-
About Us+
-
Products+
-
Solution+
-
Cases+
-
AI Lab+
-
Investor+
-
Contact+
-
Join Us +
-
-
Trial+
About Us+
Products+
Solution+
Cases+
AI Lab+
Investor+
Contact+
Join Us +
Trial+
This paper mainly introduces a semi supervised algorithm SVDD: this algorithm is an extension of traditional s Ⅵ m, which can be used for outliers (singular value) detection and classification of extremely unbalanced data. The algorithm constructs a sphere as the decision boundary according to the characteristics of the marked samples. The attributes of the remaining unmarked samples are judged by using the sphere. If the unmarked sample can fall in the ball, it will be judged as positive sample, otherwise it is negative sample. In the field of financial risk control, we often encounter the data with mark on one sample and the other without mark. We can build the model by using the marked sample, and then use the model to identify whether the remaining unmarked samples have risks
Download