22nd February 2010, 2 min read Most common questions about recommender systems… 3 thoughts on “Most common questions about recommender systems…” Daniel Haran says: February 22, 2010 at 12:48 pm My answer to #5 in commercial settings is “whichever makes you the most money”. Measuring profitability is no trivial task. Conversion rates, usage, churn and word of mouth can all be affected by the quality of the recommendations. There are hints in the literature about ways that recommenders can indirectly add revenue, see e.g. http://portal.acm.org/citation.cfm?id=1454054&dl=GUIDE&coll=GUIDE&CFID=77217953&CFTOKEN=60022310 Daniel Lemire says: February 22, 2010 at 1:18 pm Excellent point, profitability is important. Marcel Blattner says: February 23, 2010 at 2:21 am Nice post. Mixing methods is a good strategy, because the system is able ‘to learn’ (user feedback), which strategy fits best to customers (on a user level). However, it is important to ‘justify’ and to explain to users, why a particular item is recommended.
My answer to #5 in commercial settings is “whichever makes you the most money”.
Measuring profitability is no trivial task. Conversion rates, usage, churn and word of mouth can all be affected by the quality of the recommendations.
There are hints in the literature about ways that recommenders can indirectly add revenue, see e.g. http://portal.acm.org/citation.cfm?id=1454054&dl=GUIDE&coll=GUIDE&CFID=77217953&CFTOKEN=60022310
Excellent point, profitability is important.
Nice post.
Mixing methods is a good strategy, because the system is able ‘to learn’ (user feedback), which strategy fits best to customers (on a user level).
However, it is important to ‘justify’ and to explain to users, why a particular item is recommended.