This event is hosted by Elektroteknisk Gruppe and IT-Gruppen, IDA ØST /IDA IT and will be presented in English.
Recommender Systems have become a favorite tool used to stay afloat in a world of information overload. End-users are guided and lead by Recommender systems to relevant information - most notably applied in large-scale operations Like Netflix and Amazon where Recommender systems are vital in user interactions - it is a concept that is becoming defused into most online businesses, where a large selection of products. To use recommenders in practice touches on every aspect of the interaction between users and content. From data collection, understanding users by doing data analysis, predicting with machine learning and finally the presentation of the content.
Attendees are expected to have a basic understanding of maths and programming. Nevertheless, Kim Falk will exemplify with drawings and figures during his presentations.
Talk 1: Recommender Systems - What do the users want and what can a Recommender System do.
2/3's of the content watched on Netflix are recommended, likewise 35%
of the sales at Amazon. The methods and machine learning used to
calculate what to show the user is called Recommender systems.
Applying recommendations in practice touches on every aspect of the
interaction between users and content. From data collection, and
understanding users doing data analysis, machine learning and finally
presentation.
Recommender Systems has great potential, in almost every sector, but
it is not a silver bullet. In this talk, we will look at a general
overview of recommender systems and touch upon some promises, and some
obstacles.
Talk 2: Recommender systems in Ads and other practical examples
This talk will go into more details of one (or two) recommender
systems implementations. We will look at how an e-shop uses
recommender systems in ads.
Speaker:
Kim Falk is a data scientist, passionate about recommender systems and machine learning in general. He has trained recommender systems to provide movies choices to end-users as well as ads to people. He has worked with Big Data solutions and Machine learning since 2010.