| Field | Value |
|---|---|
| Title | STL-DADLM |
| Type | Source Code |
| Language | Python |
| License | |
| Status | Research Code |
| Update Frequency | NO |
| Date Published | 2020-03-20 |
| Date Updated | 2020-03-20 |
| Portal | https://github.com/tulip-lab/open-code |
| URL | https://github.com/tulip-lab/open-code/tree/master/STL-DADLM |
| Publisher | TULIP Lab |
| Point of Contact | A/Prof. Gang Li |
This package (STL-DADLM) is the deep learning algorithm for tourism demand forecasting with STL decomposition. Please be aware that:
- The training of DLM needs extra efforts based on specific data set, and direct running of the provided code DOES NOT always generate the promised performance.
- For the training of the model on the data set, please spend your own patient time and the code publisher will NOT provide assistance on this issue.
If you use it for a scientific publication, please include a reference to this paper.
- Yishuo Zhang, Gang Li, Birgit Muskat, Rob Law (2020). Tourism Demand Forecasting: A Decomposed Deep Learning Approach. Journal of Travel Research, June 2020
BibTex information:
@article{ZLML2020,
title = {Tourism Demand Forecasting: A Decomposed Deep Learning Approach},
volume = {0},
doi = {10.1177/0047287520919522},
journal = {Journal of Travel Research},
author = {Zhang, Yishuo and Li, Gang and Muskat, Birgit and Law, Rob},
month = {06},
year = {2020},
keywords = {Tourism demand forecasting, tourism planning, AI-based forecasting, deep learning, decomposing method, over-fitting},
}
The related dataset for above paper can be found at TULIP Lab Open-Data:
HK2012-2018: Tourism Demand Forecasting Data for Hong Kong on six visitor markets from January 2012 to December 2018
- Python 3.6
- Keras
- Tensorflow