by Gabriel Aldamiz-echevarría
-
October 16, 2008 at 9:06 am
· Filed under RecSys07-08
We are very excited to announce the five finalists for the first edition of the Strands $100K Call for Recommender Start-Ups.
As you know, Strands seeks to identify the best early-stage project in the area of recommendation technologies. Finalists are invited to present their projects during the ACM Conference on Recommender Systems (RecSys08) this October 23rd to 25th, 2008 in Lausanne, Switzerland. The Winner will be offered an investment of $100,000 from Strands, Inc.
A total of 68 scientists and entrepreneurs from 24 teams in 15 countries presented 26 projects. Countries included Australia, Austria, Canada, China, Finland, Germany, Greece, Hungary, Japan, Mexico, Romania, Spain, The Netherlands, Turkey and the USA. Areas where the recommendation technologies were applied are: pharmaceutical industry, cosmetics, video content, online media content, people and relations, TV content, search, events, social enterprise, advertising, location based services, travel and food.
It is interesting to note that two of the finalists, Commendo and Gravity R&D, are groups formed as a result of their efforts at the
Netflix Prize competition, and
now rank 2nd and 5th respectively.
And here is the list of the five finalists. Good luck to you all in Switzerland, this is very exiting for all of us!!
IMPRESS – Instant Entertainment, by Gravity R&D
IMPRESS “magic button” provides TV viewers instant personalized entertainment at any given time with relevant program tips instantaneously on customer demand. It automatically schedules recordings with the highest probability on user’s interest.
Gravity R&D was founded to exploit the experience learned at the Netflix Prize competition. It is specialized in algorithms, methods and practical application development for recommendation systems.
Team of 4: Domonkos Tikk, István Pilászy, Gábor Takács and Bottyán Németh. University: Budapest University of Technology. Hungary
SentiMetrix, Opinion-based Recommendations
SentiMetrix extracts and quantifies the huge and growing number of text-based opinions expressed through online media worldwide. SentiMetrix has automated sentiment extraction/analysis/scoring, the ability to find and quantify opinions in text.
SentiMetrix was founded in 2007 and it is based on the award-winning OASYS technology.
Team of 4: Patrick McBride, Vadim Kagan, V.S. Subrhamanian and Diego Reforgiato. University: University of Maryland Institute for Advanced Computer Sciences, USA
iletken, Personally Relevant Content Based On Social Connections

Iletken, which means “Conductive” in Turkish, is a hybrid recommendation engine for services like news and RSS feeds. The core system of “Iletken” mimics the dynamics behind real life information dissemination to enable friends to share and conduct relevant information.
In contrast to traditional social networks, it maintains weighted graphs of social proximity among users for different categories of interest. Analyzing user behavior, Iletken dynamically updates its social graphs and uses this data to determine information conductivity among peer.
Team of 4: Murat Deniz Oktar, Baris Can Daylik, Firat Gelbal and Selcuk Atli. Universities: Koc University Istanbul, RPI New York and TU/e, Eindhoven.
Reccoon, location-based recommender

Taking advantage of the vast amount of GPS-enabled mobile devices, Reccoon helps people discover new places based on their input, current location and current time. It also adjusts recommendations to attention profile (e.g. Last.fm listening history). Reccoon responds to the simple question: what are my options for tonight?
Team of 2: Peter Tegelaar and Dominiek ter Heide. The Netherlands
Commendo - Drug Design Recommendation

Commendo uses recommendation technologies to optimize the drug design process in the pharmaceutical industry, including speeding up drug development and the minimization of adverse drug reactions.
Commendo was founded out of the project team “BigChaos”, which still participates very successfully in the Netflix Prize competition.
Team of 4: Andreas Töscher, Michael Jahrer, Georg Preßler and Michael Schrotter. Universities: Technical University Graz and University of Applied Sciences Mittweida, Austria