Method and device for detecting clusters amongst a plurality of data arranged in orders
AssigneeSiemens Aktiengesellschaft
InventorsClemens Otte, Rudolf Sollacher, Volker Tresp
Life of this patent
Estimating the term…
Abstract
After initial clusters having only one component are formed, a conditional probability P(Ci|C′k) is determined for the cluster Ci being included in an order on condition that cluster C′k is included in the order. If P(Ci|C′k) is greater than a first threshold value S1, a new cluster Cn having all the components of clusters Ci, C′k is formed and the operations are repeated until no new clusters are formed.
Claims15
A method for detection of clusters amongst data arranged in orders in a computer, at least some of the orders being correlated to contain components including at least one of services and products, at least some of which appear on a plurality of occasions, comprising:
- initially forming clusters, each containing only one component with each cluster containing a different component than all other clusters;
- determining a conditional probability P(Ci|C′k) for a first cluster Ci included in an order on condition that a second cluster C′k is included in the order, across all instances of i and for all instances of k, where i and k run through all of the clusters in the orders;
- adding to the clusters by forming each new cluster Cn containing the components of existing clusters Ci, C′k for which the conditional probability P(Ci|C′k) is greater than a first threshold value;
- and
- repeating said determining and adding until no new cluster is formed.
At least one computer-readable medium storing instructions that when executed control a computer to perform a method for detection of clusters amongst data arranged in orders, at least some of the orders being correlated to contain components including at least one of services and products, at least some of which appear on a plurality of occasions, said method comprising:
- initially forming clusters, each containing only one component with each cluster containing a different component than all other clusters;
- determining a conditional probability P(Ci|C′k) for a first cluster Ci included in an order on condition that a second cluster C′k is included in the order, across all instances of i and for all instances of k, where i and k run through all of the clusters in the orders;
- adding to the clusters by forming each new cluster Cn containing the components of existing clusters Ci, C′k for which the conditional probability P(Ci|C′k) is greater than a first threshold value;
- and
- repeating said determining and adding until no new cluster is formed.
A control system for planning and controlling treatments, investigations and care procedures in a hospital, comprising:
- a user network;
- a database storing data arranged in orders, at least some of the orders being correlated to contain components including at least one of services and products, at least some of which appear on a plurality of occasions;
- at least one computer, connected to said database via said user network, executing a program initially forming clusters, each containing only one component with each cluster containing a different component than all other clusters;
- determining a conditional probability P(Ci|C′k) for a first cluster Ci included in an order on condition that a second cluster C′k is included in the order, across all instances of i and for all instances of k, where i and k run through all of the clusters in the orders;
- adding to the clusters by forming each new cluster Cn containing the components of existing clusters Ci, C′k for which the conditional probability P(Ci|C′k) is greater than a first threshold value;
- and repeating the determining and adding until no new cluster is formed;
- and
- at least one output device to output information representing the existing clusters when the repeating ends.
Dependencies are inferred from the claim text.