1. Αποθετήριο Παραδοτέων Υποτρόφων Δράσεων ΕΣΠΑ Τριτοβάθμιας Εκπαίδευσης 2017-2023
2. Όλο το Αποθετήριο
3. Ερευνητικές ομάδες

Τύπος:	Ανακοίνωση σε συνέδριο
Τίτλος:	Investigating the benefits of exploiting incremental learners under active learning scheme
Συγγραφέας:	[EL] Κάρλος, Σταμάτης[EN] Karlos, Stamatis [EL] Κανάς, Βασίλειος[EN] Kanas, Vasileios [EL] Φαζάκης, Νικόλαος[EN] Fazakis, Nikos [EL] Αρίδας, Χρήστος[EN] Aridas, Christos [EL] Κωτσιαντής, Σωτήριος[EN] Kotsiantis, Sotiris B. S.B.
Ημερομηνία:	10/07/2019
Περίληψη:	This paper examines the efficacy of incrementally updateable learners under the Active Learning concept, a well-known iterative semi-supervised scheme where the initially collected instances, usually a few, are augmented by the combined actions of both the chosen base learner and the human factor. Instead of exploiting conventional batch-mode learners and refining them at the end of each iteration, we introduce the use of incremental ones, so as to apply favorable query strategies and detect the most informative instances before they are provided to the human factor for annotating them. Our assumption about the benefits of this kind of combination into a suitable framework is verified by the achieved classification accuracy against the baseline strategy of Random Sampling and the corresponding learning behavior of the batch-mode approaches over numerous benchmark datasets, under the pool-based scenario. The measured time reveals also a faster response of the proposed framework, since each constructed classification model into the core of Active Learning concept is built partially, updating the existing information without ignoring the already processed data. Moreover, all the conducted comparisons are presented along with the appropriate statistical testing processes, so as to verify our claim.
Γλώσσα:	Αγγλικά
Σελίδες:	13
DOI:	10.1007/978-3-030-19823-7_3
ISBN:	978-1-4503-6610-6
Θεματική κατηγορία:	[EL] Επιστήμη πληροφόρησης[EN] Information science
Λέξεις-κλειδιά:	Incremental learners; Active Learning scheme; Stochastic Gradient Descent; Query strategy; unlabeled data
Κάτοχος πνευματικών δικαιωμάτων:	© IFIP International Federation for Information Processing 2019
Ηλεκτρονική διεύθυνση του τεκμηρίου στον εκδότη:	https://link.springer.com/chapter/10.1007/978-3-030-19823-7_3
Τίτλος πηγής δημοσίευσης:	Artificial Intelligence Applications and Innovations. AIAI 2019
Σελίδες τεκμηρίου (στην πηγή):	37-49
Σειρά δημοσίευσης:	IFIP Advances in Information and Communication Technology
Όνομα εκδήλωσης:	15th International Conference on Artificial Intelligence Applications and Innovations
Τοποθεσία εκδήλωσης:	Crete, Greece
Ημ/νία έναρξης εκδήλωσης:	24/05/2019
Ημ/νία λήξης εκδήλωσης:	26/05/2019
Σημειώσεις:	This research is implemented through the Operational Program Human Resources Development, Education and Lifelong Learning and is co-financed by the European Union (European Social Fund) and Greek national funds. Part of the IFIP Advances in Information and Communication Technology book series (IFIPAICT, volume 559)
Εμφανίζεται στις συλλογές:	Ερευνητικές ομάδες