks and the SenseWear gradient boost and random forest achieving an accuracy of 80% in classification tasks. Conclusions Algorithms trained on combined data sets demonstrated high predictive accuracy, with a tendency for superior performance of random forests and gradient boosting for most but not all wearable devices. Predictions were poorer in the between-study validations, which creates uncertainty regarding the generalizability of the tested algorithms.

scholarly journals Gender bias in machine learning for sentiment analysis

2018 ◽  
Vol 42 (3) ◽  
pp. 343-354 ◽  
Author(s):  
Mike Thelwall
Keyword(s):  
Machine Learning ◽  
Sentiment Analysis ◽  
Gender Bias ◽  
Training Data ◽  
Data Sets ◽  
Content Type ◽  
Female Authors ◽  
Gender Biases ◽  
Mixed Gender ◽  
Gender Specific

Purpose The purpose of this paper is to investigate whether machine learning induces gender biases in the sense of results that are more accurate for male authors or for female authors. It also investigates whether training separate male and female variants could improve the accuracy of machine learning for sentiment analysis. Design/methodology/approach This paper uses ratings-balanced sets of reviews of restaurants and hotels (3 sets) to train algorithms with and without gender selection. Findings Accuracy is higher on female-authored reviews than on male-authored reviews for all data sets, so applications of sentiment analysis using mixed gender data sets will over represent the opinions of women. Training on same gender data improves performance less than having additional data from both genders. Practical implications End users of sentiment analysis should be aware that its small gender biases can affect the conclusions drawn from it and apply correction factors when necessary. Users of systems that incorporate sentiment analysis should be aware that performance will vary by author gender. Developers do not need to create gender-specific algorithms unless they have more training data than their system can cope with. Originality/value This is the first demonstration of gender bias in machine learning sentiment analysis.

Do the robot

2017 ◽  
Vol 54 (2) ◽  
pp. 193-214 ◽  
Author(s):  
Michael Colaresi ◽  
Zuhaib Mahmood
Keyword(s):  
Machine Learning ◽  
Model Development ◽  
Theory Development ◽  
Training Data ◽  
Generation Process ◽  
Data Generation ◽  
Research Strategies ◽  
Conflict Processes ◽  
Iterative Model ◽  
Out Of Sample

Increasingly, scholars interested in understanding conflict processes have turned to evaluating out-of-sample forecasts to judge and compare the usefulness of their models. Research in this vein has made significant progress in identifying and avoiding the problem of overfitting sample data. Yet there has been less research providing strategies and tools to practically improve the out-of-sample performance of existing models and connect forecasting improvement to the goal of theory development in conflict studies. In this article, we fill this void by building on lessons from machine learning research. We highlight a set of iterative tasks, which David Blei terms ‘Box’s loop’, that can be summarized as build, compute, critique, and think. While the initial steps of Box’s loop will be familiar to researchers, the underutilized process of model criticism allows researchers to iteratively learn more useful representations of the data generation process from the discrepancies between the trained model and held-out data. To benefit from iterative model criticism, we advise researchers not only to split their available data into separate training and test sets, but also sample from their training data to allow for iterative model development, as is common in machine learning applications. Since practical tools for model criticism in particular are underdeveloped, we also provide software for new visualizations that build upon already existing tools. We use models of civil war onset to provide an illustration of how our machine learning-inspired research design can simultaneously improve out-of-sample forecasting performance and identify useful theoretical contributions. We believe these research strategies can complement existing designs to accelerate innovations across conflict processes.

scholarly journals Expert-augmented machine learning

2020 ◽  
Vol 117 (9) ◽  
pp. 4571-4577 ◽  
Author(s):  
Efstathios D. Gennatas ◽  
Jerome H. Friedman ◽  
Lyle H. Ungar ◽  
Romain Pirracchio ◽  
Eric Eaton ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Strategy ◽  
Expert Knowledge ◽  
Decision Rules ◽  
Total Sample ◽  
Care Patient ◽  
Training Data ◽  
Empirical Risk ◽  
Automated Method ◽  
Out Of Sample

