Error Rate Estimation Research Articles

Genotyping Error Detection and Customised Filtration for SNP Datasets.

A major challenge in analysing single-nucleotide polymorphism (SNP) genotype datasets is detecting and filtering errors that bias analyses and misinterpret ecological and evolutionary processes. Here, we present a comprehensive method to estimate and minimise genotyping error rates (deviations from the 'true' genotype) in any SNP datasets using triplicates (three repeats of the same sample) in a four-step filtration pipeline. The approach involves: (1) SNP filtering by missing data; (2) SNP filtering by error rates; (3) sample filtering by missing data and (4) detection of recaptured individuals by using estimated SNP error rates. The modular pipeline is provided in an R script that allows customised adjustments. We demonstrate the applicability of the method using non-invasive sampling from the Asiatic wild ass (Equus hemionus) population in Israel. We genotyped 756 samples using 625 SNPs, of which 255 were triplicates of 85 samples. The average SNP error rate, calculated based on the number of mismatching genotypes across triplicates before filtration, was 0.0034 and was reduced to 0.00174 following filtration. Evaluating genetic distance (GD) and relatedness (r) between triplicates before and after filtration (expected to be at the minimum and maximum respectively) showed a significant reduction in the average GD, from 58.1 to 25.3 (p = 0.0002) and a significant increase in relatedness, from r = 0.98 to r = 0.991 (p = 0.00587). We demonstrate how error rate estimation enhances recapture detection and improves genotype quality.

Estimating error rates for single molecule protein sequencing experiments.

The practical application of new single molecule protein sequencing (SMPS) technologies requires accurate estimates of their associated sequencing error rates. Here, we describe the development and application of two distinct parameter estimation methods for analyzing SMPS reads produced by fluorosequencing. A Hidden Markov Model (HMM) based approach, extends whatprot, where we previously used HMMs for SMPS peptide-read matching. This extension offers a principled approach for estimating key parameters for fluorosequencing experiments, including missed amino acid cleavages, dye loss, and peptide detachment. Specifically, we adapted the Baum-Welch algorithm, a standard technique to estimate transition probabilities for an HMM using expectation maximization, but modified here to estimate a small number of parameter values directly rather than estimating every transition probability independently. We demonstrate a high degree of accuracy on simulated data, but on experimental datasets, we observed that the model needed to be augmented with an additional error type, N-terminal blocking. This, in combination with data pre-processing, results in reasonable parameterizations of experimental datasets that agree with controlled experimental perturbations. A second independent implementation using a hybrid of DIRECT and Powell's method to reduce the root mean squared error (RMSE) between simulations and the real dataset was also developed. We compare these methods on both simulated and real data, finding that our Baum-Welch based approach outperforms DIRECT and Powell's method by most, but not all, criteria. Although some discrepancies between the results exist, we also find that both approaches provide similar error rate estimates from experimental single molecule fluorosequencing datasets.

Efficient Error-Rate Estimation for Optical Transmission Systems Using Artificial Neural Networks

ABSTRACT With the advancement in modulation formats, optical transmission systems are becoming more adaptable and dynamic. Predicting bit errors of higher-order modulation formats is a challenging problem for the optimum design of communication systems. The existing simulation methods have persisted by employing iterative procedures that frequently incur high computing expenses and time consumption. On the other hand, deep learning (DL) algorithms have demonstrated remarkable efficacy as practical computing tools, presenting a viable approach to accelerate modulation simulations. In this paper, an artificial neural network (ANN) based bit error rate (BER) estimation scheme is proposed for the popular modulation forms including 56 Gbps 16-quadrature amplitude modulation (16QAM), 100 Gbps 32QAM, and 120 Gbps 64QAM optical system. Based on constellation diagrams (CDs) acquired with different launch power, laser linewidth, transmission distance, and OSNR, amplitude histograms (AHs) are generated via a manual preprocessing method. The properly trained ANN architecture exhibits a computational speed that surpasses traditional simulation methods by a factor exceeding 347. Moreover, the design considerations including the number of layers, nodes, activation functions, learning rate, optimizers, and evolution epochs are also investigated in detail. This research paves the way for optical transmission systems to use fast DL-based optimization strategies.

