Algokghvjvgjviviviugviugvigvgvrithms.pptx

bbjhbijhfgchgcc

1. Classification ML Algorithms

2. Breast CA is the most common cancer among female and one of the leading cause of cancer death. Objective - To develop a predictive model using Multilayer Perceptron (MLP) algorithm to classify breast cancer tumors as malignant or benign based on various features extracted from diagnostic images. Approach - Data Analysis, Visualization, Data Preprocessing, Model Training, Model Evaluation, Storing the Model The Problem ‹#›

3. ‹#› The Data Consists of 31 Attributes and 569 entries Tumor size, Shape, and Texture To identify each sample. To be predicted Benign and Malignant. Area, Perimeter, Texture , Compactness, Concavity, Symmetry - Actual values, Mean and Standard Error Aspects ID & Diagnosis Attributes

4. Multilayer Perceptron ‹#› A type of Artificial Neural Network that designed to behave like human brain. It is consisted of multiple layers of nodes, each connected to the next layer. They use backpropagation to update the weights of connections between nodes during training in order to minimize the error. Predict categorical outcome of discrete values like Cat & Dog, Pneumonia or Healthy. Predicting continuous outcomes of continuous variables like Air Temperature , Stock Price Classification Regression

5. The Process Creating MLP involves several key steps in planning, data analysis , preprocessing, training and testing model and finally store the model. ‹#› _ Planning _ Plan What type of algorithm is suitable for the data we have. _ Data Analysis _ General overview of the data, obtain variables and find missing values. Preprocessing _ Adjust data types, find and remove outliers _ Training _ Train the model. _ Testing _ Perform test

6. Training Result of Python Model

7. Testing Result of Python Model

8. Result of Weka Model

9. The KNN Algorithm Principles of similarity Euclidian distance The K value How to determine? Rule of thumb Overview The nearest Neighbour? Once trained – Memorize. Lazy learner

10. Steps

11. Training Result of Python Model

12. Testing Result of Python Model

13. Result of Weka Model

14. Gaussian Naive Bayes ‹#› C lassification algorithm based on Bayes' theorem with the assumption of independence among predictors. It’s mainly useful to predict continuous data. Differs from Naive Bayes by Used when one of the attributes presence or absence affects the class or dependent variable. Used when the presence or absence one attribute can not affect the class or dependent variable. GNB Naive Bayes

15. The Process Similar to other planning, data analysis, preprocessing, training and testing model and finally store the model. _ Planning _ Plan What type of algorithm is suitable for the data we have. _ Data Analysis _ General overview of the data, obtain variables and find missing values. Preprocessing _ Adjust data types, find and remove outliers _ Training _ Train the model. ‹#› _ Testing _ Perform test

16. Training Result of Python Model

17. Testing Result of Python Model

18. Result of Weka Model

19. Dr Million (PhD) Moderator