You want to import a professor of information to calculate whether he speaks good lesson?
Suppose I import: professor of age, gender, job title etc.?
Question:
1, with the spark of the decision tree model and naive bayesian classification model to calculate the
LablePoint [1, 40 (age) 1 (gender: male) 2 (title: professor)]
40 (age) [0, 0 (gender: female) 1 (title: associate professor)]
40 (age) [1, 1 (gender: male) 3 (title: lecturer)]
For this part of the category features should I like above? Or below to handle such
LablePoint [1, 40 (age) 1 (gender: male) 0 (gender: female) 0 (title: associate professor) 1 (title: professor) 0 (title: lecturer)]
40 (age) [0, 0) (gender: male (gender: female) 1 (title: associate professor) 0 (title: professor) 0 (title: lecturer)]
40 (age) [1, 1 (gender: male) 0 (gender: female) 0 (title: associate professor) 0 (title: professor) 1 (title: lecturer)]
How to deal with the category features and text (I think is converted to vector calculation at present, how to convert vector)?
2, how to determine the characteristics and results of a relationship (has the characteristics of the structure of the more precise or have this feature and not the special clinic calculated results did not change)?
3, how to determine the relationship between the two characteristics?