The Application of Cloud Computing and Support Vector Machine in the Construction of Entrepreneurial Model

This paper studies the resource integration service of enterprise innovation team. This paper constructs a parallel support vector machine model based on multi-layer SVM, and develops an application system based on cloud platform innovation and entrepreneurship services. Then this paper studies the time series minimum optimization method based on large-scale data. The MapReduce technique is used to partition the massive training set and construct multi-layer support vector machine. Samples are collected around the optimal surface of each support vector machine. Experiments show that the proposed method is much faster in both time and recognition accuracy than the single or conventional parallel search methods. All the unsuccessful “sub-satisfactory” personnel will conduct the next round of “seeking talents”. After several rounds of screening, a fixed entrepreneurial team match was formed. This algorithm is the “delay receive” operation rule, which makes the algorithm more fast, practical and stable. It can effectively improve the resource allocation effect of the system.