Fast clustering with sparse data
AssigneeMicrosoft Corporation
InventorD. Maxwell Chickering
Life of this patent
Estimating the term…
Abstract
Efficient data modeling utilizing sparse representation of a data set. In one embodiment, a computer-implemented method such that a data set is first input. The data set has a plurality of records. Each record has at least one attribute, where each attribute has a default value. The method stores a sparse representation of each record, such that the value of each attribute of the record is stored only if the value of the attribute varies from the default value. A data model is then generated, utilizing the sparse representation, and the model is output. The generation of the data model in one embodiment is in accordance with the Expectation Maximization (EM) algorithm.
Claims27
A computer-implemented method that facilitates fast clustering of a data set defined for a plurality of attributes, comprising:
- inputting the data set, the data set having a plurality of records each having a value for each attribute, wherein the value for each attribute is selected from a plurality of possible values, one value from the plurality of possible values is designated a default value for the attribute;
- storing a sparse representation of each record, including storing the value of the attribute of the record only upon determining that the attribute of the record has a value varying from the default value;
- clustering each record using the sparse representation of the record;
- generating a data model of the data set, utilizing the sparse representation of each record;
- and,
- outputting the data model.
A machine-readable medium having processor instructions stored thereon for execution by a processor to case performance of a method that facilitates fast clustering of a data set defined for a plurality of attributes, comprising:
- inputting the data set, the data set having a plurality of records each having a value for each attribute, wherein the value assigned to each attribute is selected from a plurality of possible values, one value from the plurality of possible values is designated a default value;
- storing a sparse representation of each record, including storing the value of the attribute of the record only upon determining that the attribute of the record has a value varying from the default value;
- clustering each record using the sparse representation of the record;
- generating a data model of the data set, utilizing the sparse representation of each record;
- and,
- outputting the data model.
A computerized system comprising:
- a sparse representation of each record of a plurality of records of a data set, each record having a plurality of attributes, each record having a value for each attribute, wherein the value for each attribute is selected from a plurality of possible values, one value from the plurality of possible values is designated a default value, such that the sparse representation includes the value of the attribute of the record only upon the attribute of the record having a value varying from the default value;
- a data model of the data set, based on clusters of each record using the sparse representation of each record;
- and,
- a model generator to generate the data model from the clustered sparse representation.
Dependencies are inferred from the claim text.