Resit2022Proposed solution included

CEC 435 Machine Learning Resit 2022

CEC 435: Resit examination for CEC435, Machine Learning, College of Technology, University of Buea, instructors Tchagna A. / Foping, July 2021/2022 academic year. Part I, Exercise 1 is a 5-mark MCQ covering confusion matrices, hyperparameters in model-based learning, the effect of K in K-Nearest Neighbors, the correct preprocessing order (normalize, PCA, then train), and which statement about k-means clustering is false. Exercise 2 (25 marks) is an open-question section: defining supervised, unsupervised and reinforcement learning with two applications each, explaining Support Vector Machines for classification, and describing customer segmentation via K-Means. Part II (40 marks) is a practical linear regression exercise using a table of used-car age versus price: plotting to show the negative correlation, approximating the intercept B0 and slope B1, interpreting the results, and explaining methods for finding intercept/slope when many features are involved (multiple regression).

Open in Shoracy

Opens this paper in the app on your phone.

University of Buea

College of Technology - COT

Computer Engineering - CET

CEC4352022

Machine Learning

Resit · First Semester Resit · Level 400

From this paper

1.
Which of the following matrices can be used to show the results of model accuracy evaluation or the model’s ability to correctly predict or separate the classes? a- Confusion matrix b- Evaluation matrix c- Accuracy matrix d- Error matrix e- Identity matrix
2.
In Model based learning methods, an iterative process takes place on the ML models that are built based on various model parameters, called? a- mini-batches b- Optimized parameters c- Hyperparameters d- superparameters