Browse
Submit Data
Databases
API
Help

Dataset Information

13 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

People underestimate the errors made by algorithms for credit scoring and recidivism prediction but accept even fewer errors.

ABSTRACT: This study provides the first representative analysis of error estimations and willingness to accept errors in a Western country (Germany) with regards to algorithmic decision-making systems (ADM). We examine people's expectations about the accuracy of algorithms that predict credit default, recidivism of an offender, suitability of a job applicant, and health behavior. Also, we ask whether expectations about algorithm errors vary between these domains and how they differ from expectations about errors made by human experts. In a nationwide representative study (N = 3086) we find that most respondents underestimated the actual errors made by algorithms and are willing to accept even fewer errors than estimated. Error estimates and error acceptance did not differ consistently for predictions made by algorithms or human experts, but people's living conditions (e.g. unemployment, household income) affected domain-specific acceptance (job suitability, credit defaulting) of misses and false alarms. We conclude that people have unwarranted expectations about the performance of ADM systems and evaluate errors in terms of potential personal consequences. Given the general public's low willingness to accept errors, we further conclude that acceptance of ADM appears to be conditional to strict accuracy requirements.

SUBMITTER: Rebitschek FG

PROVIDER: S-EPMC8505498 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Similar Datasets

Reject inference methods in credit scoring.

Project description:The granting process is based on the probability that the applicant will refund his/her loan given his/her characteristics. This probability, also called score, is learnt based on a dataset in which rejected applicants are excluded. Thus, the population on which the score is used is different from the learning population. Many "reject inference" methods try to exploit the data available from the rejected applicants in the learning process. However, most of these methods are empirical and lack of formalization of their assumptions, and of their expected theoretical properties. We formalize such hidden assumptions in a general missing data setting for some of the most common reject inference methods. It reveals that hidden modelling is mostly incomplete, thus prohibiting to compare existing methods within the general model selection mechanism (except by financing "non-fundable" applicants). So, we assess performance of the methods on both simulated data and real data (from CACF, a major European loan issuer). Unsurprisingly, no method seems uniformly dominant. Both these theoretical and empirical results not only reinforce the idea to carefully use the classical reject inference methods but also to invest in future research works for designing model-based reject inference methods (without financing "non-fundable" applicants).

| S-EPMC9041715 | biostudies-literature

DNA polymerase ? proofreads errors made by DNA polymerase ?.

Project description:During eukaryotic replication, DNA polymerases ? (Pol?) and ? (Pol?) synthesize the leading and lagging strands, respectively. In a long-known contradiction to this model, defects in the fidelity of Pol? have a much weaker impact on mutagenesis than analogous Pol? defects. It has been previously proposed that Pol? contributes more to mutation avoidance because it proofreads mismatches created by Pol? in addition to its own errors. However, direct evidence for this model was missing. We show that, in yeast, the mutation rate increases synergistically when a Pol? nucleotide selectivity defect is combined with a Pol? proofreading defect, demonstrating extrinsic proofreading of Pol? errors by Pol?. In contrast, combining Pol? nucleotide selectivity and Pol? proofreading defects produces no synergy, indicating that Pol? cannot correct errors made by Pol?. We further show that Pol? can remove errors made by exonuclease-deficient Pol? in vitro. These findings illustrate the complexity of the one-strand-one-polymerase model where synthesis appears to be largely divided, but Pol? proofreading operates on both strands.

| S-EPMC7084062 | biostudies-literature

Plant breeding can be made more efficient by having fewer, better crosses.

Project description:BackgroundCrop yields have to increase to provide food security for the world's growing population. To achieve these yield increases there will have to be a significant contribution from genetic gains made by conventional plant breeding. However, the breeding process is not efficient because crosses made between parental combinations that fail to produce useful varieties consume over 99% of the resources.ResultsWe tested in a rice-breeding programme if its efficiency could be improved by using many fewer, but more judiciously chosen crosses than usual. In a 15-year programme in Nepal, with varietal testing also in India and Bangladesh, we made only six crosses that were stringently chosen on complementary parental performance. We evaluated their success by the adoption and official release of the varieties they produced. We then modelled optimum cross number using assumptions based on our experimental results.Four of the six crosses succeeded. This was a fifty-fold improvement over breeding programmes that employ many crosses where only about one, or fewer, crosses in 200 succeed. Based on these results, we modelled the optimum number of crosses by assuming there would be a decline in the reliability of the breeder's prediction of the value of each cross as more crosses were made (because there is progressively less information on the traits of the parents). Fewer-cross programmes were more likely to succeed and did so using fewer resources. Making more crosses reduced the overall probability of success of the breeding programme.ConclusionsThe efficiency of national and international breeding programmes would be increased by making fewer crosses among more carefully chosen parents. This would increase the number of higher yielding varieties that are delivered to farmers and hence help to improve food security.

