L.H. Cox. Suppression methodology and statistical disclosure analysis. Journal of the American Statistical Association, 75:377--385, 1980.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... include restricting the size of query sets (i.e. the tuples that satisfy a single query) 22] restricting the size of overlaps [18] between query sets, detecting inferences by auditing all queries asked by a specific user [12, 10, 26, 6] suppressing sensitive data in released statistical tables [13], grouping tuples and treating each group as a single tuple [11, 32] Perturbation based techniques add noise to source data or outputs [35, 5, 34] Other aspects of inference problem include the inference caused by arithmetic constraints [8, 7] inferring approximate values instead of exact ....

L.H. Cox. Suppression methodology and statistical disclosure control. Journal of American Statistical Association, 75(370):377--385, 1980.

.... techniques [20] include restricting the size of a query set (i.e. the entities that satisfy a single query) overlap control [18] in query sets, auditing all queries in order to determine when inferences are possible [12, 9, 24, 27, 6] suppressing sensitive data in released statistical tables [13], partitioning data into mutually exclusive partitions [10, 11] while In the settings of this paper, each variable can have either one value or infinitely many values. restricting each query set to range over at most one partition. Perturbation based technique includes adding noise to source ....

L.H. Cox. Suppression methodology and statistical disclosure control. Journal of American Statistic Association, 75(370):377--385, 1980. 36

.... have been studied (in the statistical database literature) which include restricting the size of a qe set (i.e. the entities that satisfy a single query) 17,13] controlling the overlap of query sets [15] suppressing sensitive data cells in a released table of statistics (i.e. query results) [9], partitioning data into mutually exclusive chunks and restricting each query set to some un divided data chunks [6,7] and (closer to our concerns in this paper) auditing all queries in order to determine whether inference is possible [8,5,20,23] For controlling statistical inference, some ....

L.H. Cox. Suppression methodology and statistical disclosure control. Journal of American Statistic Association, 75(370):377-385, 1980. 554

.... techniques [19] include restricting the size of a query set (i.e. the entities that satisfy a single query) overlap control [17] in query sets, auditing all queries in order to determine when inferences are possible [11, 8, 23, 25] suppressing sensitive data in a released statistical tables [12], partitioning data into mutually exclusive partition [9, 10] and restricting each query set to range over at most one partition. Perturbation based technique includes adding noise to source data [30] outputs [5, 26] database structure [28] or size of query sets (by sampling data to answer ....

L.H. Cox. Suppression methodology and statistical disclosure control. Journal of American Statistic Association, 75(370):377--385, 1980.

.... been studied (in the statistical database literature) which include restricting the size of a query set (i.e. the entities that satisfy a single query) 17, 13] controlling the overlap of query sets [15] suppressing sensitive data cells in a released table of statistics (i.e. query results) [9], par titioning data into mutually exclusive chunks and restricting each query set to some undivided data chunks [6, 7] and (closer to our concerns in this paper) auditing all queries in order to determine whether inference is possible [8, 5, 20, 23] For controlling statistical inference, some ....

L.H. Cox. Suppression methodology and statistical disclosure control. Journal of American Statistic Association, 75(370):377-385, 1980.

....perturbation. The query restriction family includes restricting the size of query results [13] 18] controlling the overlap among successive queries [14] keeping audit trails of all answered queries and constantly checking for possible compromises [8] suppression of data cells of small size [9], and clustering entities into mutually exclusive atomic populations [61] The perturbation family includes swapping values between records [12] replacing the original database by a sample from the same distribution [33] 42] adding noise to the values in the database [52] 57] adding noise to ....

L. Cox. Suppression methodology and statistical disclosure control. J. Am. Stat. Assoc., 75(370):377--395, April 1980.

.... This combinatorial optimization problem is known as the (secondary) Cell Suppression Problem (CSP) and belongs to the class of the strongly NP hard problems; see, e.g. Kelly, Golden and Assad [13] Geurts [8] Kao [11] Previous works on CSP mainly concentrate on heuristics; see, e.g. Cox [3, 4], Kelly, Golden and Assad [13] and Carvalho, Dellaert and Os orio [6] Kelly [12] proposed a mixed integer linear programming formulation involving a huge number of variables and constraints (for instance, the formulation involves more than 20,000,000 variables and 30,000,000 constraints for a ....

....J J J J J J J J J J J J J J J R Figure 2: Network representation of CSP. 2 A new integer linear programming model The system of linear equations (1) 3) can be represented in a natural way by the network D = V; A) of Figure 2; see, e.g. Cox [3, 4] and Kelly, Golden and Assad [13] The network has a node r i for each row i = 0; NR, and a node c j for each column j = 0; NC. We have an arc for each entry of the matrix, namely: an arc (r i ; c j ) associated with each entry a ij with i 6= 0 and j 6= 0, an arc (c 0 ; r i ) ....

L.H. Cox, \Suppression Methodology and Statistical Disclosure Control", Journal of the American Statistical Association, 75 (1980) 377-385.

....of CSP exists, which guarantees an ecient (i.e. polynomial time) performance for all possible input instances. Previous work on CSP from the literature mainly concentrate on 2 dimensional tables with marginals. Heuristic solution procedures have been proposed by several authors, including Cox [1, 2], Sande [16] Kelly, Golden and Assad [13] and Carvalho, Dellaert and Os orio [4] Kelly [12] proposed a mixed integer linear programming formulation involving a huge number of variables and constraints. Geurts [9] re ned this model, and reported computational experiences on small size instances, ....

L.H. Cox, \Suppression Methodology and Statistical Disclosure Control", Journal of the American Statistical Association, 75 (1980) 377-385.

