Reason analysis method for sudden increase of allocation and preemption work orders based on k-means clustering algorithm

Reason analysis method for sudden increase of allocation and preemption work orders based on k-means clustering algorithm

  • CN 105,608,496 B
  • Filed: 11/09/2015
  • Issued: 07/27/2021
  • Est. Priority Date: 11/09/2015
  • Status: Active Grant
First Claim
Patent Images

1. A rapid work order allocation and preemption reason analysis method based on a k-means clustering algorithm is characterized by comprising the following steps:

  • (1) establishing a keyword matrix for allocating and preempting the work order and the power failure information, namely;

    the power failure information key words are used as the transverse axis of the matrix, the allocation and preemption work order is used as the longitudinal axis of the coordinate axis, and whether each row in the allocation and preemption work order contains the key words appearing in the transverse axis is marked to form a matrix;

    (2) clustering and grouping the matrixes by a k-means clustering algorithm;

    (3) researching keywords contained in each group of the emergency work orders by using the clustered emergency work orders;

    (4) making a semantic network by using the keyword list of each group;

    the method comprises the following specific steps;

    (41) extracting each group of key words as nodes;

    (42) determining common keywords of each group, and taking the common keywords as common nodes;

    (43) carrying out layering operation on each node, wherein the layering level of the common node is higher than that of other nodes;

    the higher the common frequency is, the higher the hierarchy level is;

    (5) through the steps, a structure diagram based on the key word nodes is obtained, and in the diagram, the influence on the rapid increase of the work order is larger the more the nodes are, so that the reason of the rapid increase of the work order is analyzed.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×