22LLP132 – Principles of Artificial Intelligence and Data Analytics
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
22LLP132 – Principles of Artificial Intelligence and Data Analytics
Weighting: 100% of the module assessment Feedback
This assessment has TWO parts. Answer both the parts and submit as ONE PDF document.
22LLP132 Coursework Report (100%)
Part I (Weight - 50%)
Select an emerging application and discuss the role of data-driven learning.
Your report should contain:
a) Short abstract (100 words) – weight 5%
b) Introduction and literature review with minimum 3 recent articles (200 words) – weight 10%
c) Methodology - weight 30%
· What data is collected, or what data can be collected (200 words)
· What insights can be gained from analyzing the data (200 words)
· What type of machine learning algorithm can be used for the data analysis (200 words)
d) Conclusions and Future Directions (100 words) – weight 5%
Your answer should be less than 1000 words.
Part II (Weight - 50%)
Linear algebra is foundational in data science and machine learning (ML). It is used in several ML and artificial intelligence (AI) applications, e.g., unsupervised learning, supervised learning, classification, principle component analysis, and support vector machine. Discuss the use of linear algebra in one of the ML and AI applications. (Word Limit: 300 words)
Question 2: Weight - 10%
In a feedforward neural network, the information moves in only one direction—forward— from the input nodes, through the hidden nodes (if any) and to the output nodes. The simplest kind of neural network is a single-layer perceptron network, which consists of a single layer of output nodes; the inputs are fed directly to the outputs via a series of weights.
Use the following input X to construct a single-layer neural network with matrix operation and use programming to test your results:
|
Q: What is the value of yi in the output layer? (Hint: use inner product) (Requirement: (1) show manual calculation steps; (2) show Python programming codes and results)
Question 3: Weight - 15%
The mathematical derivative is fundamental for optimizing machine learning algorithms. In deep learning, backpropagation gradient descent is widely used for training feedforward neural networks. Explain how derivative is applied in backpropagation gradient descent to optimize neural networks. (Word Limit: 300 words)
Question 4: Weight – 15%
Bayes theorem is the foundation for a classification method, known as “Naive Bayes”. Referring to an example, discuss how Bayes Theorem is used in Naive Bayes classification. (Word Limit: 300 words)
2022-11-22