Research on the Application of User Recommendation Based on the Fusion Method of Spatially Complex Location Similarity

Algorithm 1

Recommendation algorithm based on geographic location preference.

(i)	Input: user set U, historical access record set F, u is the target user, and r is the user to be recommended
(ii)	Output: Top-u.
(iii)	Algorithm steps:
(iv)	For u in U
(v)	{ For r in U
(vi)	{Calculating the similarity between the user u and the recommended user r according to formulas (9), (11), and (13);
(vii)	Calculating the preference prediction value of the user u in U according to formulas (7) and (8);
(viii)	Calculate the similarity of the similar user recommendation with preference according to formulas (14) and (15), value;
(ix)	}
(x)	}
(xi)	For u in U
(xii)	{
(xiii)	For r in U {
(xiv)	Calculating the recommended score according to formula (16);
(xv)	}
(xvi)	}
(xvii)	Sorting in descending order according to ;
(xviii)	Pick the previous user of the list as the recommended object.