Data Powered Software

This course will provide the following:

1. Handle real world data produced from computer applications, such as user interfaces, services, and other programs.
2. Perform adequate data processes (e.g., feature extraction, selection and dimensionality reduction) to support software development.
3. Employ ML techniques (e.g., supervised, unsupervised, semi- and weak-supervised) to enrich contemporary software development.
4. Learn to Solve real-world problems by using software with data-powered modelling techniques (e.g., knowledge inference).

More concretely, we will cover the following topics:

1. Introduction to data and basic statistics.
2. Computer data pre-processing and calibration (e.g., data processing and signal processing techniques)
3. Knowledge extraction techniques from computer data (e.g., feature extraction, feature space reduction).
4. Basic modeling techniques (e.g., KNN, Naive Bayes, Gaussian process, Decision boundaries)
5. Advanced modeling techniques (e.g., SVM, random forest, ANN, and ensemble learning algorithms)
6. Empowering systems with unsupervised learning capabilities (e.g., personalization services and recommended systems).
7. Empowering systems with semi-supervised and weak-supervised learning capabilities (e.g., adaptive support).
8. Empowering systems with real-time capabilities (e.g., through time series analysis and forecasting)
9. Putting things together: Utilizing data-powered techniques to design and develop contemporary software.

Knowledge

1. Analysing real world data with statistics and machine learning
2. Feature selection and dimensionality reduction techniques
3. Application of supervised, unsupervised, semi- and weak-supervised learning

Skill

1. Translating real-world problems to machine learning space
2. Use of appropriate pipeline (series of methods)

Competence

1. Know the field of machine learning as seen from SE, IS, HCI industries

Lectures and a project. Each team has to submit project deliverables during the semester and a final report. Grading is team-based, but individual grades can be given in special cases.

The course will be held in English.