Nicht nur in der Mailingliste (Apache Mahout Status by Grant Ingersoll) des Projekts nimmt der Traffic zu, auch erste Projekte die Mahout erfolgreich einsetzen werden bekanntgegeben.

So geschehen bei dem (mobilen) News-Aggregator und Organisator Mippin aus Großbritannien. Mippin sammelt Informationen über das Surf-Verhalten seiner Nutzer und erstellt profilbezogene Vorschläge Themen-ähnlicher Websites. D.h. je mehr ein Benutzer die Software “mitlesen” lässt und je mehr Benutzer Mippin benutzen, desto besser funktioniert das Vorschlagwesen – sowas nennt man dann Collaborative Filtering.

Mippin nutzt dafür das Recommendation System in Apache Mahout und bedient sich damit denselben Tricks wie Amazons Empfehlungs-System. Laut Sean Owen (Mipping & Committer bei Apache Mahout) soll mit Apache Mahout das Ziel erreicht werden, Machine Learning Algorithmen für alle Entwicklern zur Verfügung zu stellen, damit diese Disziplin keine Raketenwissenschaft bleibt.

Hier noch einige Quellen zum Thema Apache Mahout:

• InfoQ: Apache Mahout, Highly Scalable Machine Learning Algorithms
• Cyberstrategics: Running k-means Clustering
• Iletken Blog: Mahout Review (Kritisches Benchmark)

Events like the BarCamp last weekend are always nice to get in touch with new people and learn a lot of things. Since the BarCamp wasn’t dedicated to a certain topic there was also a variety of people from different branches and I was excited wether I could get any information of my favorite topic (Information Management). Here are my take-aways, meaning interesting stuff I stumbled upon:

Yovisto (source: just from talking to someone)
Yovisto is an academic video search engine to search through educational video content mostly from universities. Beside just searching through video’s metadata, Yovisto is also able to connect presentation slides from presentations to the according video record.

QUEAP (source: session on sunday)
Queap is a web search engine which is currently in development. With it’s preprocessing approach Queap has the ability to categorize websites by using some kind of “artificial intelligence”. There is no beta yet, I’m still waiting for a demo to really test it! A brief presentation is available here.

Directed Edge (source: session on saturday)
Scott Wheeler held a presentation about collaborative filtering at the BarCamp on saturday (together with a representative from be2). He is the Co-Founder of Directed Edge, which is a company that aims to the “niche” of recommendation engines. A demo of their recommendation engine can be found here.