Ανάλυση δεδομένων αναζητήσεων τουριστικών προορισμών και αξιοποίησή τους σε μοντέλα προβλέψεων
Analysis of tourist destination search data and its utilization in predictive models
View/ Open
Keywords
Τουρισμός ; Νευρωνικά δίκτυα ; Μοντέλο πρόβλεψης ; Ανάλυση δεδομένωνAbstract
The internet serves as a critical tool for consumers and businesses alike, generating a
vast and invaluable volume of data through user searches on search engines like
Google. These searches constitute a rich data source that can be leveraged to extract
information and analyze consumer behavior patterns.
This thesis focuses on analyzing tourism movement data in Greece, specifically
investigating the correlation between Google search queries and the actual tourism
revenue recorder by Greek state. The primary objective of this research is to determine
whether Google Trends search data can be utilized to create a model that accurately
forecasts Greece’s tourism revenue. The analysis includes an examination of the
relationship between search trends and overall tourism revenue.
To this end, Principal Component Analysis was employed to consolidate search
data into a single index, named the Tourism Demand Index, representing Greece’s
overall attractiveness as a tourist destination as reflected in relevant online searches.
Following this, the correlation between the index and actual tourism revenue was
investigated, aiming to use the index as a primary factor in training the revenue
forecasting model.
Data processing was conducted using Power BI and Tableau, which enable the
visualization of data and highlight emerging trends. Furthermore, to develop the
predictive model, machine learning techniques were applied via the Orange Data
Mining tool, using neural networks to forecast future revenue based on online search
trends.
This study yielded significant findings, revealing a statistically significant
correlation between the Tourism Demand Index and actual revenue. The results suggest
that Google search data can serve as a valuable tool for forecasting tourism revenue,
providing industry stakeholders with the capacity to make data-driven decisions and
refine their strategic approaches.