Prof. Paul was out of the city for some weeks, then it's time to update our (Fanwang and I) progress made so far in the project and present it in a future meeting.

The agenda looks as following:

  • Finished tasks
    • Feature selection
    • Imbalanced subsets
  • Current activities
    • Issues with common algorithms
    • Data collection on literature
  • Issues open for discussion
    • Evaluation metrics

Partial results common algorithsm

Logistic Regression

{'C': 0.001, 'penalty': 'none', 'solver': 'newton-cg'}
{'C': 1, 'penalty': 'l2', 'solver': 'newton-cg'}
{'C': 1, 'penalty': 'l1', 'solver': 'liblinear'}
{'C': 10, 'penalty': 'l2', 'solver': 'newton-cg'}
{'C': 0.001, 'penalty': 'none', 'solver': 'newton-cg'}
{'C': 1, 'penalty': 'l1', 'solver': 'liblinear'}
{'C': 10000, 'penalty': 'l2', 'solver': 'newton-cg'}
{'C': 10, 'penalty': 'l2', 'solver': 'newton-cg'}
{'C': 1, 'penalty': 'l2', 'solver': 'lbfgs'}
{'C': 0.1, 'penalty': 'l2', 'solver': 'lbfgs'}

KNN

{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'manhattan', 'n_neighbors': 5, 'weights': 'distance'}
{'metric': 'euclidean', 'n_neighbors': 5, 'weights': 'distance'}