The module aims to provide the student with an introduction to artificial intelligence and machine learning using python. The module develops analytical skills to understand the significance of data and processes involved to convert data into valuable insights. The module gives the students a detailed understanding of both the role and purpose of machine learning techniques leading to data-driven decision making.
*Curricular information is subject to change
Semester 1
- Introduction
- Data Science, Big Data, FinTech, and Roles (Data Analyst, Data Scientist, DPO)
- Emerging Issues in FinTech
- Python Basics
- Develop understanding concerning variables, constants, data types, variable assignment, printing, etc
- Lists
- How to create lists, and perform operations specific to lists such as appending an item to the list, extending a list with another list, deleting an item, subset and slice lists, etc
- Functions and Packages
- What is the use of function? (e.g., Sum is a function)
- How to define a custom function along with arguments to achieve the desired functionality.
- How to seek Help in Python?
- How to import a library? And what are some useful ways import functions defined in python libraries
- Numpy
- Exploring mathematical array-based operations using Numpy package
- Logic, Control Flow
- Define logic, compare values
- What are comparison operators and boolean operators?
- Define rules using if, elif, else
- Making use of Boolean operators with Numpy
- Dictionaries
- Lookup data as a key-value pair. Create and Access dictionary and look through dictionaries and learn how to manipulate data inside a dictionary
- Loops
- Learning how to iterate over data using for and while loops. Understanding how loops support operations list of data items, dictionary composed of key-value pair, and Numpy array for mathematical operations.
- Visualisation
- Explore data using visualisation that helps separate data belonging to different classes.
- Exploring visualisation using Matplotlib and Seaborn
- Lambda, map, reduce, filter
- Learn the foundation of big data operations using functional programming (i.e., operations over lists)
Semester 2
- Kinds of Data and transformations
- What are different kinds of data? How can we transform data to extract meaningful insight
- Data can represent a network, a structured hierarchy, or unstructured. How to deal with different types of data?
- What are some popular data exchange formats (XML, JSON, etc.)
- Exploratory Data Analysis
- Exploring data using Pandas, and practice different strategies to plot data using Histogram, Beeswarm, ECDF, Outliers, Boxplot, Scatter & Pair Plot, Pearson Correlation Coefficient
- Classification
- What is meant by classifying data into different categories?
- What is meant by labelled data?
- What are some popular Machine Learning algorithm that allows data to be classified into different classes (e.g., Nearest Neighbour Classifiers, Decision Tree Random Forest)
- Text Analysis
- What is the role of text analysis in Machine Learning?
- How can text data (qualitative data) be transformed into quantitative data?
- What is meant by n-grams and Bag-of-words models
- What are some popular choice for text analysis
- Algorithms: Naïve Bayes, Topic models, Word Embeddings
- Applications: Sentiment Analysis and Chatbots
- Ensembles
- When performance matters the most, Ensembles help form a council of machine learning strategies. Wisdom of one vs wisdom of many
- What is Ensemble Classification and Why do ensembles work?
- Clustering
- How to identify groups and sub-group within data?
- What is meant by unlabelled data
- What are some popular clustering algorithms: K-Means, Elbow Method, Hierarchical clustering, t-SNE, PCA, NMF
- Model Evaluation
- How do we evaluate the performance of Machine Learning algorithm?
- What are some key concepts and practice?
- Concepts and practices: Confusion Matrix, Accuracy, Precision, Recall, F-Score, Precision-Recall Curve, ROC, Cross-Validation, Hyperparameter tuning
- GDPR, Ethics and laws in Data Analytics
- Ethics is essential everywhere in the world? When it comes to data, it means with data shall come responsibility. What are laws and ethics when it comes to big data analytics?
- Current and Future Trends for Machine Learning and Artificial Intelligence in Fin Tech
- Ensuring access to the latest trends and research in Machine Learning and AI and its application in the Fin Tech Industry
The module will be delivered via a combination of lectures and practical sessions where case studies will be discussed.
| Module Content & Assessment | |
|---|---|
| Assessment Breakdown | % |
| Other Assessment(s) | 100 |