|
Estimating the Final Price in Online Auctions Based on Neural Networks
|
Hamidreza Yunespour   , Mehdi Barati , Maryam Bayati |
|
|
|
Abstract: (695 Views) |
| In the last decade the importance of e-commerce has increased compared to before due to changes in the business habits of society.growth of the economy and technology has led to the production of large, yet cheap data that is easily accessible along with the development of e-commerce. Although predicting the final price in online auctions has many advantages, so far no model has been offered that can show successful performance in this regard. The proposed methods are mainly based on multiple regression which do not have promising results, in this study, deep recursive neural networks with hidden BiLSTM unit and regression feeder network and decision tree and combination of the above methods have been used. In this method, to reduce the mean square root of the error in predicting the final price, a combination of each of the above methods and a feature selection technique based on the firefly metaheuristic algorithm has been used. Experimental results in two eBay hybrid online auction databases show that the use of neural networks and a combination of traditional machine learning methods and deep networks is successful in predicting the final price. |
Article number: 4 |
| Keywords: Estimating the Final Price, Online Auctions, Neural Networks, Winner’s curse |
|
|
Full-Text [PDF 1710 kb]
(196 Downloads)
|
Type of Study: Applicable |
Subject:
General
Received: 2022/11/26 | Accepted: 2022/12/15 | Published: 2023/01/1
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
