INTERVIEW
summary
Claims a practical pipeline that uses a Bayesian network built for chikungunya detection in 48 hours, extracting and applying high-support high-confidence rules from the network’s causal structure to flag likely cases early. Uses Python tooling with scikit-learn and emphasizes need for solid statistics and ML foundations; positions Bayesian networks and decision trees as interpretable choices rather than black-box predictors; reports implementation and evaluation on disease-specific data.
AI-written summary · about
related talks
41mBayesian Network Modeling using R and Python
39mThe Duct Tape of Heroes: Bayes Rule
31mBayesian Risk Analysis For Large Multi-Modal Data
32mData science vs real world: friends or foes
44mAI in the Clinic: Prediction is Better Than Cure
47mSimple Bayesian Networks with Ruby
39mProbablistic Programming Data Science with PyMC3
14mCloud-Native AI for Fraud Prevention with Bayesian Networks