|Support type:||Scholarships in Brazil - Post-Doctorate|
|Effective date (Start):||August 01, 2012|
|Effective date (End):||July 31, 2014|
|Field of knowledge:||Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques|
|Principal Investigator:||Zhao Liang|
|Grantee:||João Roberto Bertini Junior|
|Home Institution:||Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil|
Machine learning methods are essential in the resolution of data mining related problems. Applications like clustering, classification and information retrieval, would not be properly addressed by other approaches. In the context of machine learning, only recently have graph-based methods been considered as a solution to data mining related problems. Graph-based methods consist in a powerful form for data representation and abstraction, which provides, among others advantages, representing topological relations, visualizing structures, representing groups of data with distinct formats, as well as, supplying alternative measures to characterize data. Nevertheless, if on one hand applications such as data clustering have been widely studied; on the other hand, applications which consider labeled data, such as classification, have not been received the same attention in the literature. Considering the advantages provided by graph representation and the success of such approach when addressing similar applications, which dismiss the presence of labeled data; this project aims at investigate graph-based algorithms for data classification purposes. The project addresses four related topics, differentiating regarding to (a) the kind of data; (b) the amount of labeled data and (c) the future distribution of data.