Cost-effective outbreak detection in networks

Abstract

Given a water distribution network, where should we place sensors to quickly detect contaminants? Or, which blogs should we read to avoid missing important stories? These seemingly different problems share common structure: Outbreak detection can be modeled as selecting nodes (sensor locations, blogs) in a network, in order to detect the spreading of a virus or information as quickly as possible. We present a general methodology for near optimal sensor placement in these and related problems. We demonstrate that many realistic outbreak detection objectives (e.g., detection likelihood, population affected) exhibit the property of "submodularity". We exploit submodularity to develop an efficient algorithm that scales to large problems, achieving near optimal placements, while being 700 times faster than a simple greedy algorithm. We also derive online bounds on the quality of the placements obtained by any algorithm. Our algorithms and bounds also handle cases where nodes (sensor locations, blogs) have different costs. We evaluate our approach on several large real-world problems, including a model of a water distribution network from the EPA, and real blog data. The obtained sensor placements are provably near optimal, providing a constant fraction of the optimal solution. We show that the approach scales, achieving speedups and savings in storage of several orders of magnitude. We also show how the approach leads to deeper insights in both applications, answering multicriteria trade-off, cost-sensitivity and generalization questions.

Scaling Influence Maximization with Network Abstractions

“…The simpler case, where the network topology alone dictates activation spread, has been examined by multiple research groups, seeking to improve on Kempe's early work on greedy approaches for influence maximization [14]. Examples of possible speedups include innovations such as the use of a shortest-path based influence cascade model [16] or a lazy-forward optimization algorithm [19] to reduce the number of evaluations on the influence spread of nodes. Clever heuristics have been used very successfully to speed computation in both the LT model (e.g., the PMIA algorithm [8]) and also the IC model [25].…”

Incremental Sensor Placement Optimization on Water Network

“…The objective of [8,9] is to ensure that the expected impact of a contamination event is within a pre-specified level, and [8] introduced a formulation based on set cover and solved the problem using genetic algorithm while [9] use a MIP based solution. In order to achieve the objective defined in [10], [11][12][13][14][15][16][17][18][19][20][21] adopted multi-objective optimization by different methods such as heuristic, predator-prey model or local search method and [22] used the submodular property to achieve an approximation guarantee.…”

“…Submodularity was widely used in sensor placement optimization [22,25]. But these solutions are mostlybuilt for static water network.…”

Influence Spread Maximization in Social Network

“…For the LT model, we compare our two algorithms (DAGIS and Sampling) with the CELFGreedy [4] algorithm. For the IC model, we compare our algorithm DDH with CELFGreedy [4] and Degree [2] and Distance [2]. We provide the experimental results of different algorithms for two influence cascade models.…”

“…So we can use this idea to update the influence spread of nodes lazily. In [4], Chen etc. present NewGreedyIC algorithm which is based on the greedy algorithm.…”