Intelligent Tourism Personalized Recommendation Based on Multi-Fusion of Clustering Algorithms

Table 2

Determination rules for tourist attraction categories.

Integrate and modify user attributes, contextual information, and personalized information	Gender-user	The algorithm assigns the initial recommendation probability Gij to the tourist attraction category j according to the gender i and combines the gender of the tourist user and the electronic log to modify the recommendation probability
	Age-user	The algorithm first subdivides the age into different age groups and then assigns the initial recommendation probability Aij to the tourist attraction category j according to the age group i
	Time-user	The algorithm first subdivides the time into different time periods, then assigns the initial recommendation probability Tij to the tourist attraction category j according to the time period i, and revises the recommendation probability to according to the time of the tourist user
Calculate user’s short-term preference for tourist attraction categories	External-user	The algorithm determines the above two probability values, and according to the weighted average algorithm, it is concluded that the probability value of the tourist attraction category j recommended to tourist i is Kij