Design of Library Big Data Document Service System Based on Improved Genetic Algorithm

Literature service covers data retrieval, literature standards and other aspects, and has the characteristics of large amount, many types, high value and low density. Therefore, providing efficient and accurate literature service for readers has become the main task faced by modern libraries. It is of great significance to find a balance between massive data and limited computer processing power, and to study and design a feasible document retrieval service scheme. In order to build a more intelligent library big data (BD) document service system, this article puts forward an incomplete BD filling and mining method based on improved genetic algorithm (GA). The improved GA is used to copy, hybridize and mutate the matrix, and a recommendation list with different lengths is established by ranking with mixed weights, which effectively completes the filling and mining of incomplete BD and realizes the personalized document service of university libraries. The results show that the algorithm still has a very high literature recommendation recall rate and a low MAE in the huge book system, and can meet the user’s demand for searching interested documents.