使用Google趋势预测旅游需求
发布时间:2023-02-05 13:43
通过网络爬虫获取网络空间数据以分析和预测物理世界的宏观事件是近年来十分重要的研究方向。本论文面向旅游业需求研究了如何利用谷歌趋势统计的网络搜索信息来预测物理世界的真实游客数量。世界各地的旅游业在迅猛发展中:荷兰的旅游业占国民生产总值的9%,其首都阿姆斯特丹是一个非常美丽的城市,有着郁金香、沟渠、游艇和各种展览馆。有许多著名的画家都生活在阿姆斯特丹,游客可以在这里的画廊里发现许多备受赞誉的杰作。阿姆斯特丹的旅游业对整个荷兰的经济发展贡献很大。因此,准确预测阿姆斯特丹的游客数量具有重要的实际意义。本文的研究思路是利用谷歌趋势信息来预测阿姆斯特丹的旅游业需求。具体的,我们利用Touristjourney对搜索查询词进行拓展和筛选,然后通过GoogleSearch Query利用谷歌趋势Google Trends返回的与查询词相关的搜索统计信息分析真实游客数量的相关性并训练得到隐马尔科夫模型;在测试阶段,将搜索参数归集到Google Trends中,获得查询词列表和对应的Google Trends信息,进而通过训练的隐马尔科夫模型进行游客数量预测。这项研究发现,谷歌趋势提供的信息对于确定阿姆斯...
【文章页数】:61 页
【学位级别】:硕士
【文章目录】:
ACKNOWLEDGEMENT
摘要
ABSTRACT
ABBREVIATIONS AND ACRONYMS
1. INTRODUCTION
1.1. TOURIST JOURNEY
1.2. PROBLEM STATEMENT
1.3. Research Objectives
1.4. RESEARCH QUESTIONS
1.5 THESIS OUTLINES
2. LITERATURE REVIEW
2.1. RELATED WORK
2.1.1. Econometric Models
2.1.2. Artificial Intelligence
2.1.3. Artificial Neural Networks (ANN)
2.2 CONTRIBUTION AND DIFFERENCE WITH PREVIOUS WORKS
2.3. BRIEF OVERVIEW OF SEARCH ENGINE AND SOCIAL MEDIA
2.3.1. Social Media
2.3.2. Search Engine
2.4. FORECASTING BY THE CITY OF AMSTERDAM
3. METHODOLOGY
3.1. THE WORKHOW OF OUR METHOD
3.2. DATA COLLECTION
3.3. HIDDEN MARKOV MODEL
3.4. ARTIFICIAL NEURAL NETWORK (ANN)
3.5. VECTOR AUTO-REGRESSIVE (VAR)
3.6. HIDDEN MARKOV MODEL AS A SOLUTION
3.6.1. Keywords Extraction and Evaluation
3.6.2. Hidden Markov Model Training
3.6.3. Hidden Markov Model Prediction
4. EXPERIMENTS AND RESULTS
4.1. EXPERIMENTAL SETUP
4.2. STATISTICS OF EXTRACTED KEYWORDS
4.3. RESULTS OF GRANGER CAUSALITY ANALYSIS
4.4. PREDICTION PERFORMANCE COMPARISON
5. CONCLUSION AND DISCUSSION
5.1. CONCLUSIONS
5.2. DISCUSSION
5.3. LIMITATIONS AND FUTURE RESEARCH
REFERENCES
AUTHOR PROFILE
DATASET FOR THE MASTER'S THESIS
本文编号:3735074
【文章页数】:61 页
【学位级别】:硕士
【文章目录】:
ACKNOWLEDGEMENT
摘要
ABSTRACT
ABBREVIATIONS AND ACRONYMS
1. INTRODUCTION
1.1. TOURIST JOURNEY
1.2. PROBLEM STATEMENT
1.3. Research Objectives
1.4. RESEARCH QUESTIONS
1.5 THESIS OUTLINES
2. LITERATURE REVIEW
2.1. RELATED WORK
2.1.1. Econometric Models
2.1.2. Artificial Intelligence
2.1.3. Artificial Neural Networks (ANN)
2.2 CONTRIBUTION AND DIFFERENCE WITH PREVIOUS WORKS
2.3. BRIEF OVERVIEW OF SEARCH ENGINE AND SOCIAL MEDIA
2.3.1. Social Media
2.3.2. Search Engine
2.4. FORECASTING BY THE CITY OF AMSTERDAM
3. METHODOLOGY
3.1. THE WORKHOW OF OUR METHOD
3.2. DATA COLLECTION
3.3. HIDDEN MARKOV MODEL
3.4. ARTIFICIAL NEURAL NETWORK (ANN)
3.5. VECTOR AUTO-REGRESSIVE (VAR)
3.6. HIDDEN MARKOV MODEL AS A SOLUTION
3.6.1. Keywords Extraction and Evaluation
3.6.2. Hidden Markov Model Training
3.6.3. Hidden Markov Model Prediction
4. EXPERIMENTS AND RESULTS
4.1. EXPERIMENTAL SETUP
4.2. STATISTICS OF EXTRACTED KEYWORDS
4.3. RESULTS OF GRANGER CAUSALITY ANALYSIS
4.4. PREDICTION PERFORMANCE COMPARISON
5. CONCLUSION AND DISCUSSION
5.1. CONCLUSIONS
5.2. DISCUSSION
5.3. LIMITATIONS AND FUTURE RESEARCH
REFERENCES
AUTHOR PROFILE
DATASET FOR THE MASTER'S THESIS
本文编号:3735074
本文链接:https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/3735074.html