Given the dataset below, where we want to classify whether an social networks account is real...

 

Transcribed Image Text:

Given the dataset below, where we want to classify whether an social networks account is real or not. Here "Posts", "Friends", "Photo" are the features/attributes, meaning the frequency of writing posts, number of friends, and whether use the real photo, respectively. "Real Account" is the label. Attribute values "S", "M", "L" means a "small", "medium" and "large number, respectively. Suppose we want to use Information Gain to build a decision tree model (ID3). Posts Friends Photo Real Account S S NO S YES L YES M YES L YES M YES M NO L YES M NO S YES 1) (5 pts) Compute the Information Gain if we first choose "Friends" as the attribute to split data. 2) (15 pts) Construct a decision tree from the given data. Show the computation steps. 3) (5 pts) Explain the limitation of using Information Gain as the attribute splitting measure. No. 1 2 C++ 3 4 5 6 7 8 9 10 L M M M L S M S S NO YES NO YES YES NO NO NO NO YES Activa Go to s Given the dataset below, where we want to classify whether an social networks account is real or not. Here "Posts", "Friends", "Photo" are the features/attributes, meaning the frequency of writing posts, number of friends, and whether use the real photo, respectively. "Real Account" is the label. Attribute values "S", "M", "L" means a "small", "medium" and "large number, respectively. Suppose we want to use Information Gain to build a decision tree model (ID3). Posts Friends Photo Real Account S S NO S YES L YES M YES L YES M YES M NO L YES M NO S YES 1) (5 pts) Compute the Information Gain if we first choose "Friends" as the attribute to split data. 2) (15 pts) Construct a decision tree from the given data. Show the computation steps. 3) (5 pts) Explain the limitation of using Information Gain as the attribute splitting measure. No. 1 2 C++ 3 4 5 6 7 8 9 10 L M M M L S M S S NO YES NO YES YES NO NO NO NO YES Activa Go to s