Method of referencing data via edge definition
AssigneeInternational Business Machines Corporation
InventorsTravis M. Drucker, Hoa T. Tran
Life of this patent
Estimating the term…
Abstract
Methods, systems, and articles of manufacture for referencing a selection of data including a plurality of individual data points (e.g., database cells) are provided. Rather than store a reference to each individual data point, only sufficient information necessary to describe a bounding edge of the data selection (an “edge definition”) may be stored, thus significantly reducing the amount of storage. Depending on the application, edge definitions may be used to describe two-dimensional or higher order selections of data.
Claims25
A computer-implemented method for referencing a plurality of data points, from a collection of data, comprising:
- creating an annotation associated with the plurality of data points;
- creating an edge definition for the plurality of data points comprising information which defines at least two edges that bind the plurality of data points wherein the edge definition comprises a fewer number of data points than the plurality of data points;
- storing the annotation;
- storing the edge definition in association with the annotation in a manner allowing retrieval of the annotation on the basis of the edge definition for a specified set of plurality of data points;
- receiving a request for annotations for a second selection of data;
- retrieving the annotation created for the first selection of data upon determining that the second selection of data is contained, at least partially, within the first selection of data, based on the edge definition for the first selection of data;
- and
- returning the retrieved annotation.
A computer-readable storage medium containing a program which, when executed by a processor, performs operations comprising:
- receiving a first selection of data comprising a plurality of data points and spanning at least two columns and at least two rows of an at least two-dimensional collection of data;
- creating an edge definition for the first selection of data which defines a horizontal edge spanning the at least two columns and a vertical edge spanning the at least two rows wherein the edge definition comprises a fewer number of data points than the first selection of data;
- storing the edge definition in association with an annotation created for the first selection of data, wherein the edge definition is stored in association with the annotation in a manner allowing retrieval of the annotation on the basis of the edge definition for the first selection of data;
- wherein the storing comprises:
- creating an index for the first selection of data;
- and
- storing the index with the edge definition in the edge definition table;
- receiving a request for annotations for a second selection of data;
- retrieving the annotation created for the first selection of data upon determining that the second selection of data is contained, at least partially, within the first selection of data, based on the edge definition for the first selection of data;
- wherein retrieving the annotation is done using the index created for the first selection of data;
- and
- returning the retrieved annotation.
An annotation systems comprising:
- an annotation database;
- an edge definition table;
- and
- an executable component configured to:
- create an edge definition for a selection of data comprising a plurality of data points, the edge definition comprising data points which define one or more bounding edges of the selection of data wherein the edge definition comprises a fewer number of data points than the selection of data,
- create an index for the selection of data,
- store the edge definition and index for the selection of data in the edge definition table, and store, in the annotation database, an annotation record comprising an annotation for the selection of data and the index, whereby annotation records contained in the annotation database are retrievable, for a specified set of data points of the selection of data, on the basis of the edge definition.
Dependencies are inferred from the claim text.