Nomograms for visualizing support vector machines

By Aleks Jakulin, Martin Mozina, Jaka Demsar, Ivan Bratko, Blaz Zupan. 2005

92 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Evaluation & Benchmarks, Signal Processing

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

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

In Evaluation & Benchmarks

In Signal Processing

Citations in the corpus

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, neutral): “Visual explanation Internal visualization: [68, 77, 245]”
Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation (2013, methodology): “Jakulin et al. (2005) improves the interpretability of support vector machines by using a device called “nomograms” which provide graphical representation of the contribution of variables to the model ﬁt.”
Explaining Decision-Making Algorithms through UI: Strategies to Help Non-Expert Stakeholders (2019, methodology): “Examples of the white-box technique include showing probabilities of the nodes for Bayesian networks [4], projection techniques [9] and Nomograms [28] to see the “cut” in the data points for Support Vector Machines, and the visualization of the graph of a neural network [50].”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “[302, 303, 304] introduced a method for automatically generating nomograms as the graphical tool for visual explanations of the inferential mechanisms of SVM and naı̈ve bayesian-driven models.”
Explaining prediction models and individual predictions with feature contributions (2014, methodology): “This fact has been exploited to provide an explanation for the naive Bayes classifier [3,17,23], linear SVM [14], logistic regression [21], Cox regression models [15], and additive models in general [26].”
Explanation and Justification in Machine Learning : A Survey Or (2017, methodology): “[Martens et al. , 2008] proposed to interpret the predictions of an SVM classiﬁer by extracting conjunctive rules using a small subset of features.”
A General Method for Visualizing and Explaining Black-Box Regression Models (2011, methodology): “For others, model-specific explanation and visualization methods have been developed [3,5,6].”
Learning Accurate and Interpretable Classifiers Using Optimal Multi-Criteria Rules (2013, methodology): “Although nomograms [Mozina et al. 2004; Jakulin et al. 2005] can be useful tools for enabling the visualization of these models, they are still problematic for highdimensional problems.”
An Efficient Explanation of Individual Classifications using Game Theory (2010, methodology): “Other related work focusses on explaining the SVM model, most recently in the form of visualization (Poulet, 2004; Hamel, 2006) and rule-extraction (Martens et al., 2007).”
Faithful and Customizable Explanations of Black Box Models (2019, methodology): “Furthermore, there exist tools and frameworks to visualize the functionality of different classes of models such as decision trees (Teoh and Ma 2003), SVMs etc. (Jakulin et al. 2005).”
Explanation and Reliability of Individual Predictions (2013, methodology): “Some models, such as additive models or small decision trees, are inherently transparent and require little or no additional post processing [14, 32].”
Scaling regression inputs by dividing by two standard deviations (2008, methodology): “…best be summarized graphically, either compactly through summary methods such as graphs of coefficient estimates or nomograms (Lubsen, Pool, and van der Does, 1978, Harrell, 2001, Jakulin et al., 2005), showing the (perhaps nonlinear) relation between the expected outcome as each input is varied.”
Explaining machine learning models in sales predictions (2017, methodology): “In (Jakulin et al., 2005), nomograms were developed for SVM, but they work only for a restricted class of kernels and cannot be used for general non-linear kernels.”
Nonlinear Support Vector Machine Visualization for Risk Factor Analysis Using Nomograms and Localized Radial Basis Function Kernels (2008, methodology): “introduced a nomogram approach for visualizing an SVM that can graphically expose its internal structure, and draw the effect of each feature by means of an OR such as logistic regression [16].”
Interacting with Predictions: Visual Inspection of Black-box Machine Learning Models (2016, methodology): “[17]) to see the “cut” in the data points were utilized.”