An empirical study of the naive Bayes classifier

By T. J. Watson. 2001

3,009 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/2825733f97124013e8841b3f8a0f5bd4ee4af88a

How researchers see this paper

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

Citations in the corpus

Bayesian Naïve Bayes classifiers to text classification (2018, background): “Despite apparently over-simplifier assumptions, NB classifier works quite well in many complex real-world applications, such as text classification [6,7], keyphrase extraction [8] and medical diagnosis [9].”
A Review on Machine Learning for EEG Signal Processing in Bioengineering (2020, background): “It can be corrected with a pseudocount to smooth out any outliers in the data set [91].”
A Survey of Machine Learning Techniques Applied to Self-Organizing Cellular Networks (2017, background): “ested readers, a more in-depth review of Bayesian classifiers, its advantages and disadvantages, and its two models, can be found in [46] and [47].”
Early Alert of Academically At-Risk Students: An Open Source Analytics Initiative (2014, methodology): “For further reading on the performance of Naïve Bayes, including a theoretical explanation on the optimality of the algorithm, see Rish (2001) and Zhang (2004).”
An empirical study on software defect prediction with a simplified metric set (2014, methodology): “Although the independence assumption is often violated in the real-world, the Naı̈ve Bayes classifier often competes well with more sophisticated classifiers [64].”
Spam Detection on Twitter Using Traditional Classifiers (2011, background): “The big assumption of the naïve Bayesian classifier is that the features are conditionally independent, although research shows that it is surprisingly effective in practice without the unrealistic independence assumption [7].”
A review of feature selection methods on synthetic data (2013, unknown): “5 [66], naive Bayes (NB) [67], IB1 [68] and SVM [69].”
Don't follow me: Spam detection in Twitter (2010, background): “The big assumption of naı̈ve Bayesian classifier is that the features are conditionally independent, although research shows that it is “is surprisingly effective in practice” without the unrealistic independence assumption (Rish, 2005).”
Using naive bayes to detect spammy names in social networks (2013, background): “The Naive Bayes algorithm has demonstrated great success in classiﬁcation problems (see e.g., [11, 24, 21]) even though the value s ( (cid:126)x ) of (2.3) is often highly inaccurate as an estimate of probability [4, 31, 10].”
Actively Semi-Supervised Deep Rule-based Classifier Applied to Adverse Driving Scenarios (2019, background): “The Deep Rule-Based, K-nearest Neighbors (KNN) [28], Naive Bayes (NB) [29], Support Vector Machine (SVM) [30], and the deep convolutional neural network VGG–VD–16 classifiers are compared in terms of accuracy and processing time for labeled images.”
ACAS: automated construction of application signatures (2005, methodology): “Specifically, we select three popular learning algorithms – Naive Bayes [8], AdaBoost [10] and Maximum Entropy [5], and use them to develop the application signatures.”
Towards Measuring Membership Privacy (2017, methodology): “Naive Bayes classifiers make predictions using Bayes theorem and assume conditional independence [32].”
Deep Learning for Change Detection in Remote Sensing Images: Comprehensive Review and Meta-Analysis (2020, methodology): “the analysis of models that either require a greater composition of learned concepts or functions, compared to conventional machine learning models such as naive Bayes [2], [3], support vector machine (SVM) [4], [5], random forests, [6], [7] and the decision tree [8], [9].”
Good Friends, Bad News - Affect and Virality in Twitter (2011, methodology): “We use the so-called Naive Bayes classiﬁer based on univariate discrete distributions [24], i.e., we posit a model of the probability of a tweet being news of the form p (news | w ) = p (news) p ( w | news) p (news) p ( w | news) + p ( ¬ news) p ( w |¬ news) Q where p ( w d | news) is the…”