Simultaneous estimation of genotype error and uncalled deletion rates in whole genome sequence data.

Genotype data include errors that may influence conclusions reached by downstream statistical analyses. Previous studies have estimated genotype error rates from discrepancies in human pedigree data, such as Mendelian inconsistent genotypes or apparent phase violations. However, uncalled deletions, which generally have not been accounted for in these studies, can lead to biased error rate estimates. In this study, we propose a genotype error model that considers both genotype errors and uncalled deletions when calculating the likelihood of the observed genotypes in parent-offspring trios. Using simulations, we show that when there are uncalled deletions, our model produces genotype error rate estimates that are less biased than estimates from a model that does not account for these deletions. We applied our model to SNVs in 77 sequenced White British parent-offspring trios in the UK Biobank. We use the Akaike information criterion to show that our model fits the data better than a model that does not account for uncalled deletions. We estimate the genotype error rate at SNVs with minor allele frequency > 0.001 in these data to be [Formula: see text]. We estimate that 77% of the genotype errors at these markers are attributable to uncalled deletions [Formula: see text].

ArCH: improving the performance of clonal hematopoiesis variant calling and interpretation.

The acquisition of somatic mutations in hematopoietic stem and progenitor stem cells with resultant clonal expansion, termed clonal hematopoiesis (CH), is associated with increased risk of hematologic malignancies and other adverse outcomes. CH is generally present at low allelic fractions, but clonal expansion and acquisition of additional mutations leads to hematologic cancers in a small proportion of individuals. With high depth and high sensitivity sequencing, CH can be detected in most adults and its clonal trajectory mapped over time. However, accurate CH variant calling is challenging due to the difficulty in distinguishing low frequency CH mutations from sequencing artifacts. The lack of well-validated bioinformatic pipelines for CH calling may contribute to lack of reproducibility in studies of CH. Here, we developed ArCH, an Artifact filtering Clonal Hematopoiesis variant calling pipeline for detecting single nucleotide variants and short insertions/deletions by combining the output of four variant calling tools and filtering based on variant characteristics and sequencing error rate estimation. ArCH is an end-to-end cloud-based pipeline optimized to accept a variety of inputs with customizable parameters adaptable to multiple sequencing technologies, research questions, and datasets. Using deep targeted sequencing data generated from six acute myeloid leukemia patient tumor: normal dilutions, 31 blood samples with orthogonal validation, and 26 blood samples with technical replicates, we show that ArCH improves the sensitivity and positive predictive value of CH variant detection at low allele frequencies compared to standard application of commonly used variant calling approaches. The code for this workflow is available at: https://github.com/kbolton-lab/ArCH.

Biomedical signal compression using deep learning based multi-task compressed sensing

<span>Real-time transmission of biomedical signals is immensely challenging and requires cloud and internet of things (IoT) infrastructure. Security is also an important factor; however, to accomplish this, a reconstruction method is developed in which the entire signal is supplied as an input, the primary portion is considered here, and the signal is further encoded and transmitted to the destination. Electrocardiogram (ECG) compression for the lightweight wireless network is quite challenging for long-term healthcare monitoring. Compressed sensing (CS) involves efficient encoding mechanisms for error rate estimation for reconstruction and energy consumption for wireless transmission of data. We propose a multi-task compressed sensing (MT-CS) reconstruction mechanism in this study for ECG compression of data is most chosen for a wireless network system that has various sensors embedded in it. This model further extracts the essential adaptive features for correlation existing in the ECG signals. The performance of the proposed MT-CS reconstruction mechanism is evaluated on the multiparameter intelligent monitoring in intensive care (MIMIC-II) dataset, which ensures its robustness and generalization. The results obtained upon simulation ensure that the proposed MT-CS based reconstruction approach ensures excellent reconstruction signal with fewer measurements in comparison with the existing state-of-art CS techniques.</span>

An Improved Model of Single-Event Transients Based on Effective Space Charge for Metal-Oxide-Semiconductor Field-Effect Transistor.

