Last active
March 31, 2016 14:32
-
-
Save ducnh1022/2bc3393535bf1b71557f2a2ba271f05e to your computer and use it in GitHub Desktop.
Classification Week 1 - Linear classifier and logistic regression
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Purpose | |
Data -> classifier -> intelligence | |
Input x -> sentence sentiment classifier -> y (positive or negative) | |
App: | |
Spam filtering | |
image classification | |
Impact of classification | |
Course overview | |
models: linear classifier, logistic regression, decision trees, ensembles | |
algorithms: gradient, stochastic gradient, recursive greedy, boosting | |
core ML concept: alleviating overfitting, missing data, precision recall, online learning | |
Linear Classifiers: use training data to learn a weight or coefficient for each word | |
Intuition | |
Scoring a sentence | |
output is weighted sum of input | |
Decision Boundary | |
Class probability | |
link function: suqeeze real line into 0..1 => generalize model | |
Logistic regression | |
sigmoid function => sigmoid score = 1/(1+exp(-score)) | |
find best classifier = maximize quality metric over all possible w0 w1 w2 | |
Encoding categorical inputs | |
Multiclass classification | |
1 versus all => +1 for 1 class and -1 for the rest | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment