[Retracted] Analysis of User Personalized Retrieval of Multimedia Digital Archives Dependent on BP Neural Network Algorithm

Algorithm 1 Personalized retrieval algorithm of multimedia digital archives based on the BP neural network algorithm.

	Input: domain classification model, user interest model, search keywords, and search engine. Output: user personalized search results of the multimedia digital archives.
(1)	The search engine is used to generate a preliminary search result set X based on the search keywords.
(2)	Let the number of iterations be i = 0.
(3)	The iteration operation is set. For the i-th multimedia digital archive in the set X, the equation (1) is used to calculate the probability distribution in the classification model of the field.
(4)	The probability that the multimedia digital archive is recommended to the current user is calculated based on the equation (9) and added to the list Y.
(5)	If the multimedia digital archive i is the last multimedia digital archive in the set X, proceed to the step (6); otherwise, let i = i + 1, and return to the step (3).
(6)	The multimedia digital archives are sorted according to the probability in the list Y in descending order, and the results are output.