statistical: Reviewed wang2018privacy

tables/statistical.tex

@@ -82,7 +82,10 @@
       \cite{farokhi2020temporally} & & & & & & (Laplace) & privacy \\ \hdashline
 
       \hyperlink{ma2019real}{\textbf{\emph{RPTR}}} & infinite & streaming & global & $w$-event & linkage & perturbation & differential \\
-      \cite{ma2019real} & & & & & & (Laplace) & privacy \\
+      \cite{ma2019real} & & & & & & (Laplace) & privacy \\ \hdashline
+
+      \hyperlink{wang2018privacy}{\textbf{\emph{DP-PSP}}} & infinite & streaming & global & $w$-event & linkage & perturbation & differential \\
+      \cite{wang2018privacy} & & & & & & (Laplace) & privacy \\
 
       \bottomrule
 

text/bibliography.bib

@@ -849,7 +849,6 @@
   publisher = {Springer}
 }
 
-
 @article{ji2014differential,
   title   = {Differential privacy and machine learning: a survey and review},
   author  = {Ji, Zhanglong and Lipton, Zachary C and Elkan, Charles},
@@ -1405,7 +1404,6 @@
   publisher = {Cambridge university press}
 }
 
-%  new algorithm
 @misc{russell2018fitness,
   title        = {Fitness app {Strava} exposes the location of military bases},
   author       = {Russell, Jon},
@@ -1684,6 +1682,16 @@
   organization = {IEEE}
 }
 
+@article{wang2018privacy,
+  title     = {Privacy-protected statistics publication over social media user trajectory streams},
+  author    = {Wang, Shuo and Sinnott, Richard and Nepal, Surya},
+  journal   = {Future Generation Computer Systems},
+  volume    = {87},
+  pages     = {792--802},
+  year      = {2018},
+  publisher = {Elsevier}
+}
+
 @article{wang2021current,
   title     = {Why current differential privacy schemes are inapplicable for correlated data publishing?},
   author    = {Wang, Hao and Xu, Zhengquan and Jia, Shan and Xia, Ying and Zhang, Xu},

text/related/statistical.tex

@@ -389,3 +389,22 @@ The combination of adaptive sampling and filtering can improve the accuracy when
 The mechanism detects highly frequented map regions and, using a quad-tree, it calculate the each region's privacy weight.
 In their implementation, the authors assume that highly frequented regions tend to be more privacy sensitive, and thus more noise (i.e.,~less privacy budget to invest) needs to be introduced before publicly releasing the users' data falling into these regions.
 The efficiency (both in terms of user privacy and data utility) of the mechanism depends on the number of regions that it divides the map, and therefore the challenge of its optimal division is an interesting future research topic.
+
+% Privacy-protected statistics publication over social media user trajectory streams
+% - statistical
+% - infinite
+% - streaming
+% - linkage
+% - global
+% - w-event
+% - differential privacy
+% - perturbation (Laplace)
+\hypertarget{wang2018privacy}{Wang et al.}~\cite{wang2018privacy} presented \emph{DP-PSP}, an approach for publishing differentially private statistics over infinite streams of trajectory data.
+DP-PSP segments trajectories by taking into account points of interest in road networks.
+A start and end point (anchor) represents a segment and each data point in the trajectory data is calibrated to the nearest anchor.
+This segmentation facilitates a less computationally intensive statistical processing and more efficient privacy budget allocation.
+The authors designed a private $k$ nearest neighbors algorithm by utilizing the exponential mechanism, which uses the Gaussian weighted Euclidean distance for utility function, to generate the connected segments for each segment.
+Thus, at some timestamps, they can predict accurately the upcoming statistics, and therefore save part of the available privacy budget by releasing an approximation instead of perturbing the original data.
+DP-PSP allocates the available privacy budget, in an exponentially decaying fashion, in a sliding window with a user-defined size $w$, satisfying $w$-event-level privacy.
+Statistics over the trajectory combined with Laplacian noise are released in the end of the process by DP-PSP.
+From the implementation, it is not clear how DP-PSP takes into consideration all of the user preferences regarding the size of $w$ while releasing statistics of the data of all of the sample.