Rule extraction from support vector machines: A review

By Nahla H. Barakat, Andrew P. Bradley. 2010

219 citations recorded by the source corpus.

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Interval Coded Scoring: a toolbox for interpretable scoring systems (2018, background): “Alternatively, one can construct them based on SVMs (Martens et al., 2007; Barakat and Bradley, 2010).”
Risk group detection and survival function estimation for interval coded survival methods (2013, methodology): “In order to introduce the use of more complex mathematical methods [1, 2, 4, 8, 10–12] in these domains, different methods to obtain a score system have been proposed: optimal cut-points methods [13, 14], post-processing of previously built regression models [15], classification and regression trees”
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…”
Hybrid support vector machine rule extraction method for discovering the preferences of stock market investors: Evidence from Montenegro (2015, methodology): “This disadvantage of SVMs can be resolved using rule extraction techniques (Barakat & Bradley, 2010; Diederich, 2008), which allow rules to be generated that mimic the black-box SVM model as far as possible.”
A Mathematical Model for Interpretable Clinical Decision Support with Applications in Gynecology (2012, methodology): “Although rule extraction methods have been proposed [20], the transformation of advanced mathematical models into a CDS system remains difficultly acceptable for clinicians.”
Comparative analysis of data mining methods for bankruptcy prediction (2012, background): “However, neural network models are black boxes [4,51], lacking transparency (seeing what the model is doing, or comprehensibility) and transportability (being able to easily deploy the model into a decision support system for new cases).”
Towards Extracting Faithful and Descriptive Representations of Latent Variable Models (2015, background): “More recent approaches have tackled the same problem for Support Vector Machines (Barakat and Bradley 2010).”