Skip to main content

Institute for Data Science

We process data using computer science methods to generate knowledge and added value.

Data Science supports well-founded, data-driven decision-making and improves processes. Partners from different business and research fields work with us. They benefit from IT technologies we developed in the context of our international space projects.

In addition, we are committed to shaping the digital society of the future in the areas of science communication and the aging society.

Competencies

Machine Learning

Classification and regression models, predictive modeling, time series analysis, recommendation systems, deep learning

Natural Language Processing

Topic modelling, sentiment analysis, text analysis and generation, speech recognition

Exploratory Data Science

Explorative data analysis, descriptive statistics, data visualization, analysis of (social) networks

High Performance Computing

Parallelisation on distributed systems, performance optimisation in the use of GPUs, hybrid computing, design and construction of infrastructures for data management and processing

Image Analysis and Image Processing

Image reconstruction, image segmentation, object recognition, data compression and dimension reduction

Selected projects

STIX for Solar Orbiter

The Spectrometer / Telescope for Imaging X-rays (STIX) is an X-ray telescope that makes images and spectra of solar explosions called flares. It is one of the ...

to STIX for Solar Orbiter

Yooture - The app for job seekers

The School of Engineering FHNW has developed an elaborated matching method for the innovative start-up Yooture.

to Yooture - The app for job seekers

Compressed Sensing for Solar Flares

For the RHESSI satellite we develop an algorithm to reconstruct X-ray images of solar flares.

to Compressed Sensing for Solar Flares

FHNW Myosotis Garden

Playing games with elderly people can be fun. It can facilitate communication between people from different generations.

to FHNW Myosotis Garden

IRIS Big Data

Machine learning methodology to detect, analyse and possibly predict solar flares from data provided by NASA's space telescope IRIS.

to IRIS Big Data

Optimizing hydro power plant maintenance planning

The School of Engineering FHNW has developed a method to support operators of AXPO hydro power plants.

to Optimizing hydro power plant maintenance planning

SySTEM2020: Connecting Science Learning Outside The Classroom

Research participant recruitment, data collection and mapping STEAM initiatives in Switzerland.

to SySTEM2020: Connecting Science Learning Outside The Classroom

kennwerte.ch: Web application for estimating construction costs

An innovative solution for the estimation of characteristic values for the construction and real estate industry for the start-up company ‘kennwerte ag’.

to kennwerte.ch: Web application for estimating construction costs

Projects with partners from the industry

  • Price prediction in the real estate market (construction costs, insurance prices, bid prices, rental prices etc.)
  • Recommender system for job placement solutions, auction platforms, investment instruments etc.
  • Chatbots for various industries
  • Automatic processing of medical documents (e.g. side effects of medication)
  • Predictive logistics management (for seasonally fluctuating business segments)
  • Algorithms for image reconstruction in Astrophysics and image segmentation in Neuroscience
  • Data processing pipeline development spanning multiple European supercomputing facilities
  • Prediction of customer churn to stabilize business performance
  • Predictive modeling to identify client potential for cross-selling
  • Optimization of industrial production planning

Contact

Prof. Dr. André Csillaghy
Prof. Dr. André Csillaghy Head of FHNW Institute for Data Science
Telephone : +41 56 202 76 85 (direct)

FHNW School of Engineering, Brugg-Windisch

FHNW University of Applied Sciences and Arts Northwestern Switzerland School of Engineering Bahnhofstrasse 6 CH - 5210 Windisch
More information about the location
Share this page: