Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2306.01618

Analyzing Credit Risk Model Problems through NLP-Based Clustering and Machine Learning: Insights from Validation Reports

Szymon Lis·Mariusz Kubkowski·Olimpia Borkowska·Dobromił Serwa·Jarosław Kurpanik
View original source ↗Read source document ↗Export metadata ↓

Abstract

This paper explores the use of clustering methods and machine learning algorithms, including Natural Language Processing (NLP), to identify and classify problems identified in credit risk models through textual information contained in validation reports. Using a unique dataset of 657 findings raised by validation teams in a large international banking group between January 2019 and December 2022. The findings are classified into nine validation dimensions and assigned a severity level by validators using their expert knowledge. The authors use embedding generation for the findings' titles and observations using four different pre-trained models, including "module\_url" from TensorFlow Hub and three models from the SentenceTransformer library, namely "all-mpnet-base-v2", "all-MiniLM-L6-v2", and "paraphrase-mpnet-base-v2". The paper uses and compares various clustering methods in grouping findings with similar characteristics, enabling the identification of common problems within each validation dimension and severity. The results of the study show that clustering is an effective approach for identifying and classifying credit risk model problems with accuracy higher than 60\%. The authors also employ machine learning algorithms, including logistic regression and XGBoost, to predict the validation dimension and its severity, achieving an accuracy of 80\% for XGBoost algorithm. Furthermore, the study identifies the top 10 words that predict a validation dimension and severity. Overall, this paper makes a contribution by demonstrating the usefulness of clustering and machine learning for analyzing textual information in validation reports, and providing insights into the types of problems encountered in the development and validation of credit risk models.

Explore related subjects

cs.LG

Keep this discovery

BibTeXRIS

Szymon Lis, Mariusz Kubkowski, Olimpia Borkowska, Dobromił Serwa, Jarosław Kurpanik. 2023-06-02. Analyzing Credit Risk Model Problems through NLP-Based Clustering and Machine Learning: Insights from Validation Reports. https://arxiv.org/abs/2306.01618

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

AUC Maximization from Biased Positive-unlabeled Data with Confidence

Maximizing the area under the receiver operating characteristic curve (AUC) is a standard approach to imbalanced binary classification. Although positive and negative data are required for maximizing the AUC, negative data are often difficult to collect in some real-world applications due to privacy concerns or the need for specialized expertise to annotate them. Thus, AUC maximization from positive and unlabeled (PU) data has been attracting attention. Existing methods assume that labeled positive data are unbiased samples from the true positive distribution. However, this ideal assumption is often violated in practice. In this paper, we propose a method to maximize the AUC from biased PU data. To address the bias, our key idea is to exploit {\it confidence}, i.e., the probability that an instance is positive, associated with the small number of labeled positive data. We derive an estimator of the AUC risk using biased PU data with confidence, enabling AUC maximization under such bias. We further show that the rewritten AUC risk induces a Bayes-optimal AUC ranking even when the available confidence is any strictly increasing transformation of the true posterior probability. We experimentally show the effectiveness of our method on eight real-world datasets.

cs.LG↗
arXiv2026

Measuring the Value of World-Model Updates: A Counterfactual Utility Protocol for Continual Adaptation

Continual world models must decide whether new data justify changing the model. Fixed replay schedules and prediction-error triggers specify when to update, but neither reveals the value of an individual update: one deployment run cannot show how the same model would have performed at that moment had it held its parameters. We introduce the fork ledger, which branches a deployment stream at pre-registered decision points into matched update and hold continuations under common random numbers. It evaluates both continuations on the same episodes and records $\Delta R = R_{\mathrm{update}} - R_{\mathrm{hold}}$. Always applying one fixed update mechanism lowers return on all three simulated control tasks: CartPole ($-144.0$; checkpoint-bootstrap $95\%$ CI $[-185.4,-116.1]$, against a converged return near $650$), Walker ($-82.8$; $[-101.1,-61.7]$) and Cheetah ($-18.6$; $[-29.0,-6.6]$). Divergence is an outcome of applying the update, so the estimand counts every attempted fork; restricted to the $693$ of $720$ that did not collapse, CartPole and Walker are unchanged in sign ($-113.4$ and $-82.1$) and Cheetah becomes unresolved ($-3.9$; $[-17.5,+13.0]$). The task is the unit of inference: each contributes $240$ attempted forks over five pretrained checkpoints crossed with two drift directions. The ledger makes counterfactual utility observable for a fixed mechanism, allowing triggers to be judged by the updates they select rather than by surprise detection alone.

cs.LG↗
arXiv2026

When More Is Not Better: Component Anti-Synergy in a P300 Speller

P300 brain-computer interface (BCI) spellers can provide hands-free communication for people with severe motor impairments. Modern pipelines combine multiple individually promising components, often assuming that 'more-is-better'. We tested this assumption using a four-component full-factorial experiment varying the inclusion of Euclidean Alignment (EA), xDAWN spatial filtering, subject calibration, and language model priors on a public P300 dataset. Performance was evaluated using accuracy, repetitions, and information transfer rate (ITR) with mixed-effects models. Results show that the value of components is conditional rather than additive. Calibration was the strongest singular contributor, while EA compensated for its absence in zero-calibration settings. Adding independently useful components could also reduce performance, revealing component anti-synergy. Contrary to conventional wisdom, LM support was not universally beneficial: its effect depends strongly on the strength of the underlying EEG pipeline, while results from a larger LM showed a similar pattern. Together, these findings challenge maximal 'all-on' pipeline design and highlight the value of selecting spatial and language-support components according to the quality of available EEG evidence.

cs.LG↗