Recommendation System based on tags -

i want design recommendation system goes this:

• i have many restaurants different tags

• i have users making searches using tags

• i want make recommendations based on tags users have search most

i not looking complicated algorithm.

for example:

i have user has searched:

tag1 - n1 times

tag2 - n2 times

tag3 - n3 times

tag4 - n4 times

tag5 - n5 times

and there 3 restaurants corresponding tags:

restaurant1: tag1, tag2, tag4, other_tag

restaurant2: tag5, other_tag

restaurant3: tag1, tag4, other_tag, other_tag

i thinking following logic:

let n = n1 + n2 + n3 + n4 + n5

let t_i = number of tags i_th restaurant

then i'll compute:

r1 = sum(is_tag_i_in_restaurant1 * ni) / t_1, goes 1 5

r2 = sum(is_tag_i_in_restaurant2 * ni) / t_2, goes 1 5

r3 = sum(is_tag_i_in_restaurant3 * ni) / t_3, goes 1 5

t1 = n / t_1

t2 = n / t_2

t3 = n / t_3

and each restaurant compare ri ti. let say, if ri >= ti/2 consider recommendation.

is way of doing this? can recommend me more efficient?

there's scholarly research regarding tags @ google scholar , http://grouplens.org/publications, in particular jesse vig, shilad sen, tien nguyen, , john riedl's work. lot of thought has gone predicting tag preference well. check out tagommender paper. in general, users , items (restaurants or movies or other things) have values tags on scale, such preference or appropriateness. find nearest neighbors using similarity algorithm cosine similarity.