k-nearest neighbors in uncertain graphs

Abstract

Complex networks, such as biological, social, and communication networks, often entail uncertainty, and thus, can be modeled as probabilistic graphs. Similar to the problem of similarity search in standard graphs, a fundamental problem for probabilistic graphs is to efficiently answer k-nearest neighbor queries (k-NN), which is the problem of computing the k closest nodes to some specific node.In this paper we introduce a framework for processing k-NN queries in probabilistic graphs. We propose novel distance functions that extend well-known graph concepts, such as shortest paths. In order to compute them in probabilistic graphs, we design algorithms based on sampling. During k-NN query processing we efficiently prune the search space using novel techniques.Our experiments indicate that our distance functions outperform previously used alternatives in identifying true neighbors in real-world biological data. We also demonstrate that our algorithms scale for graphs with tens of millions of edges.

On Uncertain Probabilistic Data Modeling

“…Modeling, querying and mining uncertain graphs have become an increasingly important research topic [19][20][21] recently. Probabilistic graphs are a natural model representation in many applications, such as mobile ad-hoc networks, social networks, traffic networks, biological networks, genome databases, medical records, etc.…”

Probabilistic Graphs Using Clustering Algorithm with Efficient Performance

“…Opinion mining task can be transformed into classification task, so machine learning techniques can be used for opinion mining. Machine learning approaches require a corpus containing a wide number of manually tagged.In Protein-Protein Interaction (PPI) networks, the interaction Cluster's rights two proteins is generally established with a probability property due to the limitation of observation methods [2]. In addition, it has been verified that the interaction Cluster's rights proteins A and B can influence the interaction Cluster's rights protein A and another protein C, if A, B and C have some common features.…”

Reliability Evaluation of the Minimum Spanning Tree on Uncertain Graph

“…Each possible subgraph of the uncertain graph is called implicated graph. Their research mainly focus on graph mining [10]- [12], graph queries [13]- [15] and basic graph structure [16], [17].…”

“…E implicated graphs, because each edge provides us with a binary sampling decision. Following the same assumption of the existing uncertain graph models [10], [13], [21], we assume that uncertain variables of different edges are mutually independent. Based on this assumption, the probability of sampling the implicated graph G from the uncertain graph is…”