scholarly journals Splitting chemical structure data sets for federated privacy-preserving machine learning

2021 ◽  
Author(s):  
Jaak Simm ◽  
Lina Humbeck ◽  
Adam Zalewski ◽  
Noe Sturm ◽  
Wouter Heyndrickx ◽  
...  
Keyword(s):  
Machine Learning ◽  
Quality Criteria ◽  
Privacy Preserving ◽  
Locality Sensitive Hashing ◽  
Data Sets ◽  
Data Set ◽  
Test Set ◽  
Chemical Structures ◽  
Multiple Partners ◽  
Applications Of Machine Learning

With the increase in applications of machine learning methods in drug design and related fields, the challenge of designing sound test sets becomes more and more prominent. The goal of this challenge is to have a realistic split of chemical structures (compounds) between training, validation and test set such that the performance on the test set is meaningful to infer the performance in a prospective application. This challenge is by its own very interesting and relevant,but is even more complex in a federated machine learning approach where multiple partners jointly train a model under privacy-preserving conditions where chemical structures must not be shared between the different participating parties in the federated learning. In this work we discuss three methods which provide a splitting of the data set and are applicable in a federated privacy-preserving setting, namely: a. locality-sensitive hashing (LSH), b. sphere exclusion clustering, c. scaffold-based binning (scaffold network). For evaluation of these splitting methods we consider the following quality criteria: bias in prediction performance, label and data imbalance, distance of the test set compounds to the training set and compare them to a random splitting. The main findings of the paper are a. both sphere exclusion clustering and scaffold-based binning result in high quality splitting of the data sets, b. in terms of compute costs sphere exclusion clustering is very expensive in the case of federated privacy-preserving setting.

scholarly journals Splitting chemical structure data sets for federated privacy-preserving machine learning

2021 ◽  
Vol 13 (1) ◽  
Author(s):  
Jaak Simm ◽  
Lina Humbeck ◽  
Adam Zalewski ◽  
Noe Sturm ◽  
Wouter Heyndrickx ◽  
...  
Keyword(s):  
Machine Learning ◽  
Quality Criteria ◽  
Privacy Preserving ◽  
Locality Sensitive Hashing ◽  
Data Sets ◽  
Data Set ◽  
Test Set ◽  
Chemical Structures ◽  
Multiple Partners ◽  
Applications Of Machine Learning

AbstractWith the increase in applications of machine learning methods in drug design and related fields, the challenge of designing sound test sets becomes more and more prominent. The goal of this challenge is to have a realistic split of chemical structures (compounds) between training, validation and test set such that the performance on the test set is meaningful to infer the performance in a prospective application. This challenge is by its own very interesting and relevant, but is even more complex in a federated machine learning approach where multiple partners jointly train a model under privacy-preserving conditions where chemical structures must not be shared between the different participating parties. In this work we discuss three methods which provide a splitting of a data set and are applicable in a federated privacy-preserving setting, namely: a. locality-sensitive hashing (LSH), b. sphere exclusion clustering, c. scaffold-based binning (scaffold network). For evaluation of these splitting methods we consider the following quality criteria (compared to random splitting): bias in prediction performance, classification label and data imbalance, similarity distance between the test and training set compounds. The main findings of the paper are a. both sphere exclusion clustering and scaffold-based binning result in high quality splitting of the data sets, b. in terms of compute costs sphere exclusion clustering is very expensive in the case of federated privacy-preserving setting.

scholarly journals Splitting chemical structure data sets for federated privacy-preserving machine learning

2021 ◽  
Author(s):  
Jaak Simm ◽  
Lina Humbeck ◽  
Adam Zalewski ◽  
Noe Sturm ◽  
Wouter Heyndrickx ◽  
...  
Keyword(s):  
Machine Learning ◽  
Quality Criteria ◽  
Privacy Preserving ◽  
Locality Sensitive Hashing ◽  
Data Sets ◽  
Data Set ◽  
Test Set ◽  
Chemical Structures ◽  
Multiple Partners ◽  
Applications Of Machine Learning

With the increase in applications of machine learning methods in drug design and related fields, the challenge of designing sound test sets becomes more and more prominent. The goal of this challenge is to have a realistic split of chemical structures (compounds) between training, validation and test set such that the performance on the test set is meaningful to infer the performance in a prospective application. This challenge is by its own very interesting and relevant,but is even more complex in a federated machine learning approach where multiple partners jointly train a model under privacy-preserving conditions where chemical structures must not be shared between the different participating parties in the federated learning. In this work we discuss three methods which provide a splitting of the data set and are applicable in a federated privacy-preserving setting, namely: a. locality-sensitive hashing (LSH), b. sphere exclusion clustering, c. scaffold-based binning (scaffold network). For evaluation of these splitting methods we consider the following quality criteria: bias in prediction performance, label and data imbalance, distance of the test set compounds to the training set and compare them to a random splitting. The main findings of the paper are a. both sphere exclusion clustering and scaffold-based binning result in high quality splitting of the data sets, b. in terms of compute costs sphere exclusion clustering is very expensive in the case of federated privacy-preserving setting.

scholarly journals Machine Learning Algorithms for Analysis of DNA Data Sets

2012 ◽  
pp. 47-58 ◽  
Author(s):  
John Yearwood ◽  
Adil Bagirov ◽  
Andrei V. Kelarev
Keyword(s):  
Machine Learning ◽  
Dna Sequences ◽  
Learning Algorithm ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Local Alignment ◽  
Data Sets ◽  
Data Set ◽  
Applications Of Machine Learning ◽  
New Machine

The applications of machine learning algorithms to the analysis of data sets of DNA sequences are very important. The present chapter is devoted to the experimental investigation of applications of several machine learning algorithms for the analysis of a JLA data set consisting of DNA sequences derived from non-coding segments in the junction of the large single copy region and inverted repeat A of the chloroplast genome in Eucalyptus collected by Australian biologists. Data sets of this sort represent a new situation, where sophisticated alignment scores have to be used as a measure of similarity. The alignment scores do not satisfy properties of the Minkowski metric, and new machine learning approaches have to be investigated. The authors’ experiments show that machine learning algorithms based on local alignment scores achieve very good agreement with known biological classes for this data set. A new machine learning algorithm based on graph partitioning performed best for clustering of the JLA data set. Our novel k-committees algorithm produced most accurate results for classification. Two new examples of synthetic data sets demonstrate that the authors’ k-committees algorithm can outperform both the Nearest Neighbour and k-medoids algorithms simultaneously.

scholarly journals Generation of geometric interpolations of building types with deep variational autoencoders

2020 ◽  
Vol 6 ◽  
Author(s):  
Jaime de Miguel Rodríguez ◽  
Maria Eugenia Villafañe ◽  
Luka Piškorec ◽  
Fernando Sancho Caparrini
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Large Data ◽  
Learning Model ◽  
Large Data Sets ◽  
Data Sets ◽  
Connectivity Map ◽  
Data Set ◽  
3D Objects ◽  
Machine Learning Model

Abstract This work presents a methodology for the generation of novel 3D objects resembling wireframes of building types. These result from the reconstruction of interpolated locations within the learnt distribution of variational autoencoders (VAEs), a deep generative machine learning model based on neural networks. The data set used features a scheme for geometry representation based on a ‘connectivity map’ that is especially suited to express the wireframe objects that compose it. Additionally, the input samples are generated through ‘parametric augmentation’, a strategy proposed in this study that creates coherent variations among data by enabling a set of parameters to alter representative features on a given building type. In the experiments that are described in this paper, more than 150 k input samples belonging to two building types have been processed during the training of a VAE model. The main contribution of this paper has been to explore parametric augmentation for the generation of large data sets of 3D geometries, showcasing its problems and limitations in the context of neural networks and VAEs. Results show that the generation of interpolated hybrid geometries is a challenging task. Despite the difficulty of the endeavour, promising advances are presented.

A systematic review of machine learning-based missing value imputation techniques

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Tressy Thomas ◽  
Enayat Rajabi
Keyword(s):  
Machine Learning ◽  
Selection Process ◽  
Evaluation Metrics ◽  
Correct Prediction ◽  
Data Sets ◽  
Data Set ◽  
Missing Value ◽  
Content Type ◽  
Missing Value Imputation ◽  
Literature Reviews

