Systems and methods for generating, analyzing, and storing data snippets
AssigneePalantir Technologies Inc.
InventorsPeter Wilczynski, Kendra Knittel, Andrew Elder, Anand Gupta, Jessica Headrick
Life of this patent
Estimating the term…
Abstract
Systems and methods are provided for analyzing data snippets. One or more snippets can be associated with a data object of an enterprise data platform. The one or more snippets can be organized based on metadata information associated with the one or more snippets. The organized snippets can be analyzed to determine an activity relating to an entity depicted in the one or more snippets.
Claims19
A system comprising:
- one or more processors;
- and
- a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform:
- associating, with a first data object, first portions of a first data file having a first data file type, the first data file being associated with an entity, wherein the associating of the first data object with the first portions is based on a probability of the entity matching a known entity or a different instance of the entity, wherein the first data object represents the entity;
- associating, with a second data object, second portions of a second data file having a second data file type;
- and
- aligning a first portion of the first portions with a second portion of the second portions based on metadata from the first portion and from the second portion.
A computer-implemented method, the method comprising:
- associating, with a first data object, first portions of a first data file having a first data file type, the first data file being associated with an entity;
- associating, with a second data object, second portions of a second data file having a second data file type;
- aligning a first portion of the first portions with a second portion of the second portions based on metadata from the first portion and from the second portion;
- and
- representing the first portions in higher resolution compared to a different portion of the first data file by retaining an original resolution of the first portions while decreasing an original resolution of the different portion.
A non-transitory computer readable medium comprising instructions that, when executed, cause one or more processors to perform:
- associating, with a first data object, first portions of a first data file having a first data file type, the first data file being associated with an entity;
- associating, with a second data object, second portions of a second data file having a second data file type;
- aligning a first portion of the first portions with a second portion of the second portions based on metadata from the first portion and from the second portion;
- and
- representing the first portions in higher resolution compared to a different portion of the first data file by retaining an original resolution of the first portions while decreasing an original resolution of the different portion.
Dependencies are inferred from the claim text.