Practical Machine Learning & Artificial Intelligence Course
In recent years,“Machine Learning” and “Intelligent machines”are among the most used and searched-for words. The reason for this is mainly due to the’exponential increase in the amount of data produced, to the increase in computing power and the advances made in the development of more efficient algorithms.
Machine Learning is used everywhere, often without our knowledge, with the aim of creating new value from data, for companies in every sector. A large portion of the tools we use daily, from recommendation systems to facial recognition, from fitness trackers to home assistants, analyse data and make decisions through these algorithms. The application cases are numerous and, often, are limited only by imagination.
The main objectives of Machine Learning consist of to understand the data structure, analyse them using intelligent algorithms and models and generate new insights which can be easily understood and used by people. Compared to traditional programming, Machine Learning algorithms can learn from the input data and, by using statistical analysis, produce predictions, classify information, make decisions, recognise images and sounds, and more.
This course, starting with an overview of the most common methodologies and models, supervised versus unsupervised learning, and the most common algorithmic approaches, aims to provide learners with the theoretical and practical foundations to begin applying the main Machine Learning algorithms to their own real-life cases. The examples are implemented in Python using the most popular M.L. libraries.
Contents
What they are, what they are used for, and the differences between the concepts: the Machine Learning process; what a model is; supervised versus unsupervised learning; introduction to the most popular Machine Learning models; criteria for choosing between models.
(Jupyter Notebook and Anaconda).
(Pandas, Matplotlib, Plotly).
Data reading, writing, and creation; data indexing, selection, assignment, and renaming; data summarisation, mapping, and reporting; data visualisation (main tabular and graphical formats).
(Pandas).
Grouping and sorting data: grouping, pivoting and joining; feature selection; handling incorrect data; handling missing values; manipulation of datasets in 1D, 2D and 3D; data normalisation; splitting and creation of training and testing datasets.
(Scikit-learn and Keras).
Supervised learning - Linear and Logistic Regression; classification (SVM, Decision Tree, Random Forest); Unsupervised learning - clustering (K-nearest neighbor); PCA; Reinforcement learning - Q-Learning.
Introduction to models and various use cases; Perceptron; CNN; LSTM.
Scoring (CM, ROC), interpretation of results.
Prerequisites
- Basic programming knowledge
- Knowledge of the main statistical concepts is recommended.
- Basic programming knowledge
Classroom requirements
- Video projector with a minimum native resolution of 1024x768 (preferably higher)
- Unfiltered internet connection for the teacher's laptop
- As this is a practical course, participants will need to be provided with a PC.
NOTEThe lecturer will use their own laptop, on which all the course examples are installed. If this is not possible, the provision of a PC supplied by the client must be agreed in advance.
Recipients