In this paper, a single-event transient model based on the effective space charge for MOSFETs is proposed. The physical process of deposited and moving charges is analyzed in detail. The influence of deposited charges on the electric field in the depletion region is investigated. The electric field decreases in a short time period due to the neutralization of the space charge. After that, the electric field increases first and then decreases when the deposited charge is moved out. The movement of the deposited charge in the body mainly occurs through ambipolar diffusion because of its high-density electrons and holes. The derivation of the variation in electric field in the depletion region is modeled in the physical process according to the analysis. In combination with the ambipolar diffusion model of excessive charge in the body, a physics-based model is built to describe the current pulse in the drain terminal. The proposed model takes into account the influence of multiple factors, like linear-energy transfer (LET), drain bias, and the doping concentration of the well. The model results are validated with the simulation results from TCAD. Through calculation, the root-mean-square error (RMSE) between the simulation and model is less than 3.7 × 10-4, which means that the model matches well with the TCAD results. Moreover, a CMOS inverter is simulated using TCAD and SPICE to validate the applicability of the proposed model in a circuit-level simulation. The proposed model captures the variation in net voltage in the inverter. The simulation result obviously shows the current plateau effect, while the relative error of the pulse width is 23.5%, much better than that in the classic model. In comparison with the classic model, the proposed model provides an RMSE of 7.59 × 10-5 for the output current curve and an RMSE of 0.158 for the output voltage curve, which are significantly better than those of the classic model. In the meantime, the proposed model does not produce extra simulation time compared with the classic double exponential model. So, the model has potential for application to flow estimation of the soft error rate (SER) at the circuit level to improve the accuracy of the results.

Characteristics of Channel Spreading Function and Performance of OTFS in High-Speed Railway

Orthogonal time frequency space (OTFS) modulation is an emerging technology to tackle time-frequency (TF) selective channel in high mobility scenarios. In OTFS, resource is multiplexed in the delay-Doppler (DD) domain. Based on the potential sparsity, separability, stability and compactness of the channel spreading function, OTFS is able to realize lower complexity of channel estimation, higher diversity and higher reliability compared with orthogonal frequency division multiplexing (OFDM). However, the channel spreading function for practical communication systems is rarely considered in the current OTFS-related literature. High-speed railway (HSR) is a typical high mobility scenario with trains travelling at over 200km/h, which has the potential to employ OTFS. To this end, the HSR channel spreading function is characterized and the performance of OTFS in HSR is evaluated based on the realistic channel measurement in this article. Firstly, the HSR channel in TF domain is measured based on the long term evolution railway (LTE-R) network. Then, the characteristics of the channel spreading function are analyzed. In particular, the impact of time domain channel fading on the spreading function is investigated. The characteristics of the measured spreading function are analyzed with the proposed metrics in railway viaduct and tunnel scenarios. Based on the above analysis, an algorithm for generating the channel spreading function is proposed. Feasibility of the proposed DD domain channel generation algorithm is verified through comparing metrics of which to those of the measured channel. By simulating the bit error rate (BER) and mean square channel estimation error performances of OTFS modulation in the practical band-limited systems, it is shown that the impacts of Doppler shift, delay, SFFT and time domain channel fading need be considered for the application of OTFS modulation, in contrast to the state-of-art DD domain channel generation scheme based on tap delay link (TDL) model. For example, compared to OTFS modulation under the ideal channel spreading function, OTFS modulation requires a signal gain greater than 5 dB under the practical channel spreading function affected by above factors, to achieve the same BER less than 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> under the parameters and algorithms given in section V.

