Method for classifying data using clustering and classification algorithm supervised
View Patent ↗A method for classifying data involves receiving a set of training data from a physical process such as a computer network ( 20 ). The training data has attribute data and target data. The target data has a class label associated with the attribute data. Dummy clusters are derived from centroid coordinates of the training data associated with the class label ( 22 ). Distance measures are determined between the training data and a plurality of clusters which include the dummy clusters ( 24 ). Real clusters are created in the plurality of clusters if the training data is closest to a dummy cluster or a cluster having a class label different than the class label associated with the training data ( 26 ). A closest match between data to be classified and the plurality of clusters is identified ( 28 ) and the data is classified as the class label of the closest match from the plurality of clusters ( 30 ).
1. A computer-implemented method for classification of data, comprising:
providing first data from a physical process, the first data including a class label associated with attributes of the first data;
deriving a temporary artificial cluster from centroid coordinates of the first data associated with the class label, wherein the temporary artificial cluster has a class-label different than the class label associated with the attributes of the first data;
determining distance measures between the first data and a plurality of clusters which include the temporary artificial cluster;
creating a real cluster in the plurality of clusters if the first data is closest to the temporary artificial cluster;
identifying a closest match between second data and the plurality of clusters; and
classifying the second data based on a class label of the closest match from the plurality of clusters.
2. A computer-implemented method for classification of data, comprising:
providing first data from a physical process, the first data including a class label associated with attributes of the first data;
deriving a dummy cluster from centroid coordinates of the first data associated with the class label, wherein deriving the dummy cluster includes,
(a) creating first and second dummy clusters associated with first and second class labels, respectively,
(b) initializing the first dummy cluster with the centroid coordinates of the first data if the class label of the first data matches the first class label, and
(c) initializing the second dummy cluster with the centroid coordinates of the first data if the class label of the first data matches the second class label;
determining distance measures between the first data and a plurality of clusters which include the dummy cluster;
creating a real cluster in the plurality of clusters if the first data is closest to the dummy cluster;
identifying a closest match between second data and the plurality of clusters;
classifying the second data based on a class label of the closest match from the plurality of clusters;
wherein deriving a dummy cluster includes updating the centroid coordinates of the first dummy cluster for each first data having a class label that matches the first class label; and
updating the centroid coordinates of the second dummy cluster for each first data having a class label that matches the second class label.
3. The method of claim 2 , wherein determining distance measures between the first data and a plurality of clusters includes using one of the group of a weighted Canberra distance, a weighted Euclidean distance, and a weighted Chi-squared distance for the distance measure.
4. The method of claim 2 , wherein determining distance measures between the first data and a plurality of clusters includes:
calculating sample variance of the first data;
calculating sample covariance of the first data;
calculating sample mean of the first data; and
calculating correlation coefficient from the sample variance, sample covariance, and sample mean of the first data.
5. The method of claim 2 , further including creating a real cluster in the plurality of clusters if the first data is closest to a cluster having a class label different than the class label associated with the first data.
6. The method of claim 2 , wherein identifying a closest match between second data and the plurality of clusters includes calculating a distance measure from one of the group of a weighted Canberra distance, a weighted Euclidean distance, and a weighted Chi-squared distance and using the closest distance measure as the closest match.
7. A computer-implemented method of classifying first data from a physical process, comprising:
providing first data which includes a class label associated with attributes of the first data;
deriving an artificial cluster from centroid coordinates of the first data associated with the class label, wherein the artificial cluster has a class label different than the class label associated with the attributes of the first data to cause the first data to separate from the artificial cluster during the classification process;
determining distance measures between the first data and a plurality of clusters which include the artificial cluster;
creating a real cluster in the plurality of cluster if the first data is closest to the artificial cluster;
identifying a closest match between second data and the plurality of clusters; and
classifying the second data based on a class label of the closest match from the plurality of clusters.
8. The method of claim 7 , wherein identifying a closest match between second data and the plurality of clusters includes calculating a distance measure from one of the group of a weighted Canberra distance, a weighted Euclidean distance, and a weighted Chi-squared distance and using tile closest distance measure as the closest match.
