Back to results
US 2014/0180998 A1Published application
Efficient searching of stationary datasets
AssigneeThe Regents of the University of California
InventorsMajid Sarrafzadeh, Jonathan S. Woodbridge
Life of this patent
Estimating the term…
Abstract
A saliency function is computed to indicate the saliency of each of a plurality of data points in a dataset. For each local maximum in the saliency function, a segment of the dataset is inserted into an index.
Claims24
A non-transitory computer-readable storage medium, comprising executable instructions to:
- receive digital data related to a stationary dataset;
- compute a saliency function indicating the saliency of each of a plurality of data points in the digital data;
- determine a set of at least one local maxima in the saliency function;
- and
- for each maximum in the set of at least one local maxima, insert a segment of the digital data into an index, the segment corresponding to a portion of the digital data including the data point associated with the local maximum.
2 more elements
A system, comprising:
- an input configured to receive time series signals;
- a conversion unit configured to convert a time series signal received at the input into a dataset of digital values;
- and
- a processing unit configured to
- determine a saliency function for the dataset, indicating the saliency of a plurality of data points in the dataset;
- identify local maxima in the saliency function;
- and
- create an index of portions of the dataset, wherein each portion of the dataset includes a data point associated with a local maximum of the saliency function.
5 more elements
A method, comprising:
- in a computing device,
- calculating a saliency function for a plurality of data points in a stationary dataset, the saliency function indicating for each of the plurality of data points the probability that the data point would be in its current state given the states of neighbor data points;
- determining from the saliency function a set of low-probability data points each of which has a low probability of occurrence;
- and
- for each data point in the set of low-probability data points, inserting into an index a portion of the stationary dataset including that data point.
2 more elements
A medical device comprising:
- an input to receive digital data relating to a stationary data set;
- means for computing a saliency function indicating the saliency of each of a plurality of data points in the digital data;
- means for determining a set of at least one local maxima in the saliency function;
- and
- means for inserting, for each maximum in the set of at least one local maxima, a segment of the digital data into an index, the segment corresponding to a portion of the digital data including the data point associated with the local maximum.
2 more elements
Dependencies are inferred from the claim text.