Event detection in activity networks
read more
Citations
When Engagement Meets Similarity: Efficient (k,r)-Core Computation on Social Networks
Dense Subgraph Discovery: KDD 2015 tutorial
A Nearly-Linear Time Framework for Graph-Structured Sparsity
A nearly-linear time framework for graph-structured sparsity
Scaling Distance Labeling on Small-World Networks
References
LOF: identifying density-based local outliers
Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming
A spatial scan statistic
SDPT3 — A Matlab software package for semidefinite programming, Version 1.3
A General Approximation Technique for Constrained Forest Problems
Related Papers (5)
Non-parametric scan statistics for event detection and forecasting in heterogeneous social media graphs
Graph based anomaly detection and description: a survey
Frequently Asked Questions (12)
Q2. What are the future works mentioned in the paper "Event detection in activity networks" ?
Their work opens many interesting directions for future research.
Q3. How do the authors plant events to the datasets?
The authors plant events to the datasets by setting to 1 the values of nodes that occur within the event and to 0 the volues of nodes outside the planted event.
Q4. What is the way to select the parameter?
One way of selecting the parameter is to execute the algorithms with different values of λ, plot the Pareto curve for those values and choose a value that yields the desired trade-off between spanned weight and distance.
Q5. How do the authors improve the quality of the events discovered by their algorithms?
To improve the quality of the events discovered by their algorithms the authors set the weight of a node to be |x −m|, where x is the current activity level of a station and m is the typical activity level of the station.
Q6. What is the objective function of the prize-collection Steiner-tree problem?
The objective is to select a subset of the nodes such that the cost of the tree to connect them plus the prizes of the nodes not in the tree is minimized.
Q7. What can be taken into account when monitoring traffic?
For instance, periodicity phenomena can be taken into account; for example, when monitoring the traffic activity in a city, the reference values for weekend 8am traffic are much lower than that of a week day.
Q8. What is the algorithm for PD?
The currently best algorithm [4] improves the approximation ratio to (2 − ), which is more important rather from a theoretical perspective.
Q9. What are the two simple greedy algorithms?
The authors show that the optimization function is submodular and the authors provide two simple greedy algorithms with provable approximation guarantees.
Q10. What is the main objective function of the Steiner-tree problem?
in a work related to the Steiner-tree problem formulation presented in this paper, Seufert et al. [26]consider the problem of finding a subtree with k nodes such that the total node weight is maximized.
Q11. What is the shifted version of the EventTree problem?
EventAllPairs+ PROBLEMIn this section the authors present their algorihtms for the EventAllPairs+ problem, starting with efficient greedy approaches and continuing with a slower but more effective approach based on the MaxCut problem.
Q12. What are the problems the authors define with respect to the activity value monitored?
Events are defined with respect to the activity value monitored: the authors are interested in finding compact subareas where the traffic measurements are abnormal, the pollution levels are unusually high, and so on.