Systems, methods and apparatus for identifying links among interactional digital data
AssigneeCQuotient, Inc.
InventorsBharath K. Krishnan, Vishwamitra S. Ramakrishnan
Life of this patent
Estimating the term…
Abstract
The invention provides in some aspects methods of digital data processor-based analysis of digital data that represent interactions to identify distinct individuals and/or the entities with which they are affiliated (e.g., households, businesses, social or other groups) involved in those interactions. The methods can be employed, for example, to analyze digital data representing retail purchase, marketing and visitor interactions for tracking and/or reporting purposes.
Claims26
A method of digital data processor-based analysis of digital data that represent interactions to identify distinct individuals and/or the entities with which they are affiliated, the method comprising the steps of:
- A) with a digital data processing system, grouping data representing a plurality of interactions into tentative groups (“data blobs”) based on one or more predesignated attributes, and
- B) with the digital data processing system, analyzing the digital data to identify and to newly designate, as attributes that are alone and/or in combination with each other and/or the predesignated attributes indicative of distinct individuals and/or the entities with which they are affiliated, one or more other attributes (i) whose quantitative usage among the data blobs does not in fact exceed a first threshold measure, and (ii) which are of a type for which a quantitative measure of reliability does not in fact exceed a second threshold measure in any data blobs in which the one or more other attributes appear,
- C) with the digital data processing system, re-grouping interactions represented by the digital data into refined groups based on (i) one or more of those newly designated attributes and (ii) zero or more predesignated attributes,
- D) where the quantitative usage of an attribute among the data blobs is any of a count and an index of a number of data blobs in which that attribute appears, and where the quantitative measure of reliability of the type of a given attribute in a given data blob is a ratio, in that given data blob, of a count of distinct attributes of that type in that blob to a count of distinct combinations of the predesignated attributes in that blob.
A non-transitory computer readable medium containing computer-executable instructions that when executed perform the following steps:
- A) with a digital data processing system, grouping a plurality of interactions represented by the digital data into tentative groups (“data blobs”) based on one or more predesignated attributes,
- B) with the digital data processing system, analyzing the digital data to identify and to newly designate, as attributes that are —alone and/or in combination with each other and/or the predesignated attributes —indicative of distinct individuals and/or the entities with which they are affiliated, one or more other attributes (i) whose quantitative usage among the data blobs does not in fact exceed a first threshold measure, and (ii) which are of a type for which a quantitative measure of reliability does not in fact exceed a second threshold measure in any data blobs in which the one or more other attributes appear,
- C) with the digital data processing system, re-grouping interactions represented by the digital data into refined groups based on (i) one or more of those newly designated attributes and (ii) zero or more predesignated attributes, and
- D) where the quantitative usage of an attribute among the data blobs is any of a count and an index of a number of data blobs in which that attribute appears, and where the quantitative measure of reliability of the type of a given attribute in a given data blob is a ratio, in that given data blob, of a count of distinct attributes of that type in that blob to a count of distinct combinations of the predesignated attributes in that blob.
A digital data processing system for analysis of digital data that represent interactions to identify distinct individuals and/or the entities with which they are affiliated, the system comprising a hardware digital data processor executing:
- A) first functionality that groups data representing a plurality of interactions into tentative groups (“data blobs”) based on one or more predesignated attributes, and
- B) second functionality that is coupled with the first functionality and that analyzes the digital data to identify and to newly designate, as attributes that are—alone and/or in combination with each other and/or the predesignated attributes—indicative of distinct individuals and/or the entities with which they are affiliated, one or more other attributes (i) whose quantitative usage among the data blobs does not exceed a first threshold measure, and (ii) which are of a type for which a quantitative measure of reliability does not exceed a second threshold measure in any data blobs in which the one or more other attribute appear,
- C) with third functionality that is coupled to the second functionality, re-grouping interactions represented by the digital data into refined groups based on (i) one or more of those newly designated attributes and (ii) zero or more predesignated attributes, and
- D) where the quantitative usage of an attribute among the data blobs is any of a count and an index of a number of data blobs in which that attribute appears, and where the quantitative measure of reliability of the type of a given attribute in a given data blob is a ratio, in that given data blob, of a count of distinct attributes of that type in that blob to a count of distinct combinations of the predesignated attributes in that blob.
Dependencies are inferred from the claim text.