B-095 Determination of Error Rates with Duplicate Analysis in Anti-Drug Antibody Testing

Abstract Background Duplicate measurements can be used to characterize assay imprecision, identify outliers, and estimate error rates. The benefits of duplicate analysis, however, need to be balanced against the increased financial cost. The objective of this study was to evaluate the need for continued replicate measurement in a laboratory developed assay measuring anti-drug antibodies to infliximab (ATI). To this end, duplicate patient sample measurements from the ATI assay over a two-year period were analyzed to calculate outlier and error frequencies. Methods ATI were measured by electrochemiluminescence (ECL) on the MesoScale Discovery (MSD) platform, with a standard curve from 20–1250 U/mL using commercially available monoclonal ATIs. All patient samples tested for ATI over a 27-month period of time (n = 9582) were subject to inclusion in this study. Only patient samples with ATI levels within the analytical measuring range (AMR) of the assay, however, were selected for further analysis. Samples were divided equally into 7 bins across the measuring range based on mean replicate concentration. Within-run imprecision was determined for each bin using the standard deviation (combined from all replicates using sum of squares) and mean concentration derived from all replicates. Outliers were defined as replicate measurements with a percent difference greater than ±3 SD of the calculated assay imprecision. Errors were defined as samples which had replicate values on opposite sides of the clinical cutoff of positive ATI (≥50 U/mL) and a percent difference greater than 3SD of the assay imprecision. Results Of the 9582 total patient samples analyzed, 2467 had ATI values within the AMR and 1655 had levels over the clinical cut-off of 50 U/mL (17.3% positivity rate). Within-run imprecision determined from the mean and combined SD of the duplicate measurements ranged from 15.7% at the lowest concentration bin (20–27.5 U/mL) to 10.0% at the highest concentration bin (358–1243 U/mL). The mean percentage of outliers in each bin ranged from 0.8%–3.4% resulting in an overall outlier frequency of 2.1% when considering all samples within the AMR. There were 66 samples within the AMR which demonstrated qualitative discrepancies between the replicates. However, only 21 (31.8%) of these samples, had a percent agreement greater than 3SD of the assay imprecision resulting in an overall error rate of 0.2% when considering all 9582 samples. Conclusion In this study, the analysis of replicate patient samples enabled the establishment of a precision profile and estimation of error rates for a laboratory developed test measuring ATI. The calculated error rate of 0.2%, however, is well within the reported range of most clinical assays indicating that the continued practice of duplicate measurement may not be necessary. Additionally, updating the laboratory protocol to eliminate replicate testing would have positive impact in laboratory workflow, cost, and turn-around-time for results.

Estimation of the net error rate of population size in China’s household registration

ABSTRACT A combined three-source estimator is used to estimate the net error rate of population size in household registration. It avoids the bias in the two-source estimator due to the fact that the registration of an individual on one list affects his or her registration on another list. It makes better use of frequencies than the ordinary three-source estimator. It yields an estimated net error rate that is closest to 1% of error. The combined three-source estimator combines missing-cell estimators of ordinary three-source estimators working on the contingency table of presence or absence in the census population list, the post-enumeration survey population list, and the household population list. Matching of these three registration lists is filled into the seven non-missing cells of the contingency table. The missing-cell estimator involved in the combined three-source estimator is based on the missing-cell estimators of the ordinary three-source estimators. It requires that the three population lists fully register the post-strata population. In the case of sample registration, it is based on the sampling weights and population sizes of the sampled small census areas. It provides a lower mean square error than the ordinary three-source estimator, which itself provides a lower mean square error than the two-source estimator. The estimated 0.96% net under-registration rate of population size for the country as a whole in household registration is close to the 1.00% rate found in the rectification of household registration carried out by China’s Ministry of Public Security between May 15, 2010, and September 30, 2010. Despite the bias coming from the residual heterogeneity of individuals in each post-stratum, as the sample can only accommodate a limited total number of post-strata, the combined three-source estimator should yield a smaller mean square error than the ordinary three-source estimator of the net error rate of population size in household registration.

Single-cell mutation rate of turnip crinkle virus (-)-strand replication intermediates.

