Continuous data optimization method and system based on geographic correlation

Abstract

The invention discloses a continuous data optimization method and system based on geographic correlation. The method comprises the following steps: determining an initialization weight of a data provider by utilizing historical data which is provided by the data provider and contains a certain target object and known truth data of the target object; calculating the relevancy between the target object and other target objects at the current moment; calculating the aggregation truth of the target object at the current moment; and calculating credibility according to the data provided by the provider at the current moment, the accumulated provided data times at the current moment and the aggregation truth of the target object at the current moment, and updating the weight of the data provider according to the credibility. The invention has the advantages that a continuous data optimization method based on geographic correlation is adopted, and the problem that accurate aggregation truth cannot be obtained when the target object does not collect enough data is effectively solved; and the accuracy of the result can be greatly improved under the condition that data provided by a data provider is insufficient, and the requirement of true phase aggregation is met.

Description

technical field [0001] The invention relates to a continuous data optimization method and system based on geographical correlation, belonging to the technical field of artificial intelligence. Background technique [0002] In the information age, with the development of social platforms, there are more and more data providers, and the data they provide are also diverse. However, it is not reliable to obtain the truth from only a single data source because the authoritativeness of the information provided by the data provider is difficult to identify. Therefore, in order to obtain more accurate results, it is usually necessary to aggregate data from multiple data providers to obtain the truth. [0003] However, existing methods of multi-channel information aggregation have shortcomings for applications in some fields. If the traditional truth aggregation method is adopted without considering the geographical correlation of the target object, when only a small amount of data...

AI Technical Summary

Problems solved by technology

[0003] However, the existing methods of multi-channel information aggregation have shortcomings for the application in some fields

If the traditional truth aggregation method is adopted without considering the geographical correlation of the target object, when only a small amount of data can be collected for a target object, the purpose of multi-channel aggregation cannot be achieved, and the final aggregation result will also be inaccurate

[0004] In addition, if only the continuity of the data is considered, when the rate of change of the data is very large, only the aggregation value of the target object at the previous moment is considered as the truth at the current moment, and it is obvious that accurate results cannot be obtained

Images