Data sequence prediction method and computing device

A computing device, comprising:

• a processor;
• and
• a memory coupled to the processor and configured to store instructions that, when executed by the processor, cause the computing device to be configured to:
• receive, from a server through a communication interface coupled to the processor, a first plurality of historical data sequences of N objects;
• obtain, based on the first historical data sequences, a similarity distance between every two objects of the N objects and obtain a similarity distance set, wherein the similarity distance represents a first similarity degree of between every two objects, wherein each of the first historical data sequences comprises a plurality of pieces of data arranged according to a preset rule, and wherein N is a positive integer greater than 1;
• divide the N objects into K prediction object classes based on the similarity distance set using a clustering algorithm, wherein K is a positive integer, and wherein K≤N;
• obtain, based on a historical data sequence of an object comprised in a prediction object class of the K prediction object classes, historical common data patterns of the prediction object class;
• train a model based on the historical common data patterns of the prediction object class to obtain a data pattern model;
• predict a future data sequence of the object using the data pattern model;
• and
• send, through the communication interface, the future data sequence to the server.

A computer program product comprising computer-executable instructions for storage on a non-transitory computer-readable medium that, when executed by a processor, cause an information processing apparatus to:

• receive, from a server through a communication interface coupled to the processor, a first plurality of historical data sequences of N objects;
• obtain a similarity distance between every two objects of a plurality of objects and obtain a similarity distance set based on historical data sequences of N objects, wherein the similarity distance represents a first similarity degree between every two objects, wherein each of the historical data sequences comprises a plurality of pieces of data arranged according to a preset rule;
• divide the N objects into a plurality of prediction object classes based on the similarity distance set using a clustering algorithm;
• obtain, based on a historical data sequence of an object comprised in a prediction object class of the prediction object classes, historical common data patterns of the prediction object class;
• train a model based on the historical common data patterns of the prediction object class to obtain a data pattern model;
• predict a future data sequence of the object using the data pattern model;
• and
• send, through the communication interface, the future data sequence to the server.