Using Neural Network Rule Extraction and Decision Tables for Credit - Risk Evaluation

By Bart Baesens, Rudy Setiono, C. Mues, Jan Vanthienen. 2003

556 citations recorded by the source corpus.

Subfields: Accounting, Artificial Intelligence, Finance

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Accounting

In Artificial Intelligence

In Finance

Citations in the corpus

Interpretable Two-level Boolean Rule Learning for Classification (2015, background): “ et al., 2009; Letham et al., 2012; Vellido et al., 2012). Interpretable models can be learned directly from data or can result from the approximation of black-box models (Craven & Shavlik, 1996; Baesens et al., 2003). Boolean rules are considered to be one of the most interpretable classiﬁcatio”
Strongly agree or strongly disagree?: Rating features in Support Vector Machines (2016, methodology): “scoring (Baesens et al. 2003), fraud detection (Cecchini et al.”
Comprehensible credit scoring models using rule extraction from support vector machines (2007, neutral): “[1, 3]), very little literature is available on SVM rule extractions.”
Optimal randomized classification trees (2021, background): “Extracting knowledge from data is a crucial task in Statistics and Machine Learning, which has applications in areas such as Biomedicine [1, 2], Criminal Justice [3, 4, 5], Fraud Detection [6], Privacy Protection [7], Health Care [8, 9, 10, 11], Risk Management [12, 13], Social Networks [14], and Co”
When Is the Right Time to Refresh Knowledge Discovered from Data? (2013, background): “For example, knowledge discovered by association rule mining is either a finite set of large itemsets or a finite set of association rules (Agrawal and Srikant 1994); whereas knowledge discovered using major classification models consists of a finite set of classification rules (Quinlan 1993) or can”
Building Acceptable Classification Models (2010, methodology): “Extraction techniques have been proposed from ANNs [5], as well as from SVMs [29, 30].”
Clustering Categories in Support Vector Machines (2017, methodology): “This is a credit scoring dataset, with good customers de ning the positive class (y = +1) and bad customers de ning the negative class (y = −1), and has been used in the context of Supervised Classi cation, such as in [2].”
Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations (2017, background): “In some domains where human trust in the model is an important aspect (e.g., healthcare, justice, security), verifying the model exclusively through the lens of statistical accuracy is often not sufficient [4, 6, 8, 15, 16, 24].”
Explanation Mining: Post Hoc Interpretability of Latent Factor Models for Recommendation Systems (2018, background): “(e.g. classification [7] vs. ranking [18]), model families (e.g. support vector machines [10] vs. neural networks [8]), applications (e.g. knowledge base extraction [33] vs. recommendation systems[1]), and motivations (e.g. model debugging [22, 25] vs. user acceptance [11, 19]) as well as…”
Deep learning for limit order books (2016, unknown): “Khandani et al. (2010), Butaru et al. (2015), Brown and Mues (2012), Baesens et al. (2003), Loterman et al. (2012), and Fitzpatrick and Mues (2016) use machine learning approaches for modeling consumer loan risk.”
Sparsity in Optimal Randomized Classification Trees (2020, background): “Since they are rule-based, when small-sized, they are deemed to be leaders in terms of interpretability [1, 2, 9, 15, 17, 23, 27, 28, 32, 36].”
Domain knowledge integration in data mining using decision tables: case studies in churn prediction (2009, methodology): “The motivation is that both techniques yield simple, easy to understand models and furthermore have been shown to perform well in various problem settings, e.g. for credit scoring (Baesens et al , 2003 b ).”
Classification With Ant Colony Optimization (2007, background): “the extracted rule lists with decision tables [6] and decision diagrams [42], allowing for truly easy and user-friendly consultation in every day practices.”