Transfer k-means: a new supervised clustering approach

Postgraduate Thesis uoadl:1710277 322 Read counter

Unit:
Κατεύθυνση Λογική και Θεωρία Αλγορίθμων και Υπολογισμού
Library of the School of Science
Deposit date:
2017-07-11
Year:
2017
Author:
Teloni Pelagia
Supervisors info:
Άρης Παγουρτζής, Αναπληρωτής Καθηγητής, Τομέας Τεχνολογίας Πληροφορικής και Υπολογιστών, ΕΜΠ
Original Title:
Transfer k-means: a new supervised clustering approach
Languages:
English
Translated title:
Transfer k-means: a new supervised clustering approach
Summary:
Supervised and unsupervised learning are two fundamental learning schemes
whose difference lies in the presence and absence of a supervisor
(i.e. entity which provides examples) respectively. On the other hand,
transfer learning aims at improving the learning of a task by using auxiliary knowledge.
The goal of this thesis was to investigate how the two fundamental paradigms,
supervised and unsupervised learning, can collaborate in the setting of transfer learning.
As a result, we developed transfer-$K$means, a transfer learning variant
of the popular $K$means heuristic.

The proposed method enhances the unsupervised nature of $K$means, using
supervision from a different but related context as a seeding technique,
in order to improve the heuristic's performance towards more meaningful
results. We provide approximation guarantees based on the nature of
the input and we experimentally validate the benefits of the proposed
method using documents as a real-world example.
Main subject category:
Science
Other subject categories:
Algorithms and Theory of Computation
Keywords:
clustering, transfer learning, domain adaptation, density ratio estimation, natural language processing
Index:
No
Number of index pages:
0
Contains images:
Yes
Number of references:
81
Number of pages:
73
thesis_pelagia_teloni.pdf (3 MB) Open in new window