Multi-level Multi-source Visual Ana lysis of Government Text Data

• Grant number: 14/09599-5
• Support Opportunities: Scholarships abroad - Research
• Start date: August 01, 2015
• End date: January 31, 2016
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Rosane Minghim
• Grantee: Rosane Minghim
• Host Investigator: Evangelos Milios
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Dalhousie University, Halifax, Canada
• Associated research grant: 11/22749-8 - Challenges in Exploratory Visualization of Multidimensional Data: Paradigms, Scalability and Applications, AP.TEM

Abstract

In recent years, researchers in various areas of knowledge related with data intensive computing (such as machine learning, data m ining, vis ualiz ation and applic ations) have b een directing efforts towards dealingwith larger and more diverse data sources. Additionally, there has b een a renewed motivation to integrate research in those fields, sinc e they are complementary in the manner they approach data analysis. T hispro jec t prop oses inte grating mining and visualization techniques in order to explore and understand textual data in novel ways, seeking to tackle current problem s in Visual Text Analytics and focusing attention on a strategic applications. Current research in visual text mining has not yet succeeded inhandling multiple scales of information or to effectively asso ciate two or more sets from distinct sourcestreating the same sub jects . For instance, when p olitical discourse is carried out on a particular subject and that same sub ject is discussed in the so cial media or in the news, they are not easily asso ciated by theanalyst. In this pro ject we prop ose to devise new strategies to tackle the problem of handling asso ciation between sets of documents from different sources. By choosing a flexible representation - in out case - a network representation - we can employ the large b o dy of work in graphs and networks to p erform visual analysis, mining, and partitioning of textual data sets, as well as content match. This same framework also lends itself to multi-level representation through network partitioning, which is a valid strategy tosupp ort vis ual exploration of larger data sets. The pro ject entails the three following activities: 1 - handle data sets of separate sources (such as government debates and news) as separate and associate networks, by means of a graph represe ntation as well as a selection of algorithms to partition the space;2- adaptation of multidimensional visualization techniques to explore such data sets; 3 - development of strategies to link the two or more related data sets and incorp orate that into the visual mining set up.Our baseline data set during this first year of joint work is that of the Canadian parliament. The data collection stage is already well under way, develop ed during the last year under collaboration with the group at Dalhousie. With the collab oration with other colleagues and up on my return, our advances will be applied also to the Brazilian Congress. Target users vary from interested citize ns to news professionaland the government policy devisers . The main idea is to integrate the large text mining exp erienceof the host institution with that of visualization from the prop onent and her group, to contribute instudying the issues of how government structure is reflected in the media and in the general p opulation.