About us

DB4ALL is a Start-Up at the Database Laboratory at EPFL funded by Innogrant. We are experts on decision support systems and database collection, integration and cleaning.

Our objective is to promote the use of datasets to users and companies for making more informed decisions. By datasets we refer to the content (list of books, movies, restaurants, jobs, biochemical products, zip codes...).

The idea for this venture started when we realized that there are amazing collections of public datasets in Internet that are not just available for downloading in a structed format; we acquired the know-how to extract these datasets for latter processing and analysis.

We welcome partnerships and system integrators. Please contact us and describe your project.

 

Testimonials

We needed an example catalog of products to build our demo for one of our recent FMCG customer and DB4ALL did a great job extracting leshop.ch catalog.
-Etienne Adam, Managing partner at Fastprod.

The extensive competences of DB4ALL staff in the domain of information extracion gave us the possibility to build a very efficient environment for managing complex linguistic data with adequated XML based tools.
-Martin Rajman, EPFL.

Whenever I wanted to mine web data, I've always reverted to doing everything from scratch due to the lack of flexibility in the existing tools. DB4ALL Webminer solves this issue.
-Maxim Raya, PhD student

When conducting research in social science, we build constructs which are linked to variables that often require extracting from hundreds of company websites citations from their leaders. Typically we need to find information on kind of leadership, organic growth, innovation, sustainability, corporate social responsibility, etc. We are impressed with DB4ALL solutions for this purpose.
-Felix Cardenas, PhD Candidate in Management Control and Innovation

I consider myself fortunate to have been one of the alpha cases on which Dr David Portabella has applied the webminer software. In two occasions this semester I have asked him to scour the net on my behalf to find data which would have been of fundamental importance for the course I teach, in which his project was used as a reference. I had tried to look up that kind of information last year, for the same reasons. The exercise took me over a week to complete, using up a lot of evenings and nights. I hear it took him and his team less than 15 minutes to produce the results I then used and easily integrated in my existing database.
-Vittorio Mischi, Lecturer, Entrepreneurial Finance, College of Management, EPFL

Our laboratory consistently faced a communication problem: nobody in our lab doubts the seriousness of the current rise in CO2 emissions, but convincing the general public required well-presented, publicly available data. We often need, in particular, data on historical CO2 trends; historical global average temperatures; fuel prices; oil reserves; global energy consumption, broken down by industries; etc. Many organizations provide this data, but mostly as tables on web pages. As far as I know, nobody provides a relational database with this kind of data. I have experimented with DB4ALL's set of tools for normalizing arbitrary data sources into a relational database model, and the results have impressed me. Having such a regularly updated database at our disposal would not only facilitate communication; it would also encourage a "playful" exploration of the data at hand, and perhaps lead to the discovery of hitherto unsuspected trends or correlations.
-David Lindelöf, PhD, EPFL - Solar Energy and Building Physics Laboratory

When researchers want to validate their theories, they often use small and simple data sets that are easily available on some well known websites. The problem with such data sets is that they are over-used in research papers and their interests are limited. They are what I call “toy data sets”. The extraction tool of DB4ALL provides the access to much more interesting data on the web. These data are not yet used in research. Thanks to this extraction tool, these data can be gathered to be then studied and mined for knowledge extraction. This gives research teams the opportunity to publish articles that are validated on real industrial data sets.
-Dr Sandro Saitta, www.dataminingblog.com

 

Partners

  • Macrofocus.com Macrofocus is a leading software manufacturer of advanced Data Visualization components and applications. DB4ALL collaborates with Macrofocus for visualiazing large amounts of data extracted from public websites.
  •  Fastprod is a web-based Supply Chain platform. Utilizing proved technologies, professional methodologies and practical industry knowledge, our competences are enhancing organizations competitiveness in Supply Chain. Fastprod collaborates with DB4ALL to extended customer internal databases with external public information.
  •   We have a strong relatioship with EPFL and in particular with the Database Laboratory, allowing us to keep innovating.
  •  Venturelab and the College of Managament of Technology has provided us with several entrepreneurial training sessions.

 

Awards

IMD Startup Competition, 2010 Venture Kick I start-up competition, 2009 Innogrant, 2008

 

Careers

Currently we have no open positions, but we are always keen ho hear from exceptional talents. If you have a passion for excellence and innovation, we invite you to submit your resume and a motivational letter explaining how you can add more value to our customers.