Artificial Examples Research Articles

Investigating Credit Card Payment Fraud with Detection Methods Using Advanced Machine Learning

In the cybersecurity industry, where legitimate transactions far outnumber fraudulent ones, detecting fraud is of paramount significance. In order to evaluate the accuracy of detecting fraudulent transactions in imbalanced real datasets, this study compares the efficacy of two approaches, random under-sampling and oversampling, using the synthetic minority over-sampling technique (SMOTE). Random under-sampling aims for fairness by excluding examples from the majority class, but this compromises precision in favor of recall. To strike a balance and ensure statistical significance, SMOTE was used instead to produce artificial examples of the minority class. Based on the data obtained, it is clear that random under-sampling achieves high recall (92.86%) at the expense of low precision, whereas SMOTE achieves a higher accuracy (86.75%) and a more even F1 score (73.47%) at the expense of a slightly lower recall. As true fraudulent transactions require at least two methods for verification, we investigated different machine learning methods and made suitable balances between accuracy, F1 score, and recall. Our comparison sheds light on the subtleties and ramifications of each approach, allowing professionals in the field of cybersecurity to better choose the approach that best meets the needs of their own firm. This research highlights the need to resolve class imbalances for effective fraud detection in cybersecurity, as well as the need for constant monitoring and the investigation of new approaches to increase applicability.

A Statistical Model of Hazard Rate Change Caused by a Random Shock

In the framework of statistical reliability analysis, we propose a model describing the change of reliability of a technical component after certain event, as for instance a shock, or the failure and maintenance. Two kinds of failure rate changes are considered, namely a shift of virtual age of analyzed component and the acceleration of its subjective time. The objective is to estimate both the baseline lifetime probability distribution and the magnitude of these changes. The estimation procedure is based on the maximum likelihood method, both parametric and nonparametric distributions of lifetime are considered. Further, a graphical method of the goodness-of-fit test is proposed. The method is then checked on artificial examples, finally a real data problem is solved.

Latent growth factors as predictors of distal outcomes.

A currently overlooked application of the latent curve model (LCM) is its use in assessing the consequences of development patterns of change-that is as a predictor of distal outcomes. However, there are additional complications for appropriately specifying and interpreting the distal outcome LCM. Here, we develop a general framework for understanding the sensitivity of the distal outcome LCM to the choice of time coding, focusing on the regressions of the distal outcome on the latent growth factors. Using artificial and real-data examples, we highlight the unexpected changes in the regression of the slope factor which stand in contrast to prior work on time coding effects, and develop a framework for estimating the distal outcome LCM at a point in the trajectory-known as the aperture-which maximizes the interpretability of the effects. We also outline a prioritization approach developed for assessing incremental validity to obtain consistently interpretable estimates of the effect of the slope. Throughout, we emphasize practical steps for understanding these changing predictive effects, including graphical approaches for assessing regions of significance similar to those used to probe interaction effects. We conclude by providing recommendations for applied research using these models and outline an agenda for future work in this area. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Deducing neighborhoods of classes from a fitted model

In this article, a new kind of interpretable machine learning method is presented, which can help to understand the partition of the feature space into predicted classes in a classification model using quantile shifts, and this way make the underlying statistical or machine learning model more trustworthy. Basically, real data points (or specific points of interest) are used and the changes of the prediction after slightly raising or decreasing specific features are observed. By comparing the predictions before and after the shifts, under certain conditions the observed changes in the predictions can be interpreted as neighborhoods of the classes with regard to the shifted features. Chord diagrams are used to visualize the observed changes. For illustration, this quantile shift method (QSM) is applied to an artificial example with medical labels and a real data example.