Viruses with single-stranded, positive-sense (+) RNA genomes incur high numbers of errors during replication, thereby creating diversified genome populations from which new, better adapted viral variants can emerge. However, a definitive error rate is known for a relatively few (+) RNA plant viruses, due to challenges to account for perturbations caused by natural selection and/or experimental set-ups. To address these challenges, we developed a new approach that exclusively profiled errors in the (-)-strand replication intermediates of turnip crinkle virus (TCV), in singly infected cells. A series of controls and safeguards were devised to ensure errors inherent to the experimental process were accounted for. This approach permitted the estimation of a TCV error rate of 8.47 X 10-5 substitution per nucleotide site per cell infection. Importantly, the characteristic error distribution pattern among the 50 copies of 2,363-base-pair cDNA fragments predicted that nearly all TCV (-) strands were products of one replication cycle per cell. Furthermore, some of the errors probably elevated error frequencies by lowering the fidelity of TCV RNA-dependent RNA polymerase, and/or permitting occasional re-replication of progeny genomes. In summary, by profiling errors in TCV (-)-strand intermediates incurred during replication in single cells, this study provided strong support for a stamping machine mode of replication employed by a (+) RNA virus.

МОДЕЛИРОВАНИЕ ПРОЦЕССОВ ВОЗДЕЙСТВИЯ ИМПУЛЬСНЫХ ПОМЕХ ПЕРЕМЕННОЙ ДЛИТЕЛЬНОСТИ НА СИСТЕМУ СВЯЗИ СТАНДАРТА IEEE 802.15.4-2020

The noise immunity of radio communication systems based on IEEE 802.15.4–2020 and IEEE 802.15.4z–2020 standards under interference effect of variable-duration impulse noise is investigated. The simulation model of a radio communication channel for estimation of the bit error rate for noise with different parameters is developed using the MATLAB. The estimates of the noise immunity for different standard modes taking into account the structure of transmitted data are obtained. The comparison between the characteristics of standard modes is performed. A developed model can be modified for estimation of the noise immunity under interference effect of others types of noise. The obtained results of noise immunity estimation can be used for development of perspective ultra-wideband radio communication systems and increasing the efficiency of existing radio communication systems.

Estimation of eyewitness error rates in fair and biased lineups.

The risk of mistaken identification for innocent suspects in lineups can be estimated by correcting the overall error rate by the number of people in the lineup. We compared this nominal size correction to a new effective size correction, which adjusts the error rate for the number of plausible lineup members. We hypothesized that (a) increasing lineup bias would increase misidentifications of a designated innocent suspect; (b) with the effective size correction, increasing lineup bias would also increase the estimate of innocent-suspect misidentifications; and (c) with the nominal size correction, lineup bias would have no effect on the estimate of innocent-suspect misidentifications. In a reanalysis of previous literature, we obtained 10 data sets from Open Science Framework. In three new experiments (Ns = 686, 405, and 1,531, respectively), participants observed a staged crime and completed a fair or biased lineup. In the reanalysis of previous literature, less than four of six lineup members were identified frequently enough to be classified as plausible, M = 3.78, 95% confidence interval [CI: 2.20, 5.36]. In the new experiments, increasing lineup bias increased mistaken identifications of a designated innocent suspect, odds ratio (OR) = 5.50, 95% CI [2.77, 10.95] and also increased the effective size-corrected estimate of innocent-suspect misidentifications, OR = 3.04, 95% CI [2.13, 4.33]. With the nominal size correction, lineup bias had no effect on the estimate of innocent-suspect misidentifications, OR = 0.84, 95% CI [0.60, 1.18]. Most lineups include a combination of plausible and implausible lineup members. Contrary to the nominal size correction, which ignores implausible lineup members, the effective size correction is sensitive to implausible lineup members and accounts for lineup bias when estimating the risk of innocent suspect misidentifications. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Counting and boundary limit theorems for representations of Gromov‐hyperbolic groups

