Determining journalist risk of a dataset using population equivalence class distribution estimation

Abstract

Methods and systems to de-identify a longitudinal dataset of personal records based on journalistic risk computed from a sample set of the personal records, including determining a similarity distribution of the sample set based on quasi-identifiers of the respective personal records, converting the similarity distribution of the sample set to an equivalence class distribution, and computing journalistic risk based on the equivalence distribution. In an embodiment, multiple similarity measures are determined for a personal record based on comparisons with multiple combinations of other personal records of the sample set, and an average of the multiple similarity measures is rounded. In an embodiment, similarity measures are determined for a subset of the sample set and, for each similarity measure, the number of records having the similarity measure is projected to the subset of personal records. Journalistic risk may be computed for multiple types of attacks.

1 . A computer-implemented method, comprising:
 receiving a data set from one or more databases, wherein the received data set represents patient records of individuals;   determining a first distribution from the received data set, wherein the first distribution reflects a level of resemblance shared among the individuals in the received data set, and wherein one or more values are assigned to the patient records;   producing a second distribution based on the first distribution, wherein the second distribution represents combinations of one or more individuals in the received data set who share at least one trait;   identifying a subset of the second distribution representing a population distribution;   identifying one or more risks associated with the subset of the second distribution; and   de-identifying the received data set based on the identified one or more risks.   
     
     
         2 . The method of  claim 1 , further comprising:
 1. determining a probability of the subset of the second distribution being representative of the population distribution.   
     
     
         3 . The method of  claim 1 , wherein producing the second distribution based on the first distribution includes identifying multiple combinations of the patient records. 
     
     
         4 . The method of  claim 1 , wherein the one or more risks are identified based on the level of resemblance among the individuals in the first distribution. 
     
     
         5 . The method of  claim 1 , wherein the received data set includes patient records of individuals having different medical conditions. 
     
     
         6 . The method of  claim 1 , wherein the second distribution is produced based on the first distribution by assigning a value to each individual in the first distribution reflecting the level of resemblance. 
     
     
         7 . The method of  claim 1 , wherein
 identifying the one or more risks includes performing one or more measurements.   
     
     
         8 . A computer program product comprising a tangible storage medium encoded with processor-readable instructions that, when executed by one or more processors, enable the computer program product to:
 receive a data set from one or more databases, wherein the received data set represents patient records of individuals;   determine a first distribution from the received data set, wherein the first distribution reflects a level of resemblance shared among the individuals in the received data set, and wherein one or more values are assigned to the patient records;   produce a second distribution based on the first distribution, wherein the second distribution represents combinations of one or more individuals in the received data set who share at least one trait;   identify a subset of the second distribution representing a population distribution;   identify one or more risks associated with the subset of the second distribution; and
 de-identify the received data set based on the identified one or more risks. 
   
     
     
         9 . The computer program product of  claim 8 , wherein multiple combinations of the patient records are identified in the received data set to produce the second distribution based on the first distribution. 
     
     
         10 . The computer program product of  claim 8 , wherein the one or more risks are identified by inverting at least one patient class size. 
     
     
         11 . The computer program product of  claim 8 , wherein the second distribution is produced by assigning a value to each individual in the first distribution reflecting the level of resemblance. 
     
     
         12 . The computer program product of  claim 8 , wherein the one or more risks are identified by inverting one or more patient populations. 
     
     
         13 . The computer program product of  claim 8 , wherein the one or more risks are determined by dividing a proportion of patient records of the second distribution by a size of the second distribution. 
     
     
         14 . The computer program product of  claim 8 , wherein a value is assigned to each set of data within the received data set before the second distribution is produced. 
     
     
         15 . A computer system connected to a network, the system comprising:
 a memory configured to store instructions;   one or more processors configured to execute the instructions to:
 receive a data set from one or more databases, wherein the received data set represents patient records of individuals; 
 determine a first distribution from the received data set, wherein the first distribution reflects a level of resemblance shared among the individuals in the received data set, and wherein one or more values are assigned to the patient records; 
 produce a second distribution based on the first distribution, wherein the second distribution represents combinations of one or more individuals in the received data set who share at least one trait; 
 identify a subset of the second distribution representing a population distribution; 
 identify one or more risks associated with the subset of the second distribution; and 
 de-identify the received data set based on the identified one or more risks. 
   
     
     
         16 . The system of  claim 15 , wherein the second distribution is produced by assigning a value to each individual within the first distribution reflecting the level of resemblance. 
     
     
         17 . The system of  claim 15 , wherein the one or more risks are identified by inverting the second distribution. 
     
     
         18 . The system of  claim 15 , wherein the one or more risks are determined based on an estimation of the second distribution. 
     
     
         19 . The system of  claim 15 , wherein the second distribution is produced by assigning values to each of the patient records within the first distribution. 
     
     
         20 . The system of  claim 15 , wherein multiple combinations of the patient records are identified within the received data set.