....the need of considering external bounds on the cell values when computing realistic protection intervals. A more detailed description of the problem is as follows. Given a set of sensitive cells PS (primary suppressions) along with the required protection levels, de ned e.g. according to [1], the statistical oce aims at nding a set of complementary suppressions protecting all the sensitive cells against the attacker, and such that the 3 loss of information associated with the suppressed entries is minimized. More speci cally, let T = 1; n be the index set of a given table ....

....(CSP) and belongs to the class of the strongly NP hard problems; see, e.g. Kelly, Golden and Assad [12] Geurts [8] Kao [10] Previous work on CSP mainly concentrate on 2 dimensional tables with marginals. Heuristic solution procedures have been proposed by several authors, including Cox [1, 2], Sande [16] Kelly, Golden and Assad [12] and Carvalho, Dellaert and Os orio [4] Kelly [11] proposed a mixed integer linear programming formulation involving a huge number of variables and constraints. Geurts [8] re ned this model, and reported computational experiences on small size instances, ....

[Article contains additional citation context not shown here]

L.H. Cox, \Suppression Methodology and Statistical Disclosure Control", Journal of the American Statistical Association, 75 (1980) 377-385. 11

.... is known as the (secondary) Cell Suppression Problem (CSP) and belongs to the class of the strongly NP hard problems; see, e.g. Kelly, Golden and Assad [10] Geurts [6] Kao [8] Previous works on CSP mainly concentrate on heuristics for 2 dimensional tables with marginals; see, e.g. Cox [1, 2], Sande [14] Kelly, Golden and Assad [10] and Carvalho, Dellaert and Os orio [4] Robertson [13] extended the algorithm of Sande [14] to 3 dimensional tables. Kelly [9] proposed a mixed integer linear programming formulation involving a huge number of variables and constraints. Geurts [6] re ....

....case in which table entries y i are linked by a generic system of linear equations Ay = b. In the case of k dimensional tables with marginals, A is a f0; 1; 1g matrix and b = 0. Moreover, for the case k = 2 the linear system can be represented in a natural way as a network; see, e.g. Cox [1, 2] and Kelly, Golden and Assad [10] Unfortunately, this nice structure is not preserved for k 3, unless the table decomposes into a set of independent 2 dimensional subtables. We say that a vector [y i ] de nes a feasible table whenever Ay = b, and a i lb i y i a i ub i holds for all i. ....

L.H. Cox, \Suppression Methodology and Statistical Disclosure Control", Journal of the American Statistical Association, 75 (1980) 377-385.

.... This combinatorial optimization problem is known as the (secondary) Cell Suppression Problem (CSP) and belongs to the class of the strongly NP hard problems; see, e.g. Kelly, Golden and Assad [14] Geurts [9] Kao [12] 3 Previous works on CSP mainly concentrate on heuristics; see, e.g. Cox [3, 4], Kelly, Golden and Assad [14] and Carvalho, Dellaert and Os orio [6] Kelly [13] proposed a mixed integer linear programming formulation involving a huge number of variables and constraints (for instance, the formulation involves more than 20,000,000 variables and 30,000,000 constraints for a ....

....cells, by far the largest ever tried. Finally, Section 6 draws some conclusions and addresses future directions of work. 2 A new integer linear programming model The system of linear equations (1) 3) can be represented in a natural way by a complete bipartite network D = V; A) see, e.g. Cox [3, 4] and Kelly, Golden and Assad [14] The network has a node r i for each row i = 0; NR, and a node c j for each column j = 0; NC. We have an arc for each entry of the matrix, namely: an arc (r i ; c j ) associated with each entry a ij with i 6= 0 and j 6= 0, an arc (c 0 ; r i ) ....

L.H. Cox, \Suppression Methodology and Statistical Disclosure Control", Journal of the American Statistical Association, 75 (1980) 377-385.

.... family includes restricting the size of query result (e.g. Fel72] DDS79] controlling the overlap amongst successive queries (e.g. DJL79] keeping audit trail of all answered queries and constantly checking for possible compromise (e.g. CO82] suppression of data cells of small size (e.g. Cox80] and clustering entities into mutually exclusive atomic populations (e.g. YC77] The perturbation family includes swapping values between records (e.g. Den82] replacing the original database by a sample from the same distribution (e.g. LST83] LCL85] Rei84] adding noise to the values ....

L.H. Cox. Suppression methodology and statistical disclosure control. J. Am. Stat. Assoc., 75(370):377--395, April 1980.

....data, or to make specific data views accessible to interested subgroups of individuals within the collecting organization. In order to maintain confidentiality of the respondent, data providers may suppress cell values if release would allow unauthorized inference of sensitive information [3, 7]. Inferential disclosure occurs when two or more data tables, taken together, enable a user to identify information pertaining to individual respondents even though none of the data, taken by itself, is a direct disclosure [12] This can occur when a linear combination of released cells results in ....

Lawrence H. Cox. Suppression methodology and statistical disclosure control. Journal of the American Statistical Association, 75(370):377-- 385, June 1980.

No context found.

L.H. Cox. Suppression methodology and statistical disclosure analysis. Journal of the American Statistical Association, 75:377--385, 1980.

No context found.

Cox, L.H.: Suppression methodology and statistical disclosure control. J. American Statistical Association 75 (1980) 377-385.

No context found.

Cox, L. H. (1980), \Suppression Methodology and Statistical Disclosure Control," Journal of the American Statistical Association, 75, 377-385.

No context found.

Cox, L. H. (1980), \Suppression Methodology and Statistical Disclosure Control," Journal of the American Statistical Association, 75, 377-385.

No context found.

L. H. Cox, ""Suppression methodology and statistical disclosure control,"" Journal of th merican Statistical Association, 75, 370, 377-385, June 1980. e c

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC