A Book Recommendation Algorithm Based on Data Cleaning and Association Rules

摘要

Using data mining technology can extract valuable information from users’ book borrowing data, obtain users’ borrowing behavior, and provide personalized book recommendation service for users. The traditional association rule algorithms don’t carry out data cleaning before use, resulting in a single user’s single lending records becoming outliers in the overall data set, which makes the running time of the Apriori algorithm increase significantly. In this paper, according to the support threshold, confidence threshold and filtering threshold of the data set firstly, then the Apriori algorithm is used to analyze the association rules of the cleaned data set. The experimental results show that in the case of both large and small amount of data, the analysis time of Apriori algorithm with data cleaning is shorter, the strong association rules are stronger, and the effect is remarkable in the field of personalized book recommendation.