{"id":1384,"date":"2021-01-17T12:45:20","date_gmt":"2021-01-17T12:45:20","guid":{"rendered":"http:\/\/learnlearn.uk\/alevelcs\/?page_id=1384"},"modified":"2021-01-17T13:16:06","modified_gmt":"2021-01-17T13:16:06","slug":"classification-regression-clustering","status":"publish","type":"page","link":"https:\/\/learnlearn.uk\/alevelcs\/classification-regression-clustering\/","title":{"rendered":"Classification, Regression, Clustering & Reinforcement"},"content":{"rendered":"
\n

Classification<\/h2>\n
\n\n

Classification<\/h3>\n

The aim of the classification is to split the data into two or more predefined groups. A common example is spam email filtering where emails are split into either spam or not spam.<\/p>\n

\"\"<\/a><\/p>\n\n<\/div>

Linear Regression<\/h2>\n
\n\n

Regression<\/h3>\n

The aim of the regression is to predict the value of a dependent<\/strong> variable based upon another explanatory<\/strong> variable. Linear Regression is used where there is a straight line correlation between variables, for example between poverty and teen birth rates<\/a>.<\/p>\n

\"\"<\/a>

Here is an example of linear regression, as can be seen by the straight line.<\/p><\/div>\n\n<\/div>

Non Linear Regression<\/h2>\n
\n\n

Non Linear Regression<\/strong><\/p>\n

Non-linear regression is used where there is a correlation but it is not linear, for example between life expectancy and per capita income.<\/a><\/p>\n

\"\"<\/a>

Life expectancy vs income. [Click to enlarge]<\/p><\/div>\n\n<\/div>

Clustering<\/h2>\n
\n\n

Clustering<\/h3>\n

The objective of a clustering algorithm is to split the data into smaller groups or clusters based on certain features. The programmer might specify a target number of groups or let the algorithm decide.<\/p>\n

\"\"<\/a>

The training data splits the data into clusters<\/p><\/div>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n\n<\/div>

Reinforcement<\/h2>\n
\n\n

Reinforcement Learning<\/h3>\n

Reinforcement learning is a reward based system where an agent are not given specific instructions but are rewarded for how well they perform. Often this learning follows a Darwinian model where multiple agents attempt the task (each with different slightly randomly parameters) and those that perform the best form the base settings for their child (slightly mutated versions)<\/p>\n

With each generation the agents improve in their performance.<\/p>\n

\n
https:\/\/www.youtube.com\/watch?v=gn4nRCC9TwQ<\/a>