Abstract:

In the big data era, the data which operator can collect and process is quickly growing. At the same time, due to the high dimension of the data, it is very difficult to extract and process information from them. This makes the waste of the computing capability of operators, even may lead to the decrease of performance due to the growth of the data dimension. The main reason of this phenomenon is that not all dimensions are interested with the information. Therefore, it is necessary to select the appropriate dimensions according to the data characteristics. Based on this, a data processing method based on projection pursuit was proposed. The result demonstrates the effectiveness of the scheme.

Key words: network management, dimension reduction, projection pursuit