How to choose the right algorithm for the right task?

Nowadays, a wide variety of algorithms is available in different data analytics libraries or toolkits for data analytics. The question is therefore not whether an algorithm exists to solve your problem, but rather which one is the right fit for the job. Learn how to choose the correct algorithm for a specific task in our new two-session webinar on 18 and 21 May.

The next webinar in our Mastercourse 'Data innovation' 2021 will focus on one of the final and central steps in the data science workflow: choosing the most suitable algorithm to solve a specific problem. But what is the best way to get started? And what are the most common pitfalls? Our two-part webinar session on 18 and 21 May 2021 will shed light on this matter and provide clear answers.

Why join this webinar?

Choosing the right algorithm is crucial when applying data science. Due to the wealth of algorithms included in data analytics libraries and toolkits, the question is not if there is an algorithm available for the setting at hand, but rather which one is most fit to solve your specific problem. In addition, the way your business objective is formulated as a data science task can determine the type of algorithm you can apply.

The goal of this webinar is to introduce participants to the most important data science tasks (classification, clustering, regression, etc.) and provide an overview of the most commonly used algorithms and techniques to solve each of these tasks. It is open to anyone interested in applying data science.

For each of the methods, its characteristics, advantages and disadvantages will be explained in order to guide you in making a conscious choice in terms of the available data (dimensionality, attribute types, etc.) and the expected model requirements (interpretability, accuracy, scalability, etc.). Finally, the guiding principles to train and evaluate the resulting models, including an overview of common pitfalls and frequently-used evaluation measures, will be presented.

In this session, the following questions will be answered:

  • How to translate your business objective(s) to a data science task?
  • What are the most important data science tasks, and which machine learning algorithms and techniques exist to solve these tasks?
  • How to choose the appropriate algorithm based on important characteristics of the available data and expected model requirements such as accuracy, interpretability, scalability, etc.?
  • How to train and evaluate the resulting models, in order to arrive at the most optimal performance?
> Register for this webinar


Use case: Identifying sensor issues of solar panels

Industrial solar panels are equipped with sensors that measure a range of operational and environmental conditions. Problems with these sensors can arise for a variety of reasons, e.g. an incorrect clock setting, poor orientation, a sensor that is dirty or needs to be recalibrated, etc.

Industrial solar parks often contain hundreds or thousands of solar panels, which are closely and continuously monitored. In case there is an issue with one of the panels, it is important to know whether the fault is due to the panel itself, the sensor, the inverter converting the energy, etc.

In case the error is caused by the sensor, it is also helpful to know the nature of the error, so that it can be rectified quickly – and maintenance technicians know in advance exactly which specific issue has occurred. For this problem, we have developed a machine learning model.

The machine learning model is a classification model: from a series of known causes for errors, the model tries to identify the correct cause based on measured values and derived characteristics: the features. This means that the model is taught from a series of examples in which we already know the cause of the error (labeled data). Subsequently, the taught model can be used to classify new (unlabeled) data on that basis (and thus assign the correct error type).

Classification is one of the data science tasks covered in the webinar, along with some commonly used algorithms to learn such models. We also deal with various other tasks, such as clustering, regression, etc. ... 

Practical information

The webinar ‘How to choose the right algorithm for a specific task’ is part of the Mastercourse ‘Data Innovation’ 2021, organised by the Sirris Data and AI Competence Lab (EluciDATA Lab). It is given over two half days and can be followed as a stand-alone session or combined with other sessions. The mastercourse consists of 6 sessions, each focusing on different steps in the data science process.

When?

  • First half day: 18 May 09:30 - 12:00
  • Second half day: 21 May 09:30 - 12:00

> Register now for the webinar

 


#industriepartnerschap #sterkondernemen