AbstractGiven a Gromov‐hyperbolic group endowed with a finite symmetric generating set, we study the statistics of counting measures on the spheres of the associated Cayley graph under linear representations of . More generally, we obtain a weak law of large numbers for subadditive functions, echoing the classical Fekete lemma. For strongly irreducible and proximal representations, we prove a counting central limit theorem with a Berry–Esseen type error rate and exponential large deviation estimates. Moreover, in the same setting, we show convergence of interpolated normalized matrix norms along geodesic rays to Brownian motion and a functional law of iterated logarithm, paralleling the analogous results in the theory of random matrix products. Our counting large deviation estimates address a question of Kaimanovich–Kapovich–Schupp. In most cases, our counting limit theorems will be obtained from stronger almost sure limit laws for Patterson–Sullivan measures on the boundary of the group.

Amino acid sequence assignment from single molecule peptide sequencing data using a two-stage classifier.

We present a machine learning-based interpretive framework (whatprot) for analyzing single molecule protein sequencing data produced by fluorosequencing, a recently developed proteomics technology that determines sparse amino acid sequences for many individual peptide molecules in a highly parallelized fashion. Whatprot uses Hidden Markov Models (HMMs) to represent the states of each peptide undergoing the various chemical processes during fluorosequencing, and applies these in a Bayesian classifier, in combination with pre-filtering by a k-Nearest Neighbors (kNN) classifier trained on large volumes of simulated fluorosequencing data. We have found that by combining the HMM based Bayesian classifier with the kNN pre-filter, we are able to retain the benefits of both, achieving both tractable runtimes and acceptable precision and recall for identifying peptides and their parent proteins from complex mixtures, outperforming the capabilities of either classifier on its own. Whatprot's hybrid kNN-HMM approach enables the efficient interpretation of fluorosequencing data using a full proteome reference database and should now also enable improved sequencing error rate estimates.

Shining a Light on Forensic Black-Box Studies

Forensic science plays a critical role in the United States criminal legal system. For decades, many feature-based fields of forensic science, such as firearm and toolmark identification, developed outside the scientific community’s purview. The results of these studies are widely relied on by judges nationwide. However, this reliance is misplaced. Black-box studies to date suffer from inappropriate sampling methods and high rates of missingness. Current black-box studies ignore both problems in arriving at the error rate estimates presented to courts. We explore the impact of each type of limitation using available data from black-box studies and court materials. We show that black-box studies rely on unrepresentative samples of examiners. Using a case study of a popular ballistics study, we find evidence that these nonrepresentative samples may commit fewer errors than the wider population from which they came. We also find evidence that the missingness in black-box studies is non-ignorable. Using data from a recent latent print study, we show that ignoring this missingness likely results in systematic underestimates of error rates. Finally, we offer concrete steps to overcome these limitations. Supplementary materials for this article areavailable online.

Semi-blind receivers based on a coupled nested Tucker-PARAFAC model for dual-polarized MIMO systems using combined TST and MSMKron codings

Over the past two decades, tensor-based approaches have generated ever-increasing interest for designing multiple-input multiple-output (MIMO) communication systems. In this paper, we propose a new double directional dual-polarized (DD-DP) MIMO system, equipped with uniform rectangular arrays (URAs) at both ends of the link, which combines a multiple Kronecker product of symbol matrices with a tensor space-time (TST) coding, called TST-MSMKron coding. It is first shown that the channel tensor is represented by a fifth-order PARAFAC decomposition, separated into two parts associated with vertically and horizontally polarized receive antennas, respectively. Then, it is established that the tensor of received signals satisfies a new coupled nested Tucker-PARAFAC model whose core tensor is the coding tensor. Capitalizing on this tensor modeling and assuming the tensor coding is known at the reception, two semi-blind receivers composed of two stages are derived for estimating the transmitted symbols and channel parameters (directions of departure (DoD) and arrival (DoA) angles, path loss coefficients). A new rectification method is proposed to ensure the Vandermonde structure of the estimated steering matrices. Parameter identifiability and computational complexity are analyzed for each receiver. Monte Carlo simulation results are provided to illustrate the effectiveness of the proposed TST-MSMKron coding and semi-blind receivers, in terms of symbol error rate (SER) and channel parameters estimation.

