Research on House Recommendation Model Based on Cosine Similarity in Deep Learning Mode in Grid Environment

Aiming at the current number of recommended models in the house recommendation system model, the recommendation accuracy and user satisfaction do not meet the needs of users, this paper proposes a house recommendation model based on cosine similarity in the deep learning mode in grid environment. The model first uses the established grid environment, combined with the deep learning mode, through the collection of a large number of housing basic data samples and user feedback information, establishes a mathematical model of housing recommendation, and then integrates the cosine similarity into the recommendation model, through training and parameters. Adjust the model framework gradually. Finally, through the derivation and verification of the simulation experiment, the proposed recommendation model can efficiently recommend the housing information that meets the requirements for the user, and improve the recommendation accuracy and user satisfaction.