Based on the predictions from the RF model, the insights for the Election are • The vote of a respondent is based on the importance of his attitude towards these features with top 5 being the most important. ✓ Hague ✓ Europe ✓ Blair ✓ age ✓ political.knowledge ✓ economic.cond.national ✓ economic.cond.household ✓ gender • The Attitude of the people towards, Hague, Blair and EU influence the voters most. • Hague needs to increase his popularity among the voters to win the election. • Blair already has huge popularity. • Blair has huge support among the voters and he has twice the vote as Hague. • Hague supporters are split into two categories 1. Supporters who have good opinion and vote for him 2. Supporters who have good opinion and didn’t vote for him. • The second case has to be analysed further by the party to find the reason for those behaviours. • Based on the predictions, Blair has the upper hand in the election. It is predicted that the Labour party might win with huge margins. • Based on the Business Problem, whether to predict which party votes accurately, the model can be used for that problem statement with the respective Precision/Recall Measures. • As a Final Model, RF and LDAB will the best ones to make predictions. With AUC also into consideration, LDAB can be used for better predictions.
-
Notifications
You must be signed in to change notification settings - Fork 0
Karthi18raj/Election-Exit-Poll-Prediction
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Classification Model to predict the Election Poll results for winning party
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published