Abstract: 【Purpose/significance】Aiming at the problems of the current small and medium-sized enterprise information collection
system model, such as the accuracy of information collection, the recall rate of information retrieval and the low efficiency of informa?
tion classification management, this paper proposes the construction of small and medium-sized enterprise information collection sys?
tem model based on LDA and fuzzy VIKOR method.【Method/process】According to LDA model, design and build the model architec?
ture of information collection system for small and medium-sized enterprises, collect knowledge resources through enterprise manage?
ment architecture, divide the acquired knowledge into corresponding modules of management architecture, and realize enterprise
knowledge integration management. According to the fuzzy VIKOR method, the steps of information classification for small and medi?
um-sized enterprises are designed. The standard method of Bayesian statistics is introduced to obtain the optimal number of topics.
Gibbs sampling algorithm is used to obtain the vector of the overall probability distribution of the hidden layer topic set, so as to realize
the classification management of the information collection system of small and medium-sized enterprises.【Result/conclusion】The
experimental results show that the system has high accuracy and can effectively improve the efficiency of information classification
management and information retrieval recall.【Innovation/limitation】In this paper, LDA model is used to integrate management enter?
prise knowledge, and fuzzy VIKOR method is used to classify and manage enterprise information collection, so as to construct an accu?
rate and efficient system model. However, the system model constructed in this paper has not been used in actual enterprises for feed? back and improvement.