A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer

The Gleason score is an important predictor of prognosis in prostate cancer. However, its subjective nature can result in over- or under-grading. Our objective was to train an artificial intelligence (AI)-based algorithm to grade prostate cancer in specimens from patients who underwent radical prostatectomy (RP) and to assess the correlation of AI-estimated proportions of different Gleason patterns with biochemical recurrence-free survival (RFS), metastasis-free survival (MFS), and overall survival (OS). Training and validation of algorithms for cancer detection and grading were completed with three large datasets containing a total of 580 whole-mount prostate slides from 191 RP patients at two centers and 6218 annotated needle biopsy slides from the publicly available Prostate Cancer Grading Assessment dataset. A cancer detection model was trained using MobileNetV3 on 0.5 mm × 0.5 mm cancer areas (tiles) captured at 10× magnification. For cancer grading, a Gleason pattern detector was trained on tiles using a ResNet50 convolutional neural network and a selective CutMix training strategy involving a mixture of real and artificial examples. This strategy resulted in improved model generalizability in the test set compared with three different control experiments when evaluated on both needle biopsy slides and whole-mount prostate slides from different centers. In an additional test cohort of RP patients who were clinically followed over 30 years, quantitative Gleason pattern AI estimates achieved concordance indexes of 0.69, 0.72, and 0.64 for predicting RFS, MFS, and OS times, outperforming the control experiments and International Society of Urological Pathology system (ISUP) grading by pathologists. Finally, unsupervised clustering of test RP patient specimens into low-, medium-, and high-risk groups based on AI-estimated proportions of each Gleason pattern resulted in significantly improved RFS and MFS stratification compared with ISUP grading. In summary, deep learning-based quantitative Gleason scoring using a selective CutMix training strategy may improve prognostication after prostate cancer surgery.

Comparison of a quasi Newton method using Broyden’s update formula and an adjoint method for determining local magnetic material properties of electrical steel sheets

PurposePerforming accurate numerical simulations of electrical drives, the precise knowledge of the local magnetic material properties is of utmost importance. Due to the various manufacturing steps, e.g. heat treatment or cutting techniques, the magnetic material properties can strongly vary locally, and the assumption of homogenized global material parameters is no longer feasible. This paper aims to present the general methodology and two different solution strategies for determining the local magnetic material properties using reference and simulation data.Design/methodology/approachThe general methodology combines methods based on measurement, numerical simulation and solving an inverse problem. Therefore, a sensor-actuator system is used to characterize electrical steel sheets locally. Based on the measurement data and results from the finite element simulation, the inverse problem is solved with two different solution strategies. The first one is a quasi Newton method (QNM) using Broyden's update formula to approximate the Jacobian and the second is an adjoint method. For comparison of both methods regarding convergence and efficiency, an artificial example with a linear material model is considered.FindingsThe QNM and the adjoint method show similar convergence behavior for two different cutting-edge effects. Furthermore, considering a priori information improved the convergence rate. However, no impact on the stability and the remaining error is observed.Originality/valueThe presented methodology enables a fast and simple determination of the local magnetic material properties of electrical steel sheets without the need for a large number of samples or special preparation procedures.

Environmentally Driven Intelligent Textiles That Reversibly Actuate to Provide Large Change in Porosity, Loft, Shape, or Their Combination

AbstractTextiles that intelligently respond to changes in the environment by reversibly providing dramatic changes in structure for comfort, safety, or appearance are widely sought. In order to facilitate applications development, here novel intelligent textile structures are described and developed by using polymer artificial muscles and examples of structures and mechanisms are provided that can be used in the future. These actuating textiles, which deploy a host of muscle yarn and fiber types, can change their porosity with or without any change in fabric lateral dimension or can change their loft or can either bend or fold upon temperature change. The presented intelligent textiles provide a porosity increase up to 38.3% from room temperature to 40 °C, while they exhibit a decrease in porosity up to 26.1% from room temperature to −5 °C. Textiles that can bend at an angle of 60° at 40 °C, while exhibiting a complete roll‐up at 70 °C, are also presented. An important way to change the direction of actuation produced by environmental exposure is to transition from homochiral fibers to heterochiral fibers. Other more complicated yarn types can usefully deploy chirality effects, such as yarns that are twisted and coiled, or yarns that are twisted, coiled, and then plied.

Rejection sampling and agent-based models for data limited fisheries

Many of the world’s fisheries are “data-limited” where the information does not allow precise determination of fish stock status and limits the development of appropriate management responses. Two approaches are proposed for use in data-limited stock management strategy evaluations to guide the evaluations and to understand the sources of uncertainty: rejection sampling methods and the incorporation of more complex socio-economic dynamics into management evaluations using agent-based models. In rejection sampling (or rejection filtering) a model is simulated many times with a wide range of priors on parameters and outcomes are compared multiple filtering criteria. Those simulations that pass all the filters form an ensemble of feasible models. The ensemble can be used to look for robust management strategies, robust to both model uncertainties. Agent-based models of fishery economics can be implemented within the rejection framework, integrating the biological and economic understanding of the fishery. A simple artificial example of a difference equation bio-economic model is given to demonstrate the approach. Then rejection sampling is applied to an agent-based model for the hairtail (Trichiurus japonicas) fishery, where an operating model is constructed with rejection/agent-based methods and compared to known data and analyses of the fishery. The usefulness of information and rejection filters are illuminated and efficacy examined. The methods can be helpful for strategic guidance where multiple states of nature are possible as a part of management strategy evaluation.

