Searching for SEM308DS Machine Learning Assignment Task? We have a team of well-qualified writers who provide expert assignment help to college and university students. We cover a wide range of Assignments, such as Digital Marketing, Programming, and Computer Engineering, at an affordable price, with AI-free content and 24/7 live chat support.
ASSESSMENT TASKS
Task 1 FORMATIVE TASK Machine Learning Processes
FORMATIVE TASK
Instruction: Your department has taken on a new trainee, who needs to be briefed about machine learning. Produce a report that explains the process of machine learning. It must contain the following things:
- An outline of the components of the machine learning process, and the data required for different machine learning models
- Identify and explain how to convert categorical data to numerical values
- Discuss the difficulties involved in ‘class imbalance’ in machine learning models
Task 2 SUMMATIVE TASK Evaluating Machine Learning
SUMMATIVE TASK
Instruction: Carry out an evaluation of at least two machine learning models. Your analysis must contain the following:
- An outline of what is understood in each case by a ‘confusion matrix’, how classification metrics can be calculated from a confusion matrix, and the meaning of the terms ‘overfitting’ and ‘underfitting’ (LO 1, 3.2. 4.1, 5.1)
- Identify and explain ROC and AUC in both cases, and draw conclusions from them (LO 3.3, 4.2)
- A judgment as to difficulties in assessing unsupervised machine learning models, and how such effectively problems can be overcome (LO 3.4. 2)
| Learning Outcomes: To achieve this unit, the learner must be able to: | Assessment Criteria: Assessment of these learning outcomes will require a learner to demonstrate that they can: |
| 3. Understand how to evaluate machine learning models. | 3.1 Explain what is meant by “a confusion matrix”. 3.2 Define the classification metrics: “Precision”, “Accuracy”, “Recall”, “Support”, and “F1”. 3.3 Explain what is meant by a “Receiver Operating Characteristic Curve (ROC)”, and the “Area under the ROC curve” (AUC). 3.4 Explain the difficulties with assessing unsupervised machine learning models. |
| 4. Be able to evaluate classification models. | 4.1 Calculate the classification metrics correctly from a confusion matrix. 4.2 Interpret a ROC curve and AUC and make reasoned conclusions. |
| 5. Understand the issues of bias and variance in models. | 5.1 Explain what is meant by “overfitting” and “underfitting”. 5.2 Analyse the features, uses, benefits and drawbacks of the methods to prevent overfitting: cross validation; removing features; bagging; boosting; early stopping. |