Research on Requirements Recommendation Method for High-end Equipment Based on Generative Adversarial Networks

High-end equipment plays an extremely important role in the competition of big countries because of its particularity. In response to the problem of fewer high-end equipment entities and a lack of sample data for requirements, this paper proposes a requirements recommendation method for high-end equipment based on generative adversarial networks (GAN). Firstly, the parametric requirement performance index is modeled and analyzed, and the requirement matrix expression method is designed. Secondly, the requirement matrix is used to train the GAN and generate new reasonable requirements, so as to enhance the requirement dataset. Then, K-means algorithm is used to cluster the requirement data and generate a variety of requirement combination schemes to recommend users. Finally, an example is given.