Combination of several predictors for forensic age estimation

Abstract Background Nearly all practical forensic age diagnostics studies estimate the probability distribution of the age conditional on the developmental status of a certain anatomical feature. Given such a probability distribution, the probability of a person exceeding a certain legal age threshold is computed. In court, forensic experts are often asked to summarize the probabilities obtained by evaluating different age indicators of the same person. Objective The present study demonstrates computation of the age probability distribution conditional on the conjunction of several different age indicators given the age probability distributions conditional on the development status of the respective single anatomical features. Material and methods Data from two distinctively different studies on age estimation were used to join their probability information via Bayes’ theorem. Each of the cited studies is based on the development status of only one of two different anatomical structures: third molar and clavicular epiphysis. Results We derive general formulae for Bayesian information joining in forensic age estimation. Posterior distributions of age class, given the simultaneous statuses of the two anatomical features are generated. Finally, the study presents the technique on an artificial case example. Conclusion Bayes’ theorem can be used in forensic age estimations to combine information from several different anatomical features to yield more precise probability values of age given development status data of several distinctly different anatomical features. Conditional stochastic independence of the single age indicators as used in our article has to be scrutinized and is not generally recommendable.

Data assimilation in 2D hyperbolic/parabolic systems using a stabilized explicit finite difference scheme run backward in time

ABSTRACTAn artificial example of a coupled system of three nonlinear partial differential equations generalizing 2D thermoelastic vibrations, is used to demonstrate the effectiveness, as well as the limitations, of a non iterative direct procedure in data assimilation. A stabilized explicit finite difference scheme, run backward in time, is used to find initial values, [u(.,0),v(.,0),w(.,0)], that can evolve into a useful approximation to a hypothetical target result [u∗(.,Tmax),v∗(.,Tmax),w∗(Tmax)], at some realistic 0 $ ]]>Tmax>0. Highly non smooth target data are considered, that may not correspond to actual solutions at time Tmax. Stabilization is achieved by applying a compensating smoothing operator at each time step. Such smoothing leads to a distortion away from the true solution, but that distortion is small enough to allow for useful results. Data assimilation is illustrated using 512×512 pixel images. Such images are associated with highly irregular non smooth intensity data that severely challenge ill-posed reconstruction procedures. Computational experiments show that efficient FFT-synthesized smoothing operators, based on (−Δ)q with real q>3, can be successfully applied, even in nonlinear problems in non-rectangular domains. However, an example of failure illustrates the limitations of the method in problems where Tmax, and/or the nonlinearity, are not sufficiently small.

Local Rule-Based Explanations Method Based on Genetic Algorithms with Fitness Sharing

The Local Rule Based Explanations method (LORE) explains decisions of black-box classifiers by using an interpretable model (Decision Tree – DT). The DT is trained with an artificial dataset generated by Genetic Algorithms (GAs). The primary objective of this approach is to replicate the decision boundaries of the black-box model in proximity to the instance under explanation. We show that the artificial examples generated by the GAs in LORE are not necessarily diverse. Consequently, we propose the integration of GAs with fitness sharing in LORE to generate a more diversified subset of artificial examples. The underlying motivation is to ensure that the local decision boundaries of the DT more closely resemble those of the black-box classifier. Experimental results with two classifiers (Multilayer Perceptron and Random Forests), and four classification problems, indicate that LORE with fitness sharing yields more diverse GA populations, consequently leading to improved local explanations. These findings underscore the effectiveness of incorporating fitness sharing into the LORE methodology for enhancing the explainability of black-box classifiers.

