https://flore.unifi.it/bitstream/2158/1087565/1/2015%20PhLifeRev.pdf

Statistical physics of crime: a review.

Social physics

The Accidental Guerrilla: Fighting Small Wars in the Midst of a Big One

Spatial clustering deals with the unsupervised grouping of places into clusters and finds important applications in urban planning and marketing. Current spatial clustering models disregard information about the people who are related to the clustered places. In this paper, we show how the density-based clustering paradigm can be extended to apply on places which are visited by users of a geo-social network. Our model considers both spatial information and the social relationships between users who visit the clustered places. After formally defining the model and the distance measure it relies on, we present efficient algorithms for its implementation, based on spatial indexing. We evaluate the effectiveness of our model via a case study on real data; in addition, we design two quantitative measures, called social entropy and community score to evaluate the quality of the discovered clusters. The results show that geo-social clusters have special properties and cannot be found by applying simple spatial clustering approaches. The efficiency of our index-based implementation is also evaluated experimentally.

/pdf/density-based-place-clustering-in-geo-social-networks-54n0wlun5y.pdf

Density-based place clustering in geo-social networks

Given a social network of users with non-uniform selection cost, and a fixed budget, which nodes should be chosen for initial activation to maximize the influence in the network, such that the total selection cost does not exceed the budget? This problem is known as the Budgeted Influence Maximization Problem or BIM Problem, in short. Though the problem appears to be realistic one, there are very few studies available in the literature. In this paper, we propose ComBIM, a community-based solution approach for solving the BIM problem. The proposed methodology has four steps: first, community detection to understand the inherent structure of the social network, second, budget distribution to divide the total budget among the communities based on the number of nodes and their selection costs, third, seed selection for influence maximization and finally, budget transfer, in which unutilized budget of one community is transferred to another community. We implement the proposed methodology with three publicly available social network datasets. We have compared the obtained results with that of other methodologies from the literature and observe that ComBIM can achieve the better influence spread, while taking reasonable time for seed set selection.

ComBIM: A community-based solution approach for the Budgeted Influence Maximization Problem

In this article we identify social communities among gang members in the Hollenbeck policing district in Los Angeles, based on sparse observations of a combination of social interactions and geographic locations of the individuals. This information, coming from Los Angeles Police Department (LAPD) Field Interview cards, is used to construct a similarity graph for the individuals. We use spectral clustering to identify clusters in the graph, corresponding to communities in Hollenbeck, and compare these with the LAPD's knowledge of the individuals' gang membership. We discuss different ways of encoding the geosocial information using a graph structure and the influence on the resulting clusterings. Finally we analyze the robustness of this technique with respect to noisy and incomplete data, thereby providing suggestions about the relative importance of quantity versus quality of collected data.

Community Detection Using Spectral Clustering on Sparse Geosocial Data

In this article we identify social communities among gang members in the Hollenbeck policing district in Los Angeles, based on sparse observations of a combination of social interactions and geographic locations of the individuals. This information, coming from LAPD Field Interview cards, is used to construct a similarity graph for the individuals. We use spectral clustering to identify clusters in the graph, corresponding to communities in Hollenbeck, and compare these with the LAPD's knowledge of the individuals' gang membership. We discuss different ways of encoding the geosocial information using a graph structure and the influence on the resulting clusterings. Finally we analyze the robustness of this technique with respect to noisy and incomplete data, thereby providing suggestions about the relative importance of quantity versus quality of collected data.

Community detection using spectral clustering on sparse geosocial data

In Hollenbeck, a gang-dominated region of Los Angeles, gang activity has been monitored by the LAPD. One manner has been in the form of non-criminal stops. We propose a spectral clustering algorithm to predict gang aliation from the information obtained from these stops. Despite the sparseness of data, the combination of geography and the record of individuals involved in a stop reveals the underlying gang structure and higher order structures (collaboration or rivalry). Having obtained positive results on the real Hollenbeck data, we then develop methods to create gang simulations that will further our understanding of this algorithm and the situation in Hollenbeck.

/pdf/social-network-clustering-an-analysis-of-gang-networks-1zxm5vu961.pdf

Raymond Ahn

Papers

Community Detection Using Spectral Clustering on Sparse Geosocial Data

Community detection using spectral clustering on sparse geosocial data

Social Network Clustering: An Analysis of Gang Networks