O'Reilly logo
live online training icon Live Online training

Machine Learning for IoT

Applications to spectral and sensor data

Pablo Maldonado

Machine Learning is disrupting many industries, but this is nothing compared to what comes ahead of us: the power of machine learning algorithms combined with data coming from IoT devices. In this course, we focus on a particular type of IoT data: spectral and sensor data. Spectrometers are cheaper and more precise, and this will have tremendous consequences in our lives and careers, opening up new opportunities for disruption. Many companies are already using data from spectrometers and sensors to help prevent and diagnose diseases, analyze food composition, and identify human activity. We will provide an overview of such algorithms and case studies through practical labs.

What you'll learn-and how you can apply it

By the end of this live, online course, you’ll understand:

  • What is the Internet of Things and why sensor data is important?
  • How is sensor data stored and used?
  • What makes sensor data somewhat different from standard tabular data.

And you’ll be able to:

  • Identify the likelihood of a heart attack from ECG data.
  • Measure how likely is a cheese to be tasty, given their spectral data.
  • Recognize human activity from sensor data.

This training course is for you because...

This session is targeted for anyone with basic software development skills (data scientists, software engineers, amateur programmers, and managers) looking to dig deeper into the applications of machine learning to sensor data. You will also gain valuable practical knowledge and a new perspective on your day-to-day challenges. Real-life code examples will be provided.

Prerequisites

Working knowledge of R and/or Python and familiarity with calculus and probability.

Recommended preparation

R Deep Learning Projects is a suggested introduction to some of the methods which we will use. A Github repository with course materials will be also available.

About your instructor

  • Pablo Maldonado is an applied mathematician and data scientist with a taste for software development since his days of programming BASIC on a Tandy 1000. As an academic and business consultant, he spends a great deal of his time building applied artificial intelligence solutions for text analytics, sensor and transactional data, and reinforcement learning. Pablo earned his PhD in applied mathematics (with focus on mathematical game theory) at the University Pierre et Marie Curie in Paris, France.

    Pablo is the founder of Maldonado Consulting which is a technology-agnostic data analytics consultancy based in Prague, Czech Republic, that leverage the latest tools and research to develop custom solutions around like Data Analytics, Mathematical Modeling, and Machine Learning and Artificial Intelligence.

    Pablo has been an adjunct professor, teaching AI (Reinforcement Learning) and Machine Learning at Czech Technical University in Prague, the oldest technical university in Central Europe. He has co-authored a book “R Deep Learning Projects” published by Packt.

Schedule

The timeframes are only estimates and may vary according to how the class is progressing

Section 1: Introduction to IoT and sensor data (20 min)

In this section, we will cover some examples and use cases of the internet of things and sensor data. We will answer questions such as: - Why is this relevant for our daily lives? - How is sensor data collected and stored? - Examples of analysis that can be done. (These examples will be detailed later during the lectures)

Section 2: Supervised Learning: Classification (30 min)

  • Reviewing the basic algorithms for classification of sensor data.
  • Logistic Regression: definition and examples.
  • Partial Least Squares classification.
  • Dimensionality reduction.
  • Neural networks and deep learning.
  • Lab 1: Classifying ECG data.

Section 3: Supervised Learning: Regression (30 min)

  • Reviewing the basic algorithms for regression in sensor data.
  • Ridge/Lasso Regression: definition and examples.
  • Partial Least Squares for regression.
  • Dimensionality reduction.
  • Neural networks and deep learning.
  • Lab 2: Measuring nutrient content in food from spectrometry data.

Section 4: Unsupervised Learning (30 min)

  • In unsupervised learning, aiming to use sensor data to identify different patterns. This is much more challenging, yet a very important application.
  • Manifold learning.
  • Topological data analysis.
  • Lab 3: Human activity recognition.

Section 5: Wrap-up and lessons learned (10 min)

  • Summarizing the main messages and suggest how you could apply your newly acquired knowledge in a number of settings.
  • How are companies benefiting from these techniques?
  • Suggestions for projects: build your portfolio.