Specific research topics include search technologies, machine learning methods and knowledge extraction from large data sets, presentation and visualization of massive amounts of information, efficient use of information in social media channels, as well as contextualization and personalization of information.
Our research areas are:
- Knowledge Discovery
- Knowledge Visualization
- Social Computing
- Ubiquitous Personal Computing & Business Models
- Data Management
- Data Security
Our Business domains are:
- Industrial Data Analytics
- Data-Driven Markets
- Strategic Intelligence
- Data-driven Process and Decision Support
- Learning 4.0
- Digital Life Science
PLATFORM AND SERVICES INFORMATION
|CPU cores||360 in 15 servers|
Our offer includes consultations, data analysis and training. The Big Data Lab offers a simple and direct access to our expertise and infrastructure. In addition to Apache Hadoop, the Big Data Lab is equipped with other big data technologies such as Apache Spark and Apache Storm on its computer clusters. Integrated within an international network around the topic of Big Data and Data Science, the Know-Center provides its partners with access to the latest trends and findings in this area.
SELECTED PROJECTS AND/OR SUCCESS STORIES
Porsche Holding Salzburg is the largest and most successful automotive distributor in Europe. The Salzburg-based company was founded in 1947 and operates today in 22 countries in Western and South-eastern Europe, as well as in China, Colombia and Chile. Its subsidiary Porsche Austria GmbH is importing and distributing cars to dealers and Customers across Austria. Together with the Know-Center, Porsche Austria is interested in analysing the quality of data from the market introduction and the market performance of new car models in the past in order to create a forecast. These models are then used to forecast upcoming market performance in terms of new car registrations or sales. This is of interest for current car models, but even more for upcoming new models. The project aimed at forecasting the demand on the number of cars of a specific brand overall or within a dedicated segment. A predictive model forecasts the demand for a period of a month up to one year. The decision on the model is based on an evaluation of a nonlinear approach from the research field of Deep Learning and a linear approach (Seasonal Autoregressive Integrated Moving Average, SARIMA). The linear model yielded the most promising results that hold true for both short-term and long-term demand forecasts.
The overall goal of the Data Market Austria (DMA) project is to develop the technological, infrastructural, regulatory, and economic foundations for a comprehensive, innovation-supporting, sustainable Austrian Data-Services Ecosystem, building on existing initiatives. One central part of DMA will be its brokering functionality which shall foster interactions between dataset owners, service providers (e.g., analytic services) and data market customers. Know-Center is leading the corresponding work package “Brokering Technology Foundation” which aims to generate recommendations for possible collaborations between these three groups. This is done by the development of a matchmaking framework (*) that extracts information from dataset and service descriptions as well as interactions with the DMA. The extracted information is the basis for recommendations and matchmaking algorithms. With regard to datasets, services for the analysis of this data are suggested or other data for enrichment and combination might be recommended. Similarly, with regard to services, potential input data as well as pre- and post-processing services or use case scenarios , i.e., potential usage of a dataset in combination with services, might be proposed.
(*) Basis for this framework is ScaR, a scalable recommender framework developed by Know-Center.