Case Study booklet 2023 - May I recommend the following? (Required for HL paper 3)
Introduction
By next May, you need to THOROUGHLY UNDERSTAND the three main types of machine learning that is OUTLINED in the Case Study. The Case Study is not an explanation or a text-book, but rather a description of a problem to be STUDIED and INVESTIGATED. After doing some RESEARCH and READING, your level of understanding should become considerably deeper.
Structure of the paper
30 marks, 60 minutes, worth 20% of your final CS grade.
3 x Short answer questions worth ~6 marks and ~12 minutes each
Command terms: Define (2 marks), Outline (2 marks), Explain (4 marks), Describe (4 marks)
Compare, evaluate, discuss, to what extent
1 x Extended answer question worth 12 marks - 24 minutes
Through their investigation of the case study, students should be able to:
demonstrate an understanding of the computer science concepts fundamental to the system(s) in the case study (objective 1)
demonstrate an understanding of how the system(s) in the case study work (objective 1)
apply material from the course syllabus in the context of the case study (objective 2)
explain how scenarios specified in the case study may be related to other similar local and global scenarios (objective 3)
discuss the social impacts and ethical issues relevant to the case study (objective 3)
explain technical issues relating to the case study (objective 3)
evaluate information that may be gathered from local and global sources including field trips, interviews, primary and secondary research, invited guest speakers and online interviews (objective 3)
evaluate, formulate or justify strategic solutions based on the synthesis of information from the case study itself, additional research and new stimulus material provided in the examination paper (objective 3).
Behavioural data
Cloud delivery models:
Infrastructure as a service (IaaS)
Platform as a service (PaaS)
Software as a service (SaaS)
Cloud deployment models
Collaborative filtering
Content-based filtering
Cost function
F-measure
Hyperparameter
K-nearest neighbour (k-NN) algorithm
Matrix factorisation
Mean absolute error (MAE)
Overfitting
Popularity bias
Precision
Recall
Reinforcement learning
Right to anonymity
Right to privacy
Root-mean-square error (RMSE)
Stochastic gradient descent
Training data
YouTube videos