Extraction of Spatio-Temporal Data for Social Networks
It is often possible to understand group change over time through examining social network data in a spatial and temporal context. Providing that context via text analysis requires identifying locations and associating them with people. Our GeoRef algorit
- PDF / 759,873 Bytes
- 22 Pages / 439.36 x 666.15 pts Page_size
- 20 Downloads / 175 Views
Extraction of Spatio-Temporal Data for Social Networks Judith Gelernter, Dong Cao, and Kathleen M. Carley
Abstract It is often possible to understand group change over time through examining social network data in a spatial and temporal context. Providing that context via text analysis requires identifying locations and associating them with people. Our GeoRef algorithm too automatically does this person-to-place mapping. It involves the identification of location, and uses syntactic proximity of words in the text to link location to person’s name. We describe an application using the algorithm based upon data from the Sudan Tribune divided into three periods in 2006 for the Darfur crisis. Contributions of this paper are (1) techniques to mine for location from text (2) techniques to mine for social network edges (associations between location and person), (3) spatio-temporal maps made from mined data, and (4) social network analysis based on mined data.
15.1 Introduction Texts can help us answer the question “who is where?” Automatically identifying person–location pairs in texts requires understanding the text. This paper describes how to construct an artificially intelligent algorithm that “understands” which words are places with the help of an authoritative place list called a gazetteer. The algorithm associates entity-place pairs with one another as the basis of a two-mode, people-by-location network. Social network analysis uses advanced mathematical techniques and statistical analysis to examine the relationships among group members. These members, or “entities,” are represented by nodes, and the relationships among the nodes are
J. Gelernter () D. Cao K.M. Carley School of Computer Science, Carnegie-Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA e-mail: [email protected]; [email protected]; [email protected] T. Özyer et al. (eds.), The Influence of Technology on Social Network Analysis and Mining, Lecture Notes in Social Networks 6, DOI 10.1007/978-3-7091-1346-2__15, © Springer-Verlag Wien 2013
351
352
J. Gelernter et al.
represented by links (also called edges), that in a diagram are shown as lines between nodes. Nodes may be people, organizations, locations, events, resources, topics, etc. Identifying locations. Location is particularly valuable in analyzing certain kinds of networks. For example, in epidemiology, geographic context proves more important than personal contact in understanding the spread of disease [1]. In disaster response, location of events and how they change over time can allow relief efforts to be coordinated efficiently [2]. In crime investigation as well as prevention, location is used to spot patterns and learn where to enforce preventive measures [3]. Identifying locations in a text is a complex problem. Named Entity Recognition typically includes identifying names of locations as well as people and organizations [4]. Named Entity Recognition can reach almost 80 % accuracy [5], but automatically identifying location accurately can be ha
Data Loading...
