Method for constructing index relational network

Abstract

The invention provides a method for constructing an index relational network. The method comprises the following steps: 1, collecting the data of all management indexes in the same period of time; 2, calculating the relational degree between two indexes of the management indexes, and outputting index pairs with a strong correlation; 3, testing the causal relationship between the two indexes within the index pairs, and giving the influence direction of relevance of the index pairs according to the causal relationship. According to the hospital management index relational network constructed by means of the method for constructing an index relational network, the index relational network can automatically locate the root index which causes an abnormality when an index is abnormal, the root cause of the abnormality can be traced when the abnormality occurs, and there is no need to manually sort out the cause of the abnormality according to service relationships. In addition, it can be predicted that the other indexes pointed to by the abnormal index may be abnormal according to the index relational network, and timely control and solution of more serious problems caused by the abnormality are conducted before they occur.

Description

technical field [0001] The invention relates to the field of indicator abnormality prediction, in particular to a method for constructing an indicator relationship network. Background technique [0002] In the existing technology, when a certain index in the hospital management index relationship network is abnormal, it is necessary to manually sort out the cause of the abnormality based on the business relationship, which not only needs to rely on the experience of the manager, but also cannot predict in time which other indicators it points to Metrics can misbehave, often leading to more serious problems. Contents of the invention [0003] The present invention provides a method for constructing an index relationship network to solve the problem that in the prior art, it is necessary to manually sort out abnormalities based on business relationships, and it is impossible to predict in time which other indicators pointed to by it may cause item abnormalities, resulting in...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a method for constructing an index relationship network to solve the problem that in the prior art, it is necessary to manually sort out abnormalities based on business relationships, and it is impossible to predict in time which other indicators pointed to by it may cause item abnormalities, resulting in more serious problems. matter of consequence

Images