Estimating Continuous Distributions in Bayesian Classifiers

Abstract

When modeling a probability distribution with a Bayesian network, we are faced with the problem of how to handle continuous variables. Most previous work has either solved the problem by discretizing, or assumed that the data are generated by a single Gaussian. In this paper we abandon the normality assumption and instead use statistical methods for nonparametric density estimation. For a naive Bayesian classifier, we present experimental results on a variety of natural and artificial domains, comparing two methods of density estimation: assuming normality and modeling each conditional distribution with a single Gaussian; and using nonparametric kernel density estimation. We observe large reductions in error on several natural and artificial data sets, which suggests that kernel estimation is a useful tool for learning Bayesian models.

Appears in Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence (UAI1995)

Cited by in corpus (71)

• Fast Parallel Hypertree Decompositions in Logarithmic Recursion Depth
• EEG machine learning with Higuchi fractal dimension and Sample Entropy as features for successful detection of depression
• A Feature Subset Selection Algorithm Automatic Recommendation Method
• Structure-semantics interplay in complex networks and its effects on the predictability of similarity in texts
• Parsing Argumentation Structures in Persuasive Essays
• Feature Extraction and Soft Computing Methods for Aerospace Structure Defect Classification
• Bayes and Naive Bayes Classifier
• Different Approaches to Community Evolution Prediction in Blogosphere
• Classification of signaling proteins based on molecular star graph descriptors using Machine Learning models
• Bayesian sparse graphical models for classification with application to protein expression data
• Limiting the effects of earthquakes on gravitational-wave interferometers
• A Comparative Study of Machine Learning Methods for Verbal Autopsy Text Classification
• On the efficacy of old features for the detection of new bots
• An Improved Image Mining Technique For Brain Tumour Classification Using Efficient Classifier
• A Survey of COVID-19 Misinformation: Datasets, Detection Techniques and Open Issues
• Crowdsmelling: The use of collective knowledge in code smells detection
• Analysis of group evolution prediction in complex networks
• Synergizing Domain Expertise with Self-Awareness in Software Systems: A Patternized Architecture Guideline
• An Empirical Evaluation of GDPR Compliance Violations in Android mHealth Apps
• The Natural Selection of Words: Finding the Features of Fitness
• Negative Effects of Incentivised Viral Campaigns for Activity in Social Networks
• An Improved Naive Bayes Classifier-based Noise Detection Technique for Classifying User Phone Call Behavior
• Lifted Relational Variational Inference
• A Method for Group Extraction and Analysis in Multilayer Social Networks
• Feature Selection for MAUC-Oriented Classification Systems
• ARC: DVFS-Aware Asymmetric-Retention STT-RAM Caches for Energy-Efficient Multicore Processors
• Exponential Series Approaches for Nonparametric Graphical Models
• Application of Data Mining to Network Intrusion Detection: Classifier Selection Model
• Semi-supervised Classification for Natural Language Processing
• Classifying textual data: shallow, deep and ensemble methods
• Evaluating Classifiers in Detecting 419 Scams in Bilingual Cybercriminal Communities
• Computational Eco-Systems for Handwritten Digits Recognition
• Infant Mortality Prediction using Birth Certificate Data
• A Naive Bayes machine learning approach to risk prediction using censored, time-to-event data
• Multi-label classification search space in the MEKA software
• Coronary Heart Disease Diagnosis Based on Improved Ensemble Learning
• Classifier ensemble creation via false labelling
• Data Mining: A prediction Technique for the workers in the PR Department of Orissa (Block and Panchayat)
• Analyzing Social and Stylometric Features to Identify Spear phishing Emails
• Detecting Vietnamese Opinion Spam
• Performance Analysis of Machine Learning Techniques to Predict Diabetes Mellitus
• Needmining: Identifying micro blog data containing customer needs
• Proper likelihood ratio based ROC curves for general binary classification problems
• On-the-fly Detection of Autogenerated Tweets
• Misinformation detection in Luganda-English code-mixed social media text
• Machine Assistance for Credit Card Approval? Random Wheel can Recommend and Explain
• ITCM: A Real Time Internet Traffic Classifier Monitor
• Improving Change Prediction Models with Code Smell-Related Information
• Finding the Mode of a Kernel Density Estimate
• Linguistic Characteristics of Censorable Language on SinaWeibo
• Needmining: Designing Digital Support to Elicit Needs from Social Media
• A One-Pass Private Sketch for Most Machine Learning Tasks
• Detecting Spammers via Aggregated Historical Data Set
• Understanding and Monitoring Human Trafficking via Social Sensors: A Sociological Approach
• An empirical investigation of different classifiers, encoding and ensemble schemes for next event prediction using business process event logs
• Predicting Group Evolution in the Social Network
• Local Probabilistic Model for Bayesian Classification: a Generalized Local Classification Model
• Fast Multi-Class Probabilistic Classifier by Sparse Non-parametric Density Estimation
• Supervised Learning for Optimal Power Flow as a Real-Time Proxy
• A Feature-based Classification Technique for Answering Multi-choice World History Questions
• Fake News and Phishing Detection Using a Machine Learning Trained Expert System
• Robust Robot-assisted Tele-grasping Through Intent-Uncertainty-Aware Planning
• Empirical Study on Airline Delay Analysis and Prediction
• IMMIGRATE: A Margin-based Feature Selection Method with Interaction Terms
• On the Need of Removing Last Releases of Data When Using or Validating Defect Prediction Models
• Differentiated context-aware hook placement for different owners' smartphones
• Quantile-based classifiers
• CyberLearning: Effectiveness Analysis of Machine Learning Security Modeling to Detect Cyber-Anomalies and Multi-Attacks
• DI2: prior-free and multi-item discretization ofbiomedical data and its applications
• Predicting Failures in Multi-Tier Distributed Systems
• AppsPred: Predicting Context-Aware Smartphone Apps using Random Forest Learning