Sašo Deroski
Jožef Stefan Institute
7 Papers
11 Citations
Sašo Deroski is an academic researcher from Jožef Stefan Institute. The author has contributed to research in topics: Random forest & Cluster analysis. The author has an hindex of 6, co-authored 7 publications.
Chat about Author
Papers
An extensive experimental comparison of methods for multi-label learning
TL;DR: The results of the analysis show that for multi-label classification the best performing methods overall are random forests of predictive clustering trees (RF-PCT) and hierarchy of multi- label classifiers (HOMER), followed by binary relevance (BR) and classifier chains (CC).
835
Tree ensembles for predicting structured outputs
TL;DR: This paper develops methods for learning two types of ensembles (bagging and random forests) of predictive clustering trees for global and local predictions of different types of structured outputs, and proposes to build ensemble models consisting of predictive clustered trees, which generalize classification trees.
296
Hierarchical annotation of medical images
TL;DR: The experiments show that the proposed HMC system outperforms the best-performing approach from the literature (a collection of SVMs, each predicting one label at the lowest level of the hierarchy), both in terms of error and efficiency.
195
Two stage architecture for multi-label learning
TL;DR: This work proposes a Two Stage Architecture (TSA) for efficient multi-label learning and suggests that TSCCM and TSPCCM outperform the competing algorithms in terms of predictive accuracy, while TSVM has comparable predictive performance.
38
Development of a knowledge library for automated watershed modeling
TL;DR: A library of components for building semi-distributed watershed models is developed in a formalism compliant with the equation discovery tool ProBMoT, which can automatically construct watershed models from the components in the library, given a conceptual model specification and measured data.
13