www-ai.cs.tu-dortmund.de/LEHRE/SEMINARE/MUSIK/2006/Towards%20the%20Next%20Generation%20of%20Recommender%20Systems.pdf
cosð~xx;~yyÞ ¼ ~xx ~yy jj~xxjj2 jj~yyjj2
¼
P s2Sxy
rx;sry;s
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiP s2Sxy
r2x;s
r ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiP s2Sxy
r2y;s
r ;
ð13Þ
where ~xx [...] methods can provide more accurate recommendations than pure approaches.
2.4 Summary and Conclusions
As described in Sections 2.1, 2.2, and 2.3, there has been much research done on recommendation technologies [...] within a certain range). Then, for each user c 2 C, we want to choose such
item s0 2 S that maximizes the user’s utility. More formally:
8c 2 C; s0c ¼ argmax s2S
uðc; sÞ: ð1Þ
In recommender systems, the utility …
