|
Title:
|
CLASSIFICATION OF EBAY BIDDING CHARACTERISTICS |
|
Author(s):
|
Tobias Dietrich , Detlef Seese , Stephan K. Chalup |
|
ISBN:
|
ISSN: 1645-7641 |
|
Editors:
|
Pedro Isaías |
|
Year:
|
2006 |
|
Edition:
|
V IV, 1 |
|
Keywords:
|
Online Auctions, eBay, Bidding Characteristics, Support Vector Classification, Parameter Selection |
|
Type:
|
Journal Paper |
|
First Page:
|
111 |
|
Last Page:
|
125 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
eBay is often regarded as one single homogeneous marketplace. Fee structure and trading rules are identical for all items traded and for all users. This paper presents evidence gathered from 251,996 eBay auctions suggesting that the market is in fact split into segments with individual bidding characteristics. Using only bidding-related data as input for Support Vector Classification algorithms, classification rates between 56% and 68% were achieved for the distinction between items offered by high-volume sellers and items offered by low-volume sellers, and between items offered in different product categories. In addition, an improved method for selecting Support Vector Machine parameters is presented. |
|
|
|
|
|
|
