China UnionPay had eight patents in artificial intelligence during Q3 2023. China UnionPay Co Ltd has developed a method and apparatus for real-time data monitoring using machine learning. The method involves training a multi-layer predictor with historical indicator data, which includes different types of predictors. The trained predictor can then output predicted values for future indicator data. Alarm thresholds are calculated based on the predicted values and historical prediction errors, and an alarm is triggered if the actual value of the future indicator data exceeds the threshold. This approach improves the accuracy of the alarm thresholds and adapts them to changing indicator data, reducing the need for manual configuration and minimizing missed and false alarms. GlobalData’s report on China UnionPay gives a 360-degreee view of the company including its patenting strategy. Buy the report here.

China UnionPay grant share with artificial intelligence as a theme is 50% in Q3 2023. Grant share is based on the ratio of number of grants to total number of patents.

Recent Patents

Application: Data real-time monitoring method and apparatus based on machine learning (Patent ID: US20230222362A1)

The patent filed by China UnionPay Co Ltd describes a method and apparatus for real-time data monitoring based on machine learning. The method involves training a multi-layer predictor using historical indicator data and different types of predictors. The trained predictor is then used to output predicted values of future indicator data. Alarm thresholds are calculated based on the predicted values and historical prediction errors, and an alarm is triggered when the actual value of the future indicator data exceeds the threshold. This approach improves the accuracy of the alarm thresholds and adapts them to changing indicator data, eliminating the need for manual configuration and reducing missed and false alarms.

The method includes training a plurality of predictors of different types on the actual values of historical indicator data. The trained predictors are used to predict actual values of the historical indicator data. At least one Nth-layer predictor is then trained on target predicted values of the historical indicator data and a hybrid data set. The multi-layer predictor is constructed with a plurality of trained first-layer predictors to the trained Nth-layer predictor when the layer number N reaches a predetermined value.

The apparatus for real-time data monitoring includes a multi-layer predictor training module, a predicted value calculating module, an alarm threshold calculating module, and an alarm triggering module. The multi-layer predictor training module trains the multi-layer predictor using historical indicator data and different types of predictors. The predicted value calculating module outputs predicted values of future indicator data using the trained multi-layer predictor. The alarm threshold calculating module calculates alarm thresholds based on the predicted values and historical prediction errors. The alarm triggering module triggers an alarm when the actual value of the future indicator data exceeds the corresponding threshold.

The apparatus also includes various sub-modules for training the predictors, splitting the historical indicator data, calculating weights, and calculating learning rates. These sub-modules contribute to the overall functionality of the apparatus for real-time data monitoring.

Overall, the method and apparatus described in the patent aim to improve the accuracy of real-time data monitoring by utilizing machine learning techniques and adaptive alarm thresholds.

To know more about GlobalData’s detailed insights on China UnionPay, buy the report here.

Premium Insights

From

The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.

GlobalData

GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.