Path building in cellular partitioning networks

Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Butler

Boedihardjo

et al. 2012

Intelligence analysts grapple with many challenges, chief among them is the need for software support in storytelling, i.e., automatically 'connecting the dots' between disparate entities (e.g., people, organizations) in an effort to form hypotheses and suggest nonobvious relationships. We present a system to automatically construct stories in entity networks that can help form directed chains of relationships, with support for co-referencing, evidence marshaling, and imposing syntactic constraints on the story generation process. A novel optimization technique based on concept lattice mining enables us to rapidly construct stories on massive datasets. Using several public domain datasets, we illustrate how our approach overcomes many limitations of current systems and enables the analyst to efficiently narrow down to hypotheses of interest and reason about alternative explanations.

Section: Related Projectsmentioning

confidence: 99%

Storytelling in entity networks to support intelligence analysts

Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Butler

Boedihardjo

et al. 2012

“…However, there are some works in the domain of graph theory that explicitly concentrate on finding a suitable path between two nodes within the same network. For example, Brassard and Gecsei (1980) made the attempt to solve a ''connecting the dots'' problem for cellular partitioning networks. They proposed a method of path building to efficiently find couplings in electrical networks.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Faloutsos et al (2004) capture the relationship between two nodes of a large graph based on the electricity flow from the given start and end vertices. Although the works of Brassard and Gecsei (1980) and Faloutsos et al (2004) use aspects of graph theory to find connections between nodes in the same network, to the best of our knowledge this is the first time such an algorithmic framework is used to find connections between multiple graph objects. Additionally, we have made the first attempt to find connections between pairs of cell signaling pathways in the STKE dataset based on their graph structure.…”

Section: Literature Reviewmentioning

confidence: 99%

Narratives in the Network: Interactive Methods for Mining Cell Signaling Networks

Journal of Computational Biology

Akbar

Polys

2012

In this article, we describe our work on graph mining as applied to the cellular signaling pathways in the Signal Transduction Knowledge Environment (STKE). We present new algorithms and a graphical tool that can help biologists discover relationships between pathways by looking at structural overlaps within the database. We address the problem of determining pathway relationships by using two data mining approaches: clustering and storytelling. In the first approach, our tool brings similar pathways to the same cluster, and in the second, our tool determines intermediate overlapping pathways that can lead biologists to new hypotheses and experiments regarding relationships between the pathways. We formulate the problem of discovering pathway relationships as a subgraph discovery problem and propose a new technique called Subgraph-Extension Generation (SEG), which outperforms the traditional Frequent Subgraph Discovery (FSG) approach by magnitudes. Our tool provides an interface to compare these two approaches with a variety of similarity measures and clustering techniques as well as in terms of computational performance measures such as runtime and memory consumption.

“…The connecting building task has been used in a variety of contexts: entity networks [5,8], image collections [6], cellular networks [2,7], social networks [4], and document collections [8,9,11]. All these research efforts focus on finding connections between objects that are apparently disjoint.…”

mentioning

confidence: 99%

How to Understand Connections Based on Big Data: From Cliques to Flexible Granules

Jalal-Kamali

Kreinovich

2014

Studies in Big Data

One of the main objectives of science and engineering is to predict the future state of the world -and to come up with actions which will lead to the most favorable outcome. To be able to do that, we need to have a quantitative model describing how the values of the desired quantities change -and for that, we need to know which factors influence this change. Usually, these factors are selected by using traditional statistical techniques, but with the current drastic increase in the amount of available data -known as the advent of big data -the traditional techniques are no longer feasible. A successful semi-heuristic method has been proposed to detect true connections in the presence of big data. However, this method has its limitations. The first limitation is that this method is heuristic -its main justifications are common sense and the fact that in several practical problems, this method was reasonably successful. The second limitation is that this heuristic method is based on using "crisp" granules (clusters), while in reality, the corresponding granules are flexible ("fuzzy"). In this chapter, we explain how the known semi-heuristic method can be justified in statistical terms, and we also show how the ideas behind this justification enable us to improve the known method by taking granule flexibility into account.