Assisted analytics
AssigneeDimensional Insight Incorporated
InventorsFrederick A. Powers, James Clark
Life of this patent
Estimating the term…
Abstract
Providing user-controllable visualization of an impact of each of a set of dimensions for a set of data from which at least one outlier value is detected includes computing an outlier boundary for a dimension data value via statistical analysis for data organized as assisted analytics time frame data sets. Further a dimension data value outlier factor may be produced for a detected outlier based on a weighting associated with an assisted analytics time frame that corresponds to the outlier. Dimension outlier factors may be derived therefrom and mapped to a corresponding dimension impact rating value for the user-controllable visualization.
Claims20
A system comprising:
- at least one processor;
- and
- a memory device storing an application that adapts the at least one processor to:
- compute an outlier boundary for a dimension data value by applying statistical analysis to at least a portion of data organized under the dimension data value as assisted analytics time frame data sets;
- detect at least one outlier in the assisted analytics time frame data sets that is outside of the dimension data value outlier boundary;
- produce a dimension data value outlier factor based on a weighting associated with an assisted analytics time frame for each of the at least one outlier;
- generate one or more dimension outlier factors based at least in part on the dimension data value outlier factor derived from the detected at least one outlier;
- and
- map the one or more dimension outlier factors to a corresponding dimension impact rating value, the mapping structured to facilitate user-controllable visualization in an electronic interface of an impact of each of a set of dimensions of the data organized under the dimension data value for a portion of the assisted analytics time frame data sets.
A computer implemented method comprising:
- computing with a processor an outlier boundary for a dimension data value by applying statistical analysis to at least a portion of data organized under the dimension data value as assisted analytics time frame data sets;
- detecting with the processor at least one outlier in the assisted analytics time frame data sets that is outside of the dimension data value outlier boundary;
- producing a dimension data value outlier factor based on a weighting associated with an assisted analytics time frame for each of the at least one outlier;
- generating one or more dimension outlier factors based at least in part on the dimension data value outlier factor derived from the detected at least one outlier;
- and
- mapping with the processor the one or more dimension outlier factors to a corresponding dimension impact rating value, the mapping structured to facilitate user-controllable visualization in an electronic interface of an impact of each of a set of dimensions of the data organized under the dimension data value for a portion of the assisted analytics time frame data sets.
A computer implemented method comprising:
- computing with a processor an outlier boundary for a dimension data value by applying statistical analysis to at least a portion of data organized under the dimension data value as assisted analytics time frame data sets;
- producing a dimension data value outlier factor based on a weighting associated with each assisted analytics time frame that includes an outlier that is outside of the dimension data value outlier boundary;
- generating one or more dimension outlier factors for a dimension that is common to a plurality of dimension data values based at least in part on corresponding dimension data value outlier factors;
- and
- mapping with the processor the one or more dimension outlier factors to a corresponding dimension impact rating value, the mapping structured to facilitate user-controllable visualization in an electronic interface of an impact of each of a set of the dimensions for a portion of the assisted analytics time frame data sets.
Dependencies are inferred from the claim text.