9. The method of claim 7 , wherein deriving the artificial cluster includes:
creating first and second artificial clusters associated with first and second class labels, respectively;
initializing the first artificial cluster with the centroid coordinates of the first data if the class label of the first data matches the first class label; and
initializing the second artificial cluster with the centroid coordinates of the first data if the class label of the first data matches the second class label.
10. The method of claim 9 , wherein deriving the artificial cluster further includes:
updating the centroid coordinates of the first artificial cluster for each first data having a class label that matches the first class label; and
updating the centroid coordinates of the second artificial cluster for each first data having a class label that matches the second class label.
11. The method of claim 7 , further including creating a real cluster in the plurality of clusters if the first data is closest to a cluster having a class label different than the class label associated with the first data.
12. A computer-implemented method of classifying first data from a physical process, comprising:
providing first data which includes a class label associated with attributes of the first data;
deriving a dummy cluster from centroid coordinates of the first data associated with the class label, wherein the dummy cluster has a class label different than the class label associated with the attributes of the first data;
determining distance measures between the first data and a plurality of clusters which include the dummy cluster;
creating a real cluster in the plurality of clusters if the first data is closest to a cluster having a class label different than the class label associated with the first data;
identifying a closest match between second data and the plurality of clusters; and
classifying the second data based on a class label of the closest match from the plurality of clusters.
13. The method of claim 12 , wherein identifying a closest match between second data and the plurality of clusters includes calculating a distance measure from one of the group of a weighted Canberra distance, a weighted Euclidean distance, and a weighted Chi-squared distance and using the closest distance measure as the closest match.
14. The method of claim 12 , wherein deriving a dummy cluster includes:
creating first and second dummy clusters associated with first and second class labels, respectively;
initializing the first dummy cluster with the centroid coordinates of the first data if the class label of the first data matches the first class label; and
initializing the second dummy cluster with the centroid coordinates of the first data if the class label of the first data matches the second class label.
15. The method of claim 14 , wherein deriving a dummy cluster further includes:
updating the centroid coordinates of the first dummy cluster for each first data having a class label that matches the first class label; and
updating the centroid coordinates of the second dummy cluster for each first data having a class label that matches the second class label.
16. The method of claim 12 , further including creating a real cluster in the plurality of clusters if the first data is closest to the dummy cluster.
17. A digital storage medium encoded with a computer program which classifies data, the computer program comprising:
first instructions for providing first data from a physical process, the first data including a class label associated with attributes of the first data;
second instructions for deriving an artificial cluster from centroid coordinates of the first data associated with the class label, wherein the artificial cluster has a class label different than the class label associated with the attributes of the first data to cause the first data to separate from the artificial cluster during the classification process;
third instructions for determining distance measures between the first data and a plurality of clusters which include the artificial cluster;
fourth instructions for creating a real cluster in the plurality of clusters if the first data is closest to the artificial cluster;
fifth instructions for identifying a closest match between second data and the plurality of clusters; and
sixth instructions for classifying the second data based on a class label of the closest match from the plurality of clusters.
18. The digital storage medium of claim 17 , wherein the second instructions further include:
fifth instructions for creating first and second artificial clusters associated with first and second class labels, respectively;
sixth instructions for initializing the first artificial cluster with the centroid coordinates of the first data if the class label of the first data matches the first class label; and
seventh instructions for initializing the second artificial cluster with the centroid coordinates of the first data if the class label of the first data matches the second class label.
19. The digital storage medium of claim 18 , wherein the second instructions further include:
eighth instructions for updating the centroid coordinates of the first artificial cluster for each first data having a class label that matches the first class label; and
ninth instructions for updating the centroid coordinates of the second artificial cluster for each first data having a class label that matches the second class label.
20. The digital storage medium of claim 17 , wherein the second instructions further include creating a real cluster in the plurality of clusters if the first data is closest to a cluster having a class label different then the class label associated with the first data.