Team Control Number
For office use only T1 ________________ T2 ________________ T3 ________________ T4 ________________
12218
Problem Chosen
For office use only F1________________F2________________F3________________F4________________
C
2012 Mathematical Contest in Modeling (MCM) Summary Sheet
Type a summary of your results on this page. Do not include the name of your school, advisor, or team members on this page.
Social Network Analysis in Crime Busting
Social network analysis (SNA) has been winning more and more attention in the last twenty years. The method provides an insight into various networks. In this paper, we use SNA and related techniques to analyze crime data and try to get potential criminal gang.
After first investigating the hidden features of a conspiratorial group, we consider one’s closeness to the criminal gang as the main reason in deciding whether a staff member should be regarded as a conspirator. The introduction of cooperation distance metric (CD-metric) combined with the analysis on cooperation factor makes up our model’s major base. Within these definitions, potential conspirators can be worked out by choosing those on the top of the ascending CD-metric list. Highest values of CD-metric can be recognized as being the farthest away from the conspiracy, in other words, they are the most innocent ones. A 12 members’ criminal group is dug out.
Before identifying the leading criminals, it is essential for us to quantify the ability of people’s leadership. Centrality analysis suggests a good way in determining the focal point(s) in a network. We make some amendments of the centrality measures so that it can be applied to a directed graph. What’s more, we combine centrality measures(degree, closeness and betweenness) and figure out the fusion values. We can form a ranking list of people’s ability of leadership then. The leaders in the criminal group are sure to come to surface after comparing with the priority list derived before.
In the refinement of the network model, we appeal to the theories of semantic network analysis and text analysis. On the purpose of making a deeper exploration of identifying the intrinsic character of one topic, we reapply the centrality analysis to data computing right after the construction of a semantic web. Based on text analysis approach, we build up a vector space model and compare the final outputs with ones obtained above, the results are robust.
This model’s application in other areas such as the biomedical domain has been discussed at the end of this paper, excellent performance happens when accessed to enough amounts of data. With a well-organized structure and a wide range of application, this methodology is sure to have a bright future.
Keywords: Social Network Analysis, Centrality Measures, Semantic Web, Text Analysis
ICM PROBLEM PROBLEM C: Modeling for Crime Busting Click the title below to download a ZIP file containing the 2012 ICM Problem. Your ICM submission ...
2012美国大学生数学建模C题,2012C: Modeling for Crime BustingYour organization...Your supervisor's advice to your team is identify the guilty parties so ...
2012美国大学生数学建模C题,2012C: Modeling for Crime BustingYour organization...Your supervisor's advice to your team is identify the guilty parties so ...
热门文档