# Hypothesis Testing and Classification

Everyone
• 17 videos | 1h 50m 32s
• Includes Assessment
This course will cover the basics of anomaly detection and classification: for these tasks there are methods coming from either statistics or machine learning that are built on different principles. As well as the fundamentals of hypothesis testing, which is the formalization of scientific inquiry. This delicate statistical setup obeys a certain set of rules that will be explained and put in context with classification.

## WHAT YOU WILL LEARN

• Understand what classification is
Know what binary classification is
Know how to evaluate a classifier and understand the metrics used to do so
Know what hypothesis testing is
Understand what a confidence interval is
Understand hypothesis testing when the distribution is binomial
Know how to use the statistics obtained in an appropriate manner
• Be able to apply the p-value to your calculation
Know how to estimate likelihood
Know how to use non-statistical classifiers
Understand the perceptron algorithm
Understand the proof behind the perceptron algorithm
Be able to use svms for more complicated problems
Be able to use the methods for estimating the parameters of logistic regression
Know how statistical techniques could be misapplied

## IN THIS COURSE

• Meet the instructors and learn what classification is.
• Learn about a common classification problem.
• 3.  What Are Anomalies? What Is Fraud? Spams? Part 2
Gain more understanding of binary classification.
• 4.  What Are Anomalies? What is Fraud? Spams? Part 3
Learn about errors in classification analysis.
• 5.  False Positive/Negative, Precision/Recall, F-Score
Learn how to evaluate a classifier using the metrics presented.
• 6.  Hypothesis Testing
Learn what a hypothesis test is.
• 7.  Confidence Intervals
Learn about confidence intervals and how they relate to hypothesis testing.
• 8.  Validity Of Binomial Distribution
Learn about how hypothesis testing works with a binomial distribution.
• 9.  Misuses Of Statistics
Learn how not to misuse your statistical output.
• 10.  P-Value
Learn about the p-value and how it works.
• 11.  Methods Of Estimating Likelihood
• 12.  Support Vector Machine: Non-Statistical Classifier
Find out how non-statistical classifiers work.
• 13.  Perceptron
Learn about a simple classifier with an elegant interpretation.
• 14.  Perceptron Proof
Learn about the proof behind our claim in the previous video.
• 15.  Perceptron & Data That Is Not Linearly Separable
Apply some simple modifications to allow SVMs to be used in more complicated settings.
• 16.  Estimating The Parameters Of Logistic Regression
Find out how you can estimate the parameters of logistical regression.
• 17.  Misapplications Of Statistical Techniques
Learn what not to do with these statistical methods.