Augmented Lagrangian Tracking for distributed optimization with equality and inequality coupling constraints

In this paper we propose a novel Augmented Lagrangian Tracking distributed optimization algorithm for solving multi-agent optimization problems where each agent has its own decision variables, cost function and constraint set, and the goal is to minimize the sum of the agents’ cost functions subject to local constraints plus some additional coupling constraint involving the decision variables of all the agents. In contrast to alternative approaches available in the literature, the proposed algorithm jointly features a constant penalty parameter, the ability to cope with unbounded local constraint sets, and the ability to handle both affine equality and nonlinear inequality coupling constraints, while requiring convexity only. The effectiveness of the approach is shown first on an artificial example with complexity features that make other state-of-the-art algorithms not applicable and then on a realistic example involving the optimization of the charging schedule of a fleet of electric vehicles.

A new statistical method for the comparison of biplots with the same objects and variables

Ordinations are compared most commonly by Procrustes methods applicable to points belonging to the same domain, either the objects or the variables describing them. However, no published approach applies to biplots which visualize principal component ordinations of objects and variables simultaneously.To fill this gap, this paper provides a new procedure based on two fundamental, scale-invariant properties of biplots: the cosine of the angle between the vectors pointing to every pair of objects and variables and the mean of the relative length of the two vectors. The method applies to situations in which the number of objects is the same in all ordinations, the number of variables is also the same and, additionally, there is one-to-one correspondence between the points in the different ordinations. The new method is illustrated with two artificial examples and is also applied to the comparison of biplots obtained from real field data representing samples repeated in the same locations over time.It is demonstrated that the new method reveals the similarities and differences between alternative biplots obtained for a given set of objects and variables.We expect that, thanks to the widespread use of principal component analysis in science, the method will receive applications in any studies in which interest lies in the comparison of simultaneous ordinations of objects and variables.

Diagnostics for partially linear measurement error models

Partially linear models are useful tools to analyze data from economic, genetic, and other fields. Similar to other data analyses, the identification of influential observations that may be potential outliers is an important step beyond estimation in such models. The objective of this article is to develop some diagnostic measures for identifying influential observations in partially linear models when some of the covariates are measured with errors. Deletion measures are developed based on case deletion, mean shift outlier models, and the corrected likelihood of Nakamura (1990). The performance of the methods is illustrated by an artificial example and a real example.

The scalar-on-function modal regression for functional time series data

This paper develops a new nonparametric estimator of the scalar-on function modal regression that is used to analyse the co-variability between a functional regressor and a scalar output variable. The new estimator inherits the smoothness of the kernel method and the robustness of the quantile regression. We assume that the functional observations are structured as a strong mixing functional time series data and we establish the almost complete consistency (with rate) of the constructed estimator. A discussion highlighting the impact of this new estimator in nonparametric functional data analysis is also given. The usefulness of this new estimator is shown using an artificial data example.

This population does not exist: learning the distribution of evolutionary histories with generative adversarial networks.

Numerous studies over the last decade have demonstrated the utility of machine learning methods when applied to population genetic tasks. More recent studies show the potential of deep-learning methods in particular, which allow researchers to approach problems without making prior assumptions about how the data should be summarized or manipulated, instead learning their own internal representation of the data in an attempt to maximize inferential accuracy. One type of deep neural network, called Generative Adversarial Networks (GANs), can even be used to generate new data, and this approach has been used to create individual artificial human genomes free from privacy concerns. In this study, we further explore the application of GANs in population genetics by designing and training a network to learn the statistical distribution of population genetic alignments (i.e. data sets consisting of sequences from an entire population sample) under several diverse evolutionary histories-the first GAN capable of performing this task. After testing multiple different neural network architectures, we report the results of a fully differentiable Deep-Convolutional Wasserstein GAN with gradient penalty that is capable of generating artificial examples of population genetic alignments that successfully mimic key aspects of the training data, including the site-frequency spectrum, differentiation between populations, and patterns of linkage disequilibrium. We demonstrate consistent training success across various evolutionary models, including models of panmictic and subdivided populations, populations at equilibrium and experiencing changes in size, and populations experiencing either no selection or positive selection of various strengths, all without the need for extensive hyperparameter tuning. Overall, our findings highlight the ability of GANs to learn and mimic population genetic data and suggest future areas where this work can be applied in population genetics research that we discuss herein.

A Carbodiimide-Fueled Reaction Cycle That Forms Transient 5(4H)-Oxazolones

In life, molecular architectures, like the cytoskeletal proteins or the nucleolus, catalyze the conversion of chemical fuels to perform their functions. For example, tubulin catalyzes the hydrolysis of GTP to form a dynamic cytoskeletal network. In contrast, myosin uses the energy obtained by catalyzing the hydrolysis of ATP to exert forces. Artificial examples of such beautiful architectures are scarce partly because synthetic chemically fueled reaction cycles are relatively rare. Here, we introduce a new chemical reaction cycle driven by the hydration of a carbodiimide. Unlike other carbodiimide-fueled reaction cycles, the proposed cycle forms a transient 5(4H)-oxazolone. The reaction cycle is efficient in forming the transient product and is robust to operate under a wide range of fuel inputs, pH, and temperatures. The versatility of the precursors is vast, and we demonstrate several molecular designs that yield chemically fueled droplets, fibers, and crystals. We anticipate that the reaction cycle can offer a range of other assemblies and, due to its versatility, can also be incorporated into molecular motors and machines.

Assessing Criteria Weights by the Symmetry Point of Criterion (Novel SPC Method)–Application in the Efficiency Evaluation of the Mineral Deposit Multi-Criteria Partitioning Algorithm

Information about the relative importance of each criterion or the weights of criteria can have a significant influence on the ultimate rank of alternatives. Accordingly, assessing the weights of criteria is a very important task in solving multi-criteria decision-making problems. Three methods are commonly used for assessing the weights of criteria: objective, subjective, and integrated methods. In this study, an objective approach is proposed to assess the weights of criteria, called SPC method (Symmetry Point of Criterion). This point enriches the criterion so that it is balanced and easy to implement in the process of the evaluation of its influence on decision-making. The SPC methodology is systematically presented and supported by detailed calculations related to an artificial example. To validate the developed method, we used our numerical example and calculated the weights of criteria by CRITIC, Entropy, Standard Deviation and MEREC methods. Comparative analysis between these methods and the SPC method reveals that the developed method is a very reliable objective way to determine the weights of criteria. Additionally, in this study, we proposed the application of SPC method to evaluate the efficiency of the multi-criteria partitioning algorithm. The main idea of the evaluation is based on the following fact: the greater the uniformity of the weights of criteria, the higher the efficiency of the partitioning algorithm. The research demonstrates that the SPC method can be applied to solving different multi-criteria problems.

Improving accessibility of scientific research by artificial intelligence-An example for lay abstract generation.

The role of scientific research in modern society is essential for driving innovation, informing policy decisions, and shaping public opinion. However, communicating scientific findings to the general public can be challenging due to the technical and complex nature of scientific research. Lay abstracts are written summaries of scientific research that are designed to be easily understandable and provide a concise and clear overview of key findings and implications. Artificial intelligence language models have the potential to generate lay abstracts that are consistent and accurate while reducing the potential for misinterpretation or bias. This study presents examples of artificial intelligence-generated lay abstracts of recently published articles, which were produced using different currently available artificial intelligence tools. The generated abstracts were of high linguistic quality and accurately represented the findings of the original articles. Adopting lay summaries can increase the visibility, impact, and transparency of scientific research, and enhance scientists' reputation among peers, while currently, available artificial intelligence models offer solutions to produce lay abstracts. However, the coherence and accuracy of artificial intelligence language models must be validated before they can be used for this purpose without restrictions.

Assessment of the suitability of the chemical reaction pathway algorithm as a reduction method for plasma chemistry

Determination of chemical pathways, sets of interlinked reactions, is a well-known method to study complex chemistries. In order to have a trustworthy and accurate method for analysis of chemical processes in complicated systems such as low-temperature plasmas, a used algorithm should keep conservation of necessary parameters during all steps of an analysis. In this research, the introduced algorithm by Lehmann (2004 J. Atmos. Chem. 47 45–78) is chosen as a candidate for studying tangled low-temperature plasma chemistry. As a first step, the ability of the algorithm to keep conservation of reactions rates and net production of species is assessed by designing some artificial examples. In that stage, ambiguous parts of the algorithm are clarified, and test cases are provided to verify implementation of the algorithm in any programming infrastructure. Then, the capability of the algorithm as a reduction method is investigated by applying a newly developed semi-automated method based on pathway analysis to two plasma chemistry examples, a – microwave induced plasma and a pulsed plasma.