| S-EPMC3621587 | biostudies-literature

A novel method for credit scoring based on feature transformation and ensemble model.

Project description:Credit scoring is a very critical task for banks and other financial institutions, and it has become an important evaluation metric to distinguish potential defaulting users. In this paper, we propose a credit score prediction method based on feature transformation and ensemble model, which is essentially a cascade approach. The feature transformation process consisting of boosting trees (BT) and auto-encoders (AE) is employed to replace manual feature engineering and to solve the data imbalance problem. For the classification process, this paper designs a heterogeneous ensemble model by weighting the factorization machine (FM) and deep neural networks (DNN), which can efficiently extract low-order intersections and high-order intersections. Comprehensive experiments were conducted on two standard datasets and the results demonstrate that the proposed approach outperforms existing credit scoring models in accuracy.

| S-EPMC8189024 | biostudies-literature

Asymptotic Standard Errors of Generalized Partial Credit Model True Score Equating Using Characteristic Curve Methods.

Project description:In this study, the delta method was applied to estimate the standard errors of the true score equating when using the characteristic curve methods with the generalized partial credit model in test equating under the context of the common-item nonequivalent groups equating design. Simulation studies were further conducted to compare the performance of the delta method with that of the bootstrap method and the multiple imputation method. The results indicated that the standard errors produced by the delta method were very close to the criterion empirical standard errors as well as those yielded by the bootstrap method and the multiple imputation method under all the manipulated conditions.

| S-EPMC8361376 | biostudies-literature

Longer metaphase and fewer chromosome segregation errors in modern human than Neanderthal brain development.

Project description:Since the ancestors of modern humans separated from those of Neanderthals, around 100 amino acid substitutions spread to essentially all modern humans. The biological significance of these changes is largely unknown. Here, we examine all six such amino acid substitutions in three proteins known to have key roles in kinetochore function and chromosome segregation and to be highly expressed in the stem cells of the developing neocortex. When we introduce these modern human-specific substitutions in mice, three substitutions in two of these proteins, KIF18a and KNL1, cause metaphase prolongation and fewer chromosome segregation errors in apical progenitors of the developing neocortex. Conversely, the ancestral substitutions cause shorter metaphase length and more chromosome segregation errors in human brain organoids, similar to what we find in chimpanzee organoids. These results imply that the fidelity of chromosome segregation during neocortex development improved in modern humans after their divergence from Neanderthals.

| S-EPMC9337762 | biostudies-literature

Reducing errors made by emergency physicians in interpreting radiographs: longitudinal study.

Project description:ObjectivesTo reduce errors made in the interpretation of radiographs in an emergency department.DesignLongitudinal study.SettingHospital emergency department.InterventionsAll staff reviewed all clinically significant discrepancies at monthly meetings. A file of clinically significant errors was created; the file was used for teaching. Later a team redesigned the process. A system was developed for interpreting radiographs that would be followed regardless of the day of the week or time of day. All standard radiographs were brought directly to the emergency physician for immediate interpretation. Radiologists reviewed the films within 12 hours as a quality control measure, and if a significant misinterpretation was found patients were asked to return.Main outcome measuresReduction in number of clinically significant errors (such as missed fractures or foreign bodies) on radiographs read in the emergency department. Data on the error rate for radiologists and the effect of the recall procedure were not available so reliability modelling was used to assess the effect of these on overall safety.ResultsAfter the initial improvements the rate of false negative errors fell from 3% (95% confidence interval 2.8% to 3.2%) to 1.2% (1.03% to 1.37%). After the processes were redesigned it fell further to 0.3% (0.26% to 0.34%). Reliability modelling showed that the number of potential adverse effects per 1000 cases fell from 19 before the improvements to 3 afterwards and unmitigated adverse effects fell from 2.2/1000 before to 0.16/1000 afterwards, assuming 95% success in calling patients back.ConclusionSystems of radiograph interpretation that optimise the skills of all clinicians involved and contain reliable processes for mitigating errors can reduce error rates substantially.

| S-EPMC27314 | biostudies-literature

Small-scale forestry and carbon offset markets: An empirical study of Vermont Current Use forest landowner willingness to accept carbon credit programs.

