A paired neural network model for tourist arrival forecasting

Yao, Yuan; Cao, Yi; Ding, Xuemei; Zhai, Jia; Liu, Junxiu; Luo, Yuling; Ma, Shuai; Zou, Kailin

doi:10.1016/j.eswa.2018.08.025

Cited by 38 publications

(29 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typical examples of datadriven methods are: backpropagation neural network (BPNN) [14,15], support vector regression (SVR) [16], locally weighted learning (LWL) [17], etc. Among them, the artificial neural network (ANN) has been widely adopted to predict tourist flow [18,19], and proved to outshine model-driven methods like ES or ARIMA. However, there are several defects with the data-driven methods: the systematic modeling process of traditional NNs is lacking; the model parameters need to be selected through repeated tests; the ANN cannot predict the tourist flow in complex systems with multiple tasks.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Tourist Flow Based on Deep Belief Network and Echo State Network

Cheng¹,

Zhao²

2019

RIA

View full text Add to dashboard Cite

The accuracy of tourist flow prediction is crucial to the sustainable development of tourism industry. However, it is very difficult to forecast the highly nonlinear tourist flow in an accurate manner. The artificial neural network (ANN) has been widely adopted to predict nonlinear time series, but its shallow structure cannot effectively learn the features of high-dimensional tourist flow data. To solve the problem, this paper puts forward a tourist flow prediction model based on deep learning (DL). First, the deep belief network (DBN), with its strong ability to extract nonlinear features, was employed to extract effective features through unsupervised learning of historical tourist flow. Next, the echo state network (ESN) was effectively fused with the DBN. The ESN was placed at the top layer of the tourist flow prediction model, serving as the logic regression layer. Finally, the offline training and prediction effects of the proposed ESN-DBN were verified through experiments on the holiday tourist flow data extracted from a tourist center, and compared with those of two classical prediction models, namely, backpropagation neural network (BPNN) and autoregressive integrated moving average (ARIMA). The results show that the ESN-DBN achieved a mean absolute percentage error was below 12% and a rational computing time; the proposed model also outperformed the two classical models in prediction accuracy. The research results provide an important reference for the forecast of tourist flow and planning of tourism development.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction of Tourist Flow Based on Deep Belief Network and Echo State Network

Cheng¹,

Zhao²

2019

RIA

View full text Add to dashboard Cite

show abstract

“…The the authors of Reference [22] present a method for tourist arrival and tourist demand forecasting. The method includes a novel paired neural network model.…”

Section: Related Workmentioning

confidence: 99%

“…Heterogeneous tourist information usage to build behaviour analysis models. Researchers use text data [15,23,26,27] (user reviews, attractions ratings, attraction requests and keywords in search engines); tourist photographs [19,28] for the analysis of meta-information and the image itself; data on movements and attraction visiting [17,18,21,22,24,[29][30][31]. Using heterogeneous information will allow the extraction of different behavioural components, which contributes to an increase in the number of possible behavioural patterns.…”

mentioning

confidence: 99%

“…Neural networks usage to analyze and predict the behaviour of tourists. The authors of References [19,28] use convolutional neural networks for image analysis, References [22,[29][30][31]33] work with LSTM networks that allow memorizing previous states, Reference [27] describes recurrent networks for analyzing tourist descriptions of attractions, self-organizing maps [23,25] provide information clustering, the authors of References [16,21,30,31] use combinations of neural networks to improve prediction results. Neural networks, as a rule, work more accurately than similar models, but they require a large amount of data for correct training and revealing hidden dependencies in the provided data, and for each task the volume depends on the type of the task itself (prediction of time events requires more data than classification ) and the number of certain input parameters (the more parameters, the more data is needed to identify dependencies between them).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Tourist Behaviour Analysis Based on Digital Pattern of Life—An Approach and Case Study

Mikhailov

Kashevnik

2020

Future Internet

View full text Add to dashboard Cite

The tourism industry has been rapidly growing over the last years and IT technologies have had a great affect on tourists as well. Tourist behaviour analysis has been the subject of different research studies in recent years. This paper presents the digital pattern of life concept which simplifies the tourist behaviour models’ construction and usage. The digital pattern of life defines the general concepts of tourist behaviour, connects the tourist and the digital world and allows us to track behaviour changes over time. A literature review of the current state of the research in selected fields is performed for identifying the existing problems. The case studies of behaviour analysis based on classification, clustering and time series events behaviour models are shown. An ontological approach and artificial neural networks are used during behaviour model construction, training and evaluation. The gathered results can be used by smart tourism service developers and business stakeholders.

show abstract

“…With the emergence of artificial intelligence (AI) techniques like fuzzy logic [11]- [13] and neural networks (NNs) [14], [15], more and more scholars have attempted to predict the VTV in ports by data-driven prediction methods (DDPMs) [16], [17]. The DDPMs can autonomously learn the nonlinear, dynamic changes in the historical data on the VTV in ports.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Vessel Traffic Volume in Ports Based on Improved Fuzzy Neural Network

Tian

Wang

2020

IEEE Access

View full text Add to dashboard Cite

The accurate prediction of vessel traffic volume (VTV) is very helpful to rational scheduling of port resources and reducing vessel accidents. However, the traditional VTV prediction methods face problems like the overfitting of historical data and prediction inaccuracy. To solve these problems, this paper improved the fuzzy neural network (FNN) with quantum genetic algorithm (QGA). Firstly, the basic principles of neural network (NN), fuzzy inference and the QGA were introduced in turn. Then, the weights of the FNN were optimized by the QGA. On this basis, the authors established a VTV prediction model based on the improved FNN. Finally, the established model was applied to predict the VTV in an actual port of China, in comparison with several classic NNs. The VTV data were collected based on the length and gross tonnage of vessels. The results show that the improved FNN outperformed the contrastive methods in the accuracy of VTV prediction, thanks to the optimization by the QGA. The proposed method enjoys a great application potential in the prediction of the VTV in ports. INDEX TERMS Vessel traffic volume (VTV), prediction model, fuzzy neural network (FNN), quantum genetic algorithm (QGA).

show abstract

A paired neural network model for tourist arrival forecasting

Cited by 38 publications

References 41 publications

Prediction of Tourist Flow Based on Deep Belief Network and Echo State Network

Prediction of Tourist Flow Based on Deep Belief Network and Echo State Network

Tourist Behaviour Analysis Based on Digital Pattern of Life—An Approach and Case Study

Prediction of Vessel Traffic Volume in Ports Based on Improved Fuzzy Neural Network

Contact Info

Product

Resources

About