Andre announced a first release of CISTI's Synthese Recommender system for journal articles. It uses Taste and the article citations as a substitute for user preference data.
You can search the 1.5 million BioMed articles and add them to your 'basket'. The system then recommends new articles you should read.
I'm curious about the approach of using article citations as a substitute for user preference data. That seems quite different in spirit from applying user preference data, and if anything feels closer to content-based recommendation than collaborative filtering. I'm curious if this is a common strategy in the research community.
ReplyDeleteSame thoughts here. Journal citations are a measure of how paper authors value other papers. That's valuable, but doesn't take into account my preferences.
ReplyDelete