GO-005

2015-08-10

2025-01-31

Rozwój metod selekcji zmiennych

dr hab. Witold Rudnicki

Grant will be devoted to development of new AI algorithms for knowledge discovery and model building, that are possible due to high computational power of GPUs. It will be concentrated on algorithmic side and tested using biomedical data. The algorithms are general and can be applied to any problem described with millions of features. The ideas originated from problems in modern molecular biology, which generates datasets with thousands and even millions of features. Finding relevant features is crucial for building explanatory models. Their huge number combined with relatively small number of objects and weak signal hidden behind inevitable noise constitutes really hard problem. We have developed the feature selection scheme based on exhaustive search in two and more dimensions and applied it for analysis of Alzheimer Disease. We will extend this approach for general data, moreover we will develop also new classification algorithm based on the same idea.