Requirements of Assign1

Requirements

Instructions

Instructions

Marking Guide

The submission file

Part 1

Requirements

a) Describe a big data application that the company is using and use it to illustrate the characteristics of big data analytics. (4)
b) Suggest and describe one new big data application that would help the company improve their business performance. (4)
c) Explain why your suggested application is innovative and useful. Discuss the challenges of implementing the application that you proposed. (4)

Part2

Numerical Variables

Part A: Model (3 marks)

Requirement

1. Remove the ID attribute. Consider all attributes except the class attribute as numeric attributes.

Requirement

2. Build a Naïve Bayes model and classify benign and malignant. Show the screenshot of the model when you train the model with all records.

Part B: Explanation (3 marks)

Requirement

Use one record to explain how the model makes classification.

Part C: 10 fold cross validation (3 marks)

Requirement

Show the accuracy of the model using 10-fold cross validation and the confusion matrix. Show and explain the meaning of the precision and recall for malignant.

Categorical Variables

Part D: Discretizetion (3 marks)

Requirement

Discretise the data set using three bins (equal-frequency).

Part E: Model (3 Marks)

1. Model and Evaluation

Requirement

Build a Naïve Bayes model and classify benign and malignant using the discretised dataset. Show the accuracy of the model using 5-fold cross validation and the confusion matrix.

Part F: Explaination (3 Marks)

Requirement

Use one record to explain how the model makes classification using the discretised dataset.

Details of Assignment1