BigML Education - Supervised vs Unsupervised
1 like1,300 views
1. Supervised learning uses labeled training data that contains examples and outcomes to build a model that can predict the outcome of new data. Unsupervised learning uses only examples in the training data to find interesting patterns. 2. Supervised learning is used for problems like predicting home prices, loan defaults, or cancer diagnosis. Unsupervised learning is used for problems like clustering similar customers, detecting anomalies, or discovering associated products. 3. The key difference is that supervised learning predicts an outcome/label, while unsupervised learning performs discovery without labels in the training data.
BigML Education - Supervised vs Unsupervised
- 1. BigML Education Supervised vs Unsupervised July 2017
- 2. BigML Education Program 2Supervised versus Unsupervised Supervised Learning 1. Training data provides “examples” and “outcomes” 2. The machine learns to predict the outcome of new data based on the past examples SQFT BEDS BATHS 3.125 5 3 2.100 4 2 1.200 3 1,5 3.950 6 4 PRICE 530.000 $ 460.000 $ 250.000 $ ???
- 3. BigML Education Program 3Supervised versus Unsupervised Supervised Learning Label Training
- 4. BigML Education Program 4Supervised versus Unsupervised Supervised Learning Example Question • Will this customer default on a loan? Training Data Previous months of loans applications Previous home sales • How many customers will apply for a loan next month? • How much is this home worth? Previous loans that were paid or defaulted • Is this cancer malignant? Previous stats of benign / malignant cancers
- 5. BigML Education Program 5Supervised versus Unsupervised Supervised Learning • Training data has one feature that is the “outcome” • Sometimes referred to as the “label” or “objective” • Goal is to build a model which can predict the outcome • If categorical: model is a“classification” • if numeric: model is a “regression” • Because the data has a known value, model can be evaluated • Split the data into a training and test set • Model the training set / Predict the test test • Compares the predictions to the known values • Algorithms • Model / Ensemble • Logistic Regression • Time Series
- 6. BigML Education Program 6Supervised versus Unsupervised Unsupervised Learning 1. Training data provides “examples” - no specific “outcome” 2. The machine tries to find “interesting” patterns in the data date customer account auth class zip amount Mon Bob 3421 pin clothes 46140 135 Tue Bob 3421 sign food 46140 401 Tue Alice 2456 pin food 12222 234 Wed Sally 6788 pin gas 26339 94 Wed Bob 3421 pin tech 21350 2459 Wed Bob 3421 pin gas 46140 83 Thr Sally 6788 sign food 26339 51
- 7. BigML Education Program 7Supervised versus Unsupervised Unsupervised Learning date customer account auth class zip amount Mon Bob 3421 pin clothes 46140 135 Tue Bob 3421 sign food 46140 401 Tue Alice 2456 pin food 12222 234 Wed Sally 6788 pin gas 26339 94 Wed Bob 3421 pin tech 21350 2459 Wed Bob 3421 pin gas 46140 83 The Sally 6788 sign food 26339 51 Clustering similar
- 8. BigML Education Program 8Supervised versus Unsupervised Unsupervised Learning date customer account auth class zip amount Mon Bob 3421 pin clothes 46140 135 Tue Bob 3421 sign food 46140 401 Tue Alice 2456 pin food 12222 234 Wed Sally 6788 pin gas 26339 94 Wed Bob 3421 pin tech 21350 2459 Wed Bob 3421 pin gas 46140 83 The Sally 6788 sign food 26339 51 Anomaly Detection unusual
- 9. BigML Education Program 9Supervised versus Unsupervised Unsupervised Learning date customer account auth class zip amount Mon Bob 3421 pin clothes 46140 135 Tue Bob 3421 sign food 46140 401 Tue Alice 2456 pin food 12222 234 Wed Sally 6788 pin gas 26339 94 Wed Bob 3421 pin tech 21350 2459 Wed Bob 3421 pin gas 46140 83 Thr Sally 6788 sign food 26339 51 zip = 46140 amount < 100 {customer = Bob, account = 3421} {class = gas} Association Discovery
- 10. BigML Education Program 10Supervised versus Unsupervised Unsupervised Learning Example Question • Is this transaction unusual? Training Data Examples of previous purchases Customer profiles • Are the products purchased together? • Are these customers similar? Previous transactions
- 11. BigML Education Program 11Supervised versus Unsupervised Unsupervised Learning • Training data has only examples and no specific “outcome” • This is common - labels are typically expensive • Goal is to perform discovery, find patterns, etc • Tends to be more difficult • Algorithms • Clusters • Anomaly Detection • Association Discovery • Topic Models • Because the data has no “outcome”, can not be evaluated • Each discovery method has it’s own quality measures