search

SrijanShovit / HealthLearning

A repo comprising of various Machine Learning and Deep Learning projects in healthcare domain.
39 stars 54 forks source link

Issue: #79 Heart attack prediction model #85

Closed jain-anshika closed 6 months ago

jain-anshika commented 6 months ago

Issue: #79

Heart attack prediction model

The test accuracy score of Logistric Regression is 0.9016393442622951 The test accuracy score of SVM after hyper-parameter tuning is 0.9016393442622951 The test accuracy score of Gradient Boosting Classifier is 0.8688524590163934 The test accuracy score of Random Forest is 0.7868852459016393 The test accuracy score of Decision Tree is 0.7868852459016393

jain-anshika commented 6 months ago

@SrijanShovit Hey I added more comments and conclusion please check if it is okay now!! Please tell me if there are any other changes needed.

Thank you for giving me a chance to work.

jain-anshika commented 6 months ago

Do you mean I should only keep it till conclusions of EDA?

SrijanShovit commented 6 months ago

Do you mean I should only keep it till conclusions of EDA?

yes, you might see for other projects also. I find it good for small steps as separate issues rather than entire project as one issue. This would help in learning and points both.

SrijanShovit commented 6 months ago

Do you mean I should only keep it till conclusions of EDA?

yes, you might see for other projects also. I find it good for small steps as separate issues rather than entire project as one issue. This would help in learning and points both.

SrijanShovit commented 6 months ago

Issue: #79

Heart attack prediction model

The test accuracy score of Logistric Regression is 0.9016393442622951 The test accuracy score of SVM after hyper-parameter tuning is 0.9016393442622951 The test accuracy score of Gradient Boosting Classifier is 0.8688524590163934 The test accuracy score of Random Forest is 0.7868852459016393 The test accuracy score of Decision Tree is 0.7868852459016393

So you see you don't have great accuracy. There can be some reasons. As we proceed with steps each with 1 issue, we can have better ideas which aspect to it to take accuracy till 98% or above.