Ping An Insurance (Group) Company of China has been granted a patent for a network anomaly data detection method. The method involves receiving access request data, searching historical data, performing word segmentation, obtaining a weight matrix, inputting it into a detection model, and judging the presence of anomaly data. The patent also includes the ability to correct the initial detection model. GlobalData’s report on Ping An Insurance (Group) Company of China gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Ping An Insurance (Group) Company of China, digital lending was a key innovation area identified from patents. Ping An Insurance (Group) Company of China's grant share as of June 2023 was 1%. Grant share is based on the ratio of number of grants to total number of patents.
Network anomaly data detection method
A recently granted patent (Publication Number: US11683330B2) describes a network anomaly data detection method. The method involves receiving access request data from a client and searching historical access request data corresponding to a user session identifier. A header character string of the access request data is acquired and subjected to word segmentation processing to obtain a word segmentation set. A word segmentation weight matrix is then obtained based on the historical access request data and the word segmentation set. This weight matrix is inputted into an anomaly data detection model to determine the probability of data anomalies. The method also includes judging whether anomaly data exists in the header character string and correcting the initial anomaly data detection model accordingly.
The patent also includes additional steps for the method. After receiving the access request data, the method involves reading a user session identifier and comparing the user fingerprint identifier in the identifier with the user fingerprint identifier of the last received access request data. If inconsistent, a new request session identifier is generated and transmitted to the client when response data is returned. Another step involves acquiring the session length of the current session when the user fingerprint identifier is inconsistent with the last access request. If the session length does not exceed a preset threshold, historical access request data corresponding to the user session identifier is searched. If the session length exceeds the threshold, a new request session identifier is generated and transmitted to the client when response data is returned.
The patent also describes a computer equipment embodiment, which includes a memory and one or more processors. The computer equipment executes the steps of the network anomaly data detection method, including receiving access request data, searching historical access request data, acquiring a header character string, performing word segmentation processing, obtaining a word segmentation weight matrix, inputting the matrix into an anomaly data detection model, judging for anomaly data, and correcting the initial anomaly data detection model.
Overall, this patent presents a method and computer equipment for detecting network anomalies by analyzing access request data and historical data. The method involves word segmentation processing and the use of an anomaly data detection model to determine the probability of data anomalies. The additional steps of comparing user fingerprint identifiers and generating new request session identifiers enhance the accuracy of the anomaly detection process.
To know more about GlobalData’s detailed insights on Ping An Insurance (Group) Company of China, buy the report here.