PurposeThe primary aim of this study is to review the studies from different dimensions including type of methods, experimentation setup and evaluation metrics used in the novel approaches proposed for data imputation, particularly in the machine learning (ML) area. This ultimately provides an understanding about how well the proposed framework is evaluated and what type and ratio of missingness are addressed in the proposals. The review questions in this study are (1) what are the ML-based imputation methods studied and proposed during 2010–2020? (2) How the experimentation setup, characteristics of data sets and missingness are employed in these studies? (3) What metrics were used for the evaluation of imputation method?Design/methodology/approachThe review process went through the standard identification, screening and selection process. The initial search on electronic databases for missing value imputation (MVI) based on ML algorithms returned a large number of papers totaling at 2,883. Most of the papers at this stage were not exactly an MVI technique relevant to this study. The literature reviews are first scanned in the title for relevancy, and 306 literature reviews were identified as appropriate. Upon reviewing the abstract text, 151 literature reviews that are not eligible for this study are dropped. This resulted in 155 research papers suitable for full-text review. From this, 117 papers are used in assessment of the review questions.FindingsThis study shows that clustering- and instance-based algorithms are the most proposed MVI methods. Percentage of correct prediction (PCP) and root mean square error (RMSE) are most used evaluation metrics in these studies. For experimentation, majority of the studies sourced the data sets from publicly available data set repositories. A common approach is that the complete data set is set as baseline to evaluate the effectiveness of imputation on the test data sets with artificially induced missingness. The data set size and missingness ratio varied across the experimentations, while missing datatype and mechanism are pertaining to the capability of imputation. Computational expense is a concern, and experimentation using large data sets appears to be a challenge.Originality/valueIt is understood from the review that there is no single universal solution to missing data problem. Variants of ML approaches work well with the missingness based on the characteristics of the data set. Most of the methods reviewed lack generalization with regard to applicability. Another concern related to applicability is the complexity of the formulation and implementation of the algorithm. Imputations based on k-nearest neighbors (kNN) and clustering algorithms which are simple and easy to implement make it popular across various domains.

scholarly journals Birds Sound Classification Based on Machine Learning Algorithms

2021 ◽  
pp. 1-11
Author(s):  
Aska E. Mehyadin ◽  
Adnan Mohsin Abdulazeez ◽  
Dathar Abas Hasan ◽  
Jwan N. Saeed
Keyword(s):  
Machine Learning ◽  
Noise Suppression ◽  
Bird Species ◽  
Machine Learning Algorithms ◽  
Data Sets ◽  
Learning Technology ◽  
Species Classification ◽  
Data Set ◽  
Sound Classification ◽  
Mel Frequency Cepstral Coefficient

The bird classifier is a system that is equipped with an area machine learning technology and uses a machine learning method to store and classify bird calls. Bird species can be known by recording only the sound of the bird, which will make it easier for the system to manage. The system also provides species classification resources to allow automated species detection from observations that can teach a machine how to recognize whether or classify the species. Non-undesirable noises are filtered out of and sorted into data sets, where each sound is run via a noise suppression filter and a separate classification procedure so that the most useful data set can be easily processed. Mel-frequency cepstral coefficient (MFCC) is used and tested through different algorithms, namely Naïve Bayes, J4.8 and Multilayer perceptron (MLP), to classify bird species. J4.8 has the highest accuracy (78.40%) and is the best. Accuracy and elapsed time are (39.4 seconds).

scholarly journals GAINESIS: Generative Artificial Intelligence NEtlists SynthesIS

Electronics ◽  
2022 ◽  
Vol 11 (2) ◽  
pp. 245
Author(s):  
Konstantinos G. Liakos ◽  
Georgios K. Georgakilas ◽  
Fotis C. Plessas ◽  
Paris Kitsos
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Power Analysis ◽  
Public Libraries ◽  
Data Sets ◽  
Hardware Trojan ◽  
Generative Adversarial Network ◽  
Data Set ◽  
Encrypted Data ◽  
Adversarial Network

A significant problem in the field of hardware security consists of hardware trojan (HT) viruses. The insertion of HTs into a circuit can be applied for each phase of the circuit chain of production. HTs degrade the infected circuit, destroy it or leak encrypted data. Nowadays, efforts are being made to address HTs through machine learning (ML) techniques, mainly for the gate-level netlist (GLN) phase, but there are some restrictions. Specifically, the number and variety of normal and infected circuits that exist through the free public libraries, such as Trust-HUB, are based on the few samples of benchmarks that have been created from circuits large in size. Thus, it is difficult, based on these data, to develop robust ML-based models against HTs. In this paper, we propose a new deep learning (DL) tool named Generative Artificial Intelligence Netlists SynthesIS (GAINESIS). GAINESIS is based on the Wasserstein Conditional Generative Adversarial Network (WCGAN) algorithm and area–power analysis features from the GLN phase and synthesizes new normal and infected circuit samples for this phase. Based on our GAINESIS tool, we synthesized new data sets, different in size, and developed and compared seven ML classifiers. The results demonstrate that our new generated data sets significantly enhance the performance of ML classifiers compared with the initial data set of Trust-HUB.

scholarly journals Optimization through Bayesian Classification on the k-Anonymized Data

2012 ◽  
pp. 245-249
Author(s):  
L Mohana Tirumala ◽  
S. Srinivasa Rao
Keyword(s):  
Cost Effective ◽  
Privacy Preserving ◽  
Bayesian Classification ◽  
Data Sets ◽  
Data Set ◽  
Equal Importance ◽  
Cost Effective Approach ◽  
Effective Manner ◽  
Anonymized Data ◽  
Refinement Algorithm

Privacy preserving in Data mining & publishing, plays a major role in today networked world. It is important to preserve the privacy of the vital information corresponding to a data set. This process can be achieved by k-anonymization solution for classification. Along with the privacy preserving using anonymization, yielding the optimized data sets is also of equal importance with a cost effective approach. In this paper Top-Down Refinement algorithm has been proposed which yields optimum results in a cost effective manner. Bayesian Classification has been proposed in this paper to predict class membership probabilities for a data tuple for which the associated class label is unknown.

