search

linziyingzz / Peace-agreement

G12's repositories for mining possible pattern between peace agreements data and conflicts data
MIT License
1 stars 0 forks source link

Problem on country analysis part #10

Closed linziyingzz closed 3 years ago

linziyingzz commented 3 years ago

Hi, I might have some suggestion, there is no intuitive comparability of data put together in different dimensions and orders of magnitude. could you improve that?

s2070762 commented 3 years ago

I see the problem. Thank you! The data in the feature vector is very scattered and the small data are swallowed by the big data. I would use standardization to deal with this situation. Z-score standardization may be best suitable in this case. The data will be converted to the center by removing the mean of each feature and then be scaled via dividing the non-normal feature by its standard deviation. By doing so, there could exist more comparability of the data in different dimensions and orders of magnitude.