Machine learning is proving invaluable across disciplines. However, its success is often limited by the quality and quantity of available data, while its adoption is limited by the level of trust afforded by given models. Human vs. machine performance is commonly compared empirically to decide whether a certain task should be performed by a computer or an expert. In reality, the optimal learning strategy may involve combining the complementary strengths of humans and machines. Here, we present expert-augmented machine learning (EAML), an automated method that guides the extraction of expert knowledge and its integration into machine-learned models. We used a large dataset of intensive-care patient data to derive 126 decision rules that predict hospital mortality. Using an online platform, we asked 15 clinicians to assess the relative risk of the subpopulation defined by each rule compared to the total sample. We compared the clinician-assessed risk to the empirical risk and found that, while clinicians agreed with the data in most cases, there were notable exceptions where they overestimated or underestimated the true risk. Studying the rules with greatest disagreement, we identified problems with the training data, including one miscoded variable and one hidden confounder. Filtering the rules based on the extent of disagreement between clinician-assessed risk and empirical risk, we improved performance on out-of-sample data and were able to train with less data. EAML provides a platform for automated creation of problem-specific priors, which help build robust and dependable machine-learning models in critical applications.

Facial Attractiveness: Beauty and the Machine

2006 ◽  
Vol 18 (1) ◽  
pp. 119-142 ◽  
Author(s):  
Yael Eisenthal ◽  
Gideon Dror ◽  
Eytan Ruppin
Keyword(s):  
Machine Learning ◽  
Facial Attractiveness ◽  
Training Data ◽  
Data Sets ◽  
Learning Context ◽  
Data Set ◽  
Attractiveness Rating ◽  
Facial Beauty ◽  
Facial Images

This work presents a novel study of the notion of facial attractiveness in a machine learning context. To this end, we collected human beauty ratings for data sets of facial images and used various techniques for learning the attractiveness of a face. The trained predictor achieves a significant correlation of 0.65 with the average human ratings. The results clearly show that facial beauty is a universal concept that a machine can learn. Analysis of the accuracy of the beauty prediction machine as a function of the size of the training data indicates that a machine producing human-like attractiveness rating could be obtained given a moderately larger data set.

scholarly journals THE USE OF MACHINE LEARNING METHODS FOR BINARY CLASSIFICATION OF THE WORKING CONDITION OF BEARINGS USING THE SIGNALS OF VIBRATION ACCELERATION

2021 ◽  
pp. 15-22
Author(s):  
Ruslan Babudzhan ◽  
Konstantyn Isaienkov ◽  
Danilo Krasiy ◽  
Oleksii Vodka ◽  
Ivan Zadorozhny ◽  
...  
Keyword(s):  
Machine Learning ◽  
Binary Classification ◽  
Fractal Dimensions ◽  
Feature Space ◽  
Training Data ◽  
Supervised Machine Learning ◽  
Support Vector ◽  
Data Sets ◽  
Vibration Acceleration ◽  
K Nearest Neighbors

The paper investigates the relationship between vibration acceleration of bearings with their operational state. To determine these dependencies, a testbench was built and 112 experiments were carried out with different bearings: 100 bearings that developed an internal defect during operation and 12bearings without a defect. From the obtained records, a dataset was formed, which was used to build classifiers. Dataset is freely available. A methodfor classifying new and used bearings was proposed, which consists in searching for dependencies and regularities of the signal using descriptive functions: statistical, entropy, fractal dimensions and others. In addition to processing the signal itself, the frequency domain of the bearing operationsignal was also used to complement the feature space. The paper considered the possibility of generalizing the classification for its application on thosesignals that were not obtained in the course of laboratory experiments. An extraneous dataset was found in the public domain. This dataset was used todetermine how accurate a classifier was when it was trained and tested on significantly different signals. Training and validation were carried out usingthe bootstrapping method to eradicate the effect of randomness, given the small amount of training data available. To estimate the quality of theclassifiers, the F1-measure was used as the main metric due to the imbalance of the data sets. The following supervised machine learning methodswere chosen as classifier models: logistic regression, support vector machine, random forest, and K nearest neighbors. The results are presented in theform of plots of density distribution and diagrams.

Sign in / Sign up

Export Citation Format

Share Document