I
Indrė Žliobaitė
Researcher at Helsinki Institute for Information Technology
Publications - 25
Citations - 2851
Indrė Žliobaitė is an academic researcher from Helsinki Institute for Information Technology. The author has contributed to research in topics: Concept drift & Computer science. The author has an hindex of 12, co-authored 16 publications receiving 2175 citations. Previous affiliations of Indrė Žliobaitė include University of Waikato & Aalto University.
Papers
More filters
Journal ArticleDOI
A survey on concept drift adaptation
TL;DR: The survey covers the different facets of concept drift in an integrated way to reflect on the existing scattered state of the art and aims at providing a comprehensive introduction to the concept drift adaptation for researchers, industry analysts, and practitioners.
Book ChapterDOI
Active learning with evolving streaming data
TL;DR: This paper develops two active learning strategies for streaming data that explicitly handle concept drift, based on uncertainty, dynamic allocation of labeling efforts over time and randomization of the search space.
Book ChapterDOI
Pitfalls in benchmarking data stream classification and how to avoid them
TL;DR: It is demonstrated how a naive classifier considering the temporal component only outperforms a lot of current state-of-the-art classifiers on real data streams that have temporal dependence, i.e. data is autocorrelated.
Journal ArticleDOI
An ecometric analysis of the fossil mammal record of the Turkana Basin.
Mikael Fortelius,Indrė Žliobaitė,Ferhat Kaya,Faysal Bibi,René Bobe,Louise N. Leakey,Meave G. Leakey,David B. Patterson,Janina Rannikko,Lars Werdelin +9 more
TL;DR: It is suggested that the regionally arid Turkana Basin may between 4 and 2 Ma have acted as a ‘species factory’, generating ecological adaptations in advance of the global trend, and temporally and spatially resolved estimates of temperature and precipitation are provided.
Journal ArticleDOI
Combining similarity in time and space for training set formation under concept drift
TL;DR: A method for training set selection, particularly relevant when the expected drift is gradual, is developed, which shows the best accuracy in the peer group on the real and artificial drifting data.