Evaluation of Machine-Learning Tools for Predicting Sand Production

2021 ◽  
Author(s):  
Afungchwi Ronald Ngwashi ◽  
David O. Ogbe ◽  
Dickson O. Udebhulu
Keyword(s):  
Machine Learning ◽  
Niger Delta ◽  
Oil And Gas ◽  
Back Propagation ◽  
Oil And Gas Industry ◽  
Learning Tools ◽  
Sand Production ◽  
Data Set ◽  
Test Set ◽  
Gas Industry

Abstract Data analytics has only recently picked the interest of the oil and gas industry as it has made data visualization much simpler, faster, and cost-effective. This is driven by the promising innovative techniques in developing artificial intelligence and machine-learning tools to provide sustainable solutions to ever-increasing problems of the petroleum industry activities. Sand production is one of these real issues faced by the oil and gas industry. Understanding whether a well will produce sand or not is the foundation of every completion job in sandstone formations. The Niger Delta Province is a region characterized by friable and unconsolidated sandstones, therefore it's more prone to sanding. It is economically unattractive in this region to design sand equipment for a well that will not produce sand. This paper is aimed at developing a fast and more accurate machine-learning algorithm to predict sanding in sandstone formations. A two-layered Artificial Neural Network (ANN) with back-propagation algorithm was developed using PYTHON programming language. The algorithm uses 11 geological and reservoir parameters that are associated with the onset of sanding. These parameters include depth, overburden, pore pressure, maximum and minimum horizontal stresses, well azimuth, well inclination, Poisson's ratio, Young's Modulus, friction angle, and shale content. Data typical of the Niger Delta were collected to validate the algorithm. The data was further split into a training set (70%) and a test set (30%). Statistical analyses of the data yielded correlations between the parameters and were plotted for better visualization. The accuracy of the ANN algorithm is found to depend on the number of parameters, number of epochs, and the size of the data set. For a completion engineer, the answer to the question of whether or not a well will require sand production control is binary-either a well will produce sand or it does not. Support vector machines (SVM) are known to be better suited as the machine-learning tools for binary identification. This study also presents a comparative analysis between ANN and SVM models as tools for predicting sand production. Analysis of the Niger Delta data set indicated that SVM outperformed ANN model even when the training data set is sparse. Using the 30% test set, ANN gives an accuracy, precision, recall, and F1 - Score of about 80% while the SVM performance was 100% for the four metrics. It is then concluded that machine learning tools such as ANN with back-propagation and SVM are simple, accurate, and easy-to-use tools for effectively predicting sand production.

scholarly journals Precision-Recall versus Accuracy and the Role of Large Data Sets

2019 ◽  
Vol 33 ◽  
pp. 4039-4048 ◽  
Author(s):  
Brendan Juba ◽  
Hai S. Le
Keyword(s):  
Machine Learning ◽  
Class Imbalance ◽  
Imbalanced Data ◽  
Large Data ◽  
Constant Factor ◽  
Data Sets ◽  
Data Set ◽  
Small Constant ◽  
Classifier Performance ◽  
Necessary And Sufficient

Practitioners of data mining and machine learning have long observed that the imbalance of classes in a data set negatively impacts the quality of classifiers trained on that data. Numerous techniques for coping with such imbalances have been proposed, but nearly all lack any theoretical grounding. By contrast, the standard theoretical analysis of machine learning admits no dependence on the imbalance of classes at all. The basic theorems of statistical learning establish the number of examples needed to estimate the accuracy of a classifier as a function of its complexity (VC-dimension) and the confidence desired; the class imbalance does not enter these formulas anywhere. In this work, we consider the measures of classifier performance in terms of precision and recall, a measure that is widely suggested as more appropriate to the classification of imbalanced data. We observe that whenever the precision is moderately large, the worse of the precision and recall is within a small constant factor of the accuracy weighted by the class imbalance. A corollary of this observation is that a larger number of examples is necessary and sufficient to address class imbalance, a finding we also illustrate empirically.

Sign in / Sign up

Export Citation Format

Share Document