Using the heart.csv data provided, construct an initial ‘default’ decision tree for classifying whether a patient has less chance, or more chance of having a heart attack, and summarise the performance of the model.
Improve the performance by changing the decision tree parameters. Suggest reasons for why the performance has improved, and explain the results.
Document your processes, showing step by step screens and outputs, including an explanation of the final parameters you used, and what effect they had on your model.
Description of the Heart dataset: • Age : Age of the patient • Sex : Sex of the patient • exng: exercise induced angina (1 = yes; 0 = no) • caa: number of major vessels coloured by flourosopy (0-3) • cp : Chest Pain type chest pain type o Value 1: typical angina o Value 2: atypical angina o Value 3: non-anginal pain o Value 4: asymptomatic • trtbps : resting blood pressure (in mm Hg) • chol : cholestoral in mg/dl fetched via BMI sensor • fbs : (fasting blood sugar > 120 mg/dl) (1 = true; 0 = false) • restecg : resting electrocardiographic results o Value 0: normal o Value 1: having ST-T wave abnormality o Value 2: showing probable or definite left ventricular hypertrophy • thalachh : maximum heart rate achieved • oldpeak: ST depression induced by exercise relative to rest (-2.6 to 6.2) • slp: the slope of the peak execise ST ssegment (0=upsloping, 1=flat, 2=Down) • thall: 0=normal; 1=fixed defect; 2=reversable defect • output (ie the label) : 0= less chance of heart attack 1= more chance of heart attack
2) k-Nearest Neighbour Classification
Using the DataSet_Scoring.csv and DataSet_Training.csv provided on BB.
First use DataSet_Training.csv to build and evaluate the k-NN algorithm
The dataset relates to pupils in a sports academy. Trainer has worked with athletes over the years and developed an extensive data set. He is now wondering can he use past performance of previous clients to predict prime sports for up-and-coming high school athletes. By evaluating each athlete’s performance across a range of tests, he hopes to be able to figure out for which sport each athlete has the highest aptitude.
Dataset consists of:
Age (in years) Strength (ranking weight lifting exercises on a scale of 1-10, 10 being the highest Quickness (performance in speed ‘buzzer’ tests, on a scale of 0-6, 6 being the fastest Injury 0/1 1= serious injuries that took more than 3 weeks to heal Vision: scores of 0-4 with 4 being perfect vision Endurance: scale of 0-10, with 10 being best Agility: scale of 0-100 with 100 being highest Decision¬_Making: 3-100 (note the data has some erroneous data that needs to be filtered out using the ‘filter examples’ operator) Prime_sport: which sport the athletes went on to specialist in (Football, Basketball, Baseball or hockey)
1) First use DataSet_Training.csv to build and evaluate the K-nn algorthm (note that data needs to be filtered to remove errors, as explained above!)
2) What is the accuracy of the model?
3) Next remove the split validation operator, bring the k-means operator to the main pane, read in the second file (DateSet_Scoring.csv), filter it for errors and connect it to the ‘apply model’ operator to generate k-NN model predictions and confidences
4) What is the confidence level for each prediction?
5) What happens when you change k to 2? Or to 3?
i need this by tomorrow after noon befor 3pm London Time
Author Social Media Boost Category: Content Creation, Content Writing, Facebook Marketing, Graphic Design, Instagram Marketing, Social Media Management, Social Media Marketing, Twitter Budget: $30 - $250 AUD
