Method, apparatus, and device for determining network anomaly behavior, and readable storage medium

1. A method for determining a network anomaly behavior, comprising:

acquiring network data traffic in a target network, and determining each data packet contained in the network data traffic as a node of a complete graph, to construct a data complete graph;

optimizing the data complete graph into a minimum spanning tree by using Kruskal algorithm based on weights of all edges in the data complete graph, and determining each node of the minimum spanning tree as a cluster center, wherein for each edge in the data complete graph, a length of the edge is determined as the weight of the edge;

classifying all data packets contained in the network data traffic based on cluster centers, to acquire data element sets corresponding to the cluster centers respectively; and

determining an anomaly behavior in the target network based on a type of each of the data element sets, wherein the type of the data element set is determined by a type of a network attack contained in the network data traffic.

2. The method for determining a network anomaly behavior according to claim 1 , wherein after acquiring the data element sets corresponding to the cluster centers respectively, the method further comprises:

calculating, for each of the data element sets, a ratio of an intra-cluster distance to an inter-cluster distance of the data element set, and determining whether ratios of the intra-cluster distance to the inter-cluster distance of all the data element sets are less than a preset first threshold;

performing steps of determining the anomaly behavior in the target network based on a type of each of the data element sets, in response to determining that the ratios of the intra-cluster distance to the inter-cluster distance of all the data element sets are less than the preset first threshold.

3. The method for determining a network anomaly behavior according to claim 2 , wherein when there is a ratio greater than or equal to the preset first threshold, the method further comprises:

updating each cluster center according to an objective formula, and performing steps of classifying all data packets contained in the network data traffic based on cluster centers to acquire data element sets corresponding to the cluster centers respectively, wherein

the objective formula is: G′=G ( X,Y,Z ), X =( x 1 m 1 +x 2 m 2 + . . . +x i m i )/ M,Y =( y 1 m 1 +y 2 m 2 + . . . +y i m i )/ M,Z =( z 1 m 1 +z 2 m 2 + . . . +z i m i )/ M,

wherein G′ represents an updated cluster center, M represents, before updating, a sum of distances between the cluster center and each data element in the data element set corresponding to the cluster center, and X, Y and Z respectively represent coordinates of the updated cluster center in a three-dimensional space; and

m 1 , m 2 . . . m i respectively represent, before updating, a distance between each data element in the data element set and the cluster center corresponding to the data element set, x 1 , x 2 . . . x i respectively represent, before updating, a X coordinate value of each data element in the data element set mapped to the three-dimensional space, y 1 , y 2 . . . y i respectively represent, before updating, a Y coordinate value of each data element in the data element set mapped to the three-dimensional space, and z 1 , z 2 . . . z i respectively represent, before updating, a Z coordinate value of each data element in the data element set mapped to the three-dimensional space.

4. The method for determining a network anomaly behavior according to claim 3 , wherein before determining an anomaly behavior in the target network based on a type of each of the data element sets, the method further comprises:

calculating a function value of an objective function, and determining whether a difference between the calculated function value and an initial function value is less than a preset second threshold; and

performing steps of determining an anomaly behavior in the target network based on a type of each of the data element sets, in response to determining that the difference between the calculated function value and the initial function value is less than the preset second threshold.

5. The method for determining a network anomaly behavior according to claim 4 , wherein when the difference between the calculated function value and the initial function value is greater than or equal to the preset second threshold, the method further comprises:

determining the calculated function value as the initial function value, and performing steps of updating each cluster center according to the objective formula.

6. The method for determining a network anomaly behavior according to claim 1 , wherein the classifying all data packets contained in the network data traffic based on cluster centers, to acquire data element sets corresponding to the cluster centers respectively comprises:

acquiring, for each of the cluster centers, the data element set corresponding to the cluster center according to the following steps:

removing a target cluster center from the data packets contained in the network data traffic, to acquire a target data packet set;

for each data element in the target data packet set,

calculating a distance between the data element and the target cluster center, and determining whether the distance is less than a preset radius value; and

dividing the data element into a data element set corresponding to the target cluster center, in response to determining that the distance is less than the preset radius value.

7. The method for determining a network anomaly behavior according to claim 6 , further comprising:

calculating a first cosine similarity between a target data element and any one of the data elements in the data element set corresponding to the target cluster center, and a second cosine similarity between the target data element and any one of data elements in a data element set corresponding to an object cluster center, in a case that a first distance between the target data element and the target cluster center, and a second distance between the target data element and the object cluster center are both less than the radius value;

determining whether the first cosine similarity is greater than the second cosine similarity;

dividing the target data element into the data element set corresponding to the target cluster center, if it is determined that the first cosine similarity is greater than the second cosine similarity; and

dividing the target data element into the data element set corresponding to the object cluster center, if it is determined that the first cosine similarity is less than or equal to the second cosine similarity.

8. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 1 .

9. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 2 .

10. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 3 .

11. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 4 .

12. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 5 .

13. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 6 .

14. A device for determining a network anomaly behavior, comprising

a memory, configured to store a computer program; and

a processor, configured to execute the computer program to implement steps of the method for determining a network anomaly behavior according to claim 7 .

15. A non-transitory readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement steps of the method for determining a network anomaly behavior according to claim 1 .

16. A non-transitory readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement steps of the method for determining a network anomaly behavior according to claim 2 .

17. A non-transitory readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement steps of the method for determining a network anomaly behavior according to claim 3 .

18. A non-transitory readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement steps of the method for determining a network anomaly behavior according to claim 4 .

19. A non-transitory readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement steps of the method for determining a network anomaly behavior according to claim 5 .