Inlämning av Examensarbete / Submission of Thesis

Niklas Lavesson MSE-2003:18, pp. 45. Inst. för programvaruteknik och datavetenskap/Dept. of Software Engineering and Computer Science, 2003.

The work

Författare / Author: Niklas Lavesson
Titel / Title: Evaluation of classifier performance and the impact of learning algorithm parameters
Abstrakt Abstract:

Much research has been done in the fields of
classifier performance evaluation and optimization.
This work summarizes this research and tries to answer
the question if algorithm parameter tuning has more
impact on performance than the choice of algorithm. An
alternative way of evaluation; a measure function is also
demonstrated. This type of evaluation is compared with
one of the most accepted methods; the cross-validation
test. Experiments, described in this work, show that
parameter tuning often has more impact on performance
than the actual choice of algorithm and that the measure
function could be a complement or an alternative to the
standard cross-validation tests.

Ämnesord / Subject: Datavetenskap - Computer Science\Artificial Intelligence

Nyckelord / Keywords: classifier performance, evaluation, optimization

Publication info

Dokument id / Document id:
Program:/ Programme Magisterprogram Programvaruteknik, 40 poäng/Masters programme Software Engineering
Registreringsdatum / Date of registration: 09/16/2004
Uppsatstyp / Type of thesis: D-Uppsats/Magister/Master

Context

Handledare / Supervisor: Paul Davidsson
paul.davidsson@bth.se
Organisation / Organisation: Blekinge Institute of Technology
Institution / School: Inst. för programvaruteknik och datavetenskap/Dept. of Software Engineering and Computer Science
Inst. för Programvaruteknik och Datavetenskap S-372 25 Ronneby
+46 455 38 50 00
http://www.ipd.bth.se/

Files & Access

Bifogad uppsats fil(er) / Files attached: nla_mse2003_18.pdf (240 kB, öppnas i nytt fönster)