Idiot's Bayes—Not So Stupid After All?

By David J. Hand, Keming Yu. 2001

1,137 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/bd6a9d35dabbba8132f48835f636cf7c3b3e9c80

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

1. Philip S. Yu
2. Charu C. Aggarwal
3. B. Liu
4. Qiang Yang
5. Zhi-Hua Zhou
6. Bing Liu
7. Vipin Kumar
8. Xindong Wu

In Information Systems

1. Philip S. Yu
2. Vipin Kumar
3. Zhi-Hua Zhou
4. B. Liu
5. Kevin Swingler
6. Xindong Wu
7. Qiang Yang
8. Heikki Mannila

Citations in the corpus

• Transaction aggregation as a strategy for credit card fraud detection (2009, background): “• Naive Bayes Classifier (see e.g. Hand and Yu 2001; Duda and Hart 1973).”
• The Perils of Ignoring Data Suitability - The Suitability of Data used to Train Neural Networks Deserves More Attention (2011, methodology): “Wu et al. found that the most popular classification methods were k Nearest Neighbours (kNN), Naïve Bayes (Hand, 2001), Support Vector Machines (Vapnik, 1995), and two tree building algorithms: C4.5 (Quinlan, 1993) and CART (Breiman, 1984).”
• BANYAN. II. VERY LOW MASS AND SUBSTELLAR CANDIDATE MEMBERS TO NEARBY, YOUNG KINEMATIC GROUPS WITH PREVIOUSLY KNOWN SIGNS OF YOUTH (2013, methodology): “However, since we use a naive Bayesian classifier in the sense that we treat input parameters as independent variables, we expect that the Bayesian probabilities P(Hk|{Oi}) we derive this way will be biased (Hand & Yu 2001).”
• Web-Scale Bayesian Click-Through rate Prediction for Sponsored Search Advertising in Microsoft's Bing Search Engine (2010, methodology): “We are currently exploring the use of adPredictor for related tasks within Bing including organic search, display and contextual advertising.”
• A human functional protein interaction network and its application to cancer data analysis (2010, background): “NBCs can frequently yield better results than more sophisticated techniques [80].”
• Encrypted statistical machine learning: new privacy preserving methods (2015, background): “Essentially, although it produces biased probability estimates, this does not necessarily translate into a high classification error (Hand and Yu, 2001).”
• Prediction of Slope Stability using Naive Bayes Classifier (2018, background): “Despite the naive design and the apparently oversimplified assumptions, many examples are reported exhibiting the surprising effectiveness of the NBCs even with strong dependencies (Hand and Yu, 2001; Zhang, 2004).”
• BANYAN. VII. A NEW POPULATION OF YOUNG SUBSTELLAR CANDIDATE MEMBERS OF NEARBY MOVING GROUPS FROM THE BASS SURVEY (2015, background): “Probabilities generated from a naive Bayesian classifier can be biased when the input parameters are not independent (which is the case here); however, the relative ranking of hypotheses for a given object overcomes this bias (Hand & Yu 2001).”
• EffectorP: predicting fungal effector proteins from secretomes using machine learning. (2016, background): “Nevertheless, Na€ıve Bayes classification is known to be a robust approach that often outperforms more sophisticated machine learning classifiers (Hand & Yu, 2001) and surprisingly, correlation between variables is not necessarily detrimental to its performance (Wu et al., 2008).”
• Principles of Data Mining (2001, methodology): “Discussions of the independence Bayes method include Russek et al 1983, Hilden 1984, Kohavi 1996, Domingos and Pazzani 1997, and Hand and Yu 1999.”
• A naive Bayes model for robust remaining useful life prediction of lithium-ion battery (2014, background): “The naive Bayes has a long history in the literature, sometimes under various names in earlier articles [26].”
• Top 10 algorithms in data mining (2007, methodology): “General discussion of the naive Bayes method and its merits are given in [22,33].”