CANCER DETECTION

I have used various machine learning algorithm to predict the type of cancer(Malignant and Benign) in Patients. I have compared the results of neural network and Decision tree algorithm with the logistic regression algorithm.

Here, Mr. John Hughes suggested that a neural network is the best choice to detect the cancer.

Data source - Kaggle

SUMMARY OF KEY FINDINGS

1. Key Statistics
   
2. Implemented and evaluated different algorithms
   
3. compared the results of final NN algo to the Logistic regression algo with below given output
   

ASSUMPTIONS FOR NN(NEURAL NETWORK)

1. Used 20% for the test size, i.e test_size = 0.2
   
2. Hidden_layer_size = (9,4,2)
   
3. max_iter = 10000
   
4. random_state = 100
   

OUTCOMES

NN

DECISION TREE

LOGISTIC REGRESSION

CONFUSION MATRIX