Maple species identification based on leaf hyperspectral imaging data

The present study aimed to identify the leaves of Acer campestre L., A. negundo L. and A. saccharinum L. The experiment was carried out under controlled laboratory conditions in 2021. A Cubert UHD-185 camera was used for hyperspectral imaging (HSI). The content of photosynthetic pigments was determined by the spectrophotometric method. HSI and determination of pigments were carried out during the growing season of maples with intervals between experiments of 7–10 days. The time interval in the vegetation of maples was determined, in which the identification of their species by the random forest (RF) method occurs most accurately. Based on the strength of the correlation with photosynthetic pigments, factor loadings of principal component analysis (PCA), results of the t-test and analysis of variance (ANOVA), a group of vegetation indices (VIs) was selected, the values of which are associated with the species characteristics of maples. It has been established that VIs CRI1, Carter4, Datt3, DPI, EVI, Gitelson2, GMI1, Green_NDVI, MTVI, PRI, REP_Li, TCARI, TVI are significant for the identification of A. campestre, A. negundo and A. saccharinum. The OOB estimate of error rate does not exceed 7% when classified by the RF method using selected VIs. When training on the data of Jul 02, the testing error in the time interval from Jul 09 to Aug 24 did not exceed 20%.

Line Balancing Based on Error Rate Estimation with Artificial Neural Networks in Assembly Line Operations

In this study, in the assembly line systems consisting of the operations in interaction with each other; To reduce the number of faulty products, to prevent poor quality and to reduce the production time, Error Ratio Estimation with Artificial Neural Networks and probabilistic Line Balancing method have been performed. The error rate estimation provides information on which jeans models should be applied in the improvement work to eliminate existing errors in place. In the study, using the Levenberg - Marquardt Learning Algorithm, machine learning was determined by the experimental design method. At the same time, it has been used as an artificial intelligence algorithm in the multi-directional decision making stages, estimation and line balancing parts. In Assembly Line Equilibration, it has been aimed to re-stabilize the unbalanced line with the influence of post-forecasting process recovery. The Probabilistic Line Balancing method has been used because the processing times are stochastic (variable) and statistical data and mathematical algorithms (digital algorithms can be created). When the results are examined, a successful forecasting process has been carried out for two different five-pocket jeans models which has been selected and it has been seen that the work components of the probabilistic line balancing method enable it to be precisely assigned to work stations. And it has given reliable results.

Hierarchical Bayesian non-response models for error rates in forensic black-box studies.

Forensic science plays a critical role in the United States criminal legal system. Historically, however, most feature-based fields of forensic science, including firearms examination and latent print analysis, have not been shown to be scientifically valid. Recently, black-box studies have been proposed as a means of assessing whether these feature-based disciplines are valid, at least in terms of accuracy, reproducibility and repeatability. In these studies, forensic examiners frequently either do not respond to every test item or select an answer equivalent to 'don't know'. Current black-box studies do not account for these high levels of missingness in statistical analyses. Unfortunately, the authors of black-box studies typically do not share the data necessary to meaningfully adjust estimates for the high proportion of missing responses. Borrowing from work in the context of small area estimation, we propose the use of hierarchical Bayesian models that do not require auxiliary data to adjust for non-response. Using these models, we offer the first formal exploration of the impact that missingness is playing in error rate estimations reported in black-box studies. We show that error rates currently reported as low as 0.4% could actually be at least 8.4% in models accounting for non-response where inconclusive decisions are counted as correct, and over 28% when inconclusives are counted as missing responses. These proposed models are not the answer to the missingness problem in black-box studies. But with the release of auxiliary information, they can be the foundation for new methodologies to adjust for missingness in error rate estimations. This article is part of the theme issue 'Bayesian inference: challenges, perspectives, and prospects'.