Project description:This study investigates the preferences of small forest landowners regarding forest carbon credit programs while documenting characteristics of potentially successful frameworks. We designed hypothetical carbon credit programs with aggregated carbon offset projects and requirements of existing voluntary and compliance protocols in mind. We administered a mail survey to 992 forest landowners in Vermont's Current Use Program utilizing best-worst choice, a novel preference elicitation technique, to elicit their preferences about these programs. We found that small forest landowners see revenue as the most important factor in a carbon credit program and the duration of the program as the least important factor. Landowners reported that shorter program duration, higher revenue, and lower withdrawal penalties positively impact their willingness to accept forest carbon credit programs. Notably, our study includes carbon credit program implementer as a key program attribute, allowing us to quantify landowners' tradeoffs between non-profit, for-profit, and government organizations. Overall, we found that landowners significantly prefer working with a non-profit organization. Based on monetary estimates of willingness-to-accept compensation, our results suggest that aggregated forest carbon offset projects incorporating small forest landowners could be piloted successfully in Vermont by non-profit organizations while maintaining relatively strict guidelines of existing carbon offset protocols.

| S-EPMC6091951 | biostudies-literature

A credit scoring model based on the Myers-Briggs type indicator in online peer-to-peer lending.

Project description:Although psychometric features have been considered for alternative credit scoring, they have not yet been applied to peer-to-peer (P2P) lending because such information is not available on platforms. This study proposed an alternative credit scoring model for P2P lending by extracting typical personality types inferred from the borrowers' job category. We projected a virtual space of borrowers by using the affinity matrix based on the Myers-Briggs type indicator (MBTI) that fits each job category. Applying the distance in this space to Lending Club data, we used locally weighted logistic regression to vary the coefficients of the variables, which affect loan repayments, with each MBTI type for predicting the default probability. We found that each MBTI type's credit scoring model has different significant variables. This study provides insights into breakthroughs in developing alternative credit scoring for P2P lending.Supplementary informationThe online version contains supplementary material available at 10.1186/s40854-022-00347-4.

| S-EPMC9060850 | biostudies-literature

Assessing How Residual Errors of Scoring Functions Correlate to Ligand Structural Features.

Project description:Scoring functions (SFs) are ubiquitous tools for early stage drug discovery. However, their accuracy currently remains quite moderate. Despite a number of successful target-specific SFs appearing recently, up until now, no ideas on how to systematically improve the general scope of SFs have been formulated. In this work, we hypothesized that the specific features of ligands, corresponding to interactions well appreciated by medicinal chemists (e.g., hydrogen bonds, hydrophobic and aromatic interactions), might be responsible, in part, for the remaining SF errors. The latter provides direction to efforts aimed at the rational and systematic improvement of SF accuracy. In this proof-of-concept work, we took a CASF-2016 coreset of 285 ligands as a basis for comparison and calculated the values of scores for a representative panel of SFs (including AutoDock 4.2, AutoDock Vina, X-Score, NNScore2.0, ΔVina RF20, and DSX). The residual error of linear correlation of each SF value, with the experimental values of affinity and activity, was then analyzed in terms of its correlation with the presence of the fragments responsible for certain medicinal chemistry defined interactions. We showed that, despite the fact that SFs generally perform reasonably, there is room for improvement in terms of better parameterization of interactions involving certain fragments in ligands. Thus, this approach opens a potential way for the systematic improvement of SFs without their significant complication. However, the straightforward application of the proposed approach is limited by the scarcity of reliable available data for ligand-receptor complexes, which is a common problem in the field.

| S-EPMC9739603 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data