Using Large Data Sets

瘋狂的一週開始! 很多要補完! 速度上工拉~!

Data For Machine Learning

Dsigning a high accuracy learning system

研究這主題的大神 : Banko and Brill 2011

E.G. Classify between confusable words.
   {to, two, too}, {then, than}
For breakfast I ate _____ eggs.

Algorithms

• Perceptron (Logistic regression)
• Winnow
• Memory-based

• Naive Bayes

• 

  “It’s not who has the best algorithm that wins. It’s who has the most data.”
  So, When is this true and when is this not true?

Large data rationale

Useful test: Given the input x, can a human expert confidently predict y ?

Use a learning algorithm with many parameters (e.g. logistic regression/linear regression with many features; neural network with many hidden units)

Use a very large training set(unlikely to overfit)

The large training set is unlikely to help when:

• The features x do not contain enough information to predict y accurately (such as predicting a house’s price from only its size), and we are using a simple learning algorithm such as logistic regression

• The features x do not contain enough information to predict y accurately (such as predicting a house’s price from only its size), even if we are using a neural network with a large number of hidden units.

For reference:

• Accuracy = (true positives + true negatives) / (total examples)
• Precision = (true positives) / (true positives + false positives)
• Recall = (true positives) / (true positives + false negatives)
• F1 score = (2 * precision * recall) / (precision + recall)