And this is our project, Correlation of Temperature and Tourism in Baguio. Our main objective is to use Data science techniques in gaining new insights regarding climate change and global warming and form recommendations to positively impact tourism in Baguio.
One of the effects of climate change and global warming is a change in tourism patterns. For example, rising temperatures in Southern Europe could prompt tourists to choose cooler destinations or shift their vacation to spring or autumn instead of summer. A study entitled “High Temperatures and Tourism: Findings from China” also supports the idea. High temperature influences travel patterns, making tourists more inclined to seek relief from the heat. The same can be observed locally. Baguio is one of the popular tourist destinations as it offers a cooler temperature compared to the rest of the country.
Is there a correlation between the number of tourists and the average temperature of Baguio?
What are the individual trends of the average temperature of Baguio and the number of tourists?
Null Hypothesis: There is no correlation between the temperature and the number of tourists in Baguio.
Alternative Hypothesis: There is a negative correlation between the temperature and the number of tourists in Baguio.
We propose making a correlation model as well as a predictive model on the two variables to see their correlation and future trends. Insights will be gathered from the result to help in forming recommendations regarding climate change and global warming as well as tourism in Baguio.
This is done with the help of Python!
We used python libraries such as BeautifulSoup for retrieving HTML content and Selenium for importing the webdriver module in automation.
We simulated accessing of the Weather Underground dataset using Chrome browser. All the information from the dataset were collected such as daily temperature, humidity, pressure, dew point, wind speed, and precipitation.
For the number of tourists, this is simply accessed from the Freedom of Information Philippines website. The dataset also has other information such as length of stay, occupancy details, and so on.
The irrelevant variables have been removed from the complete dataset. This is done manually via the editing of the spreadsheets generated by the data collection. The other variables for the weather data are kept in the daily spreadsheet in the case that it is needed in the future. The other data regarding occupancy details is saved on our group drive should we need it in the future.
We used some preprocessing techniques such as normalization and standardization.
We used seaborn and matplotlib to generate plots about the relationships of our data.
We used cross-correlational analysis via Python to confirm the initial findings from the correlation matrix.
We created a time series model using SARIMA to forecast the number of tourists in the upcoming years. According to the prediction of the model, there is a slow increase in the number of tourists during peak season. However, further testing and validation should be conducted to improve the accuracy of the model. One of the difficulties enocuntered was making the data stationary for SARIMA due to the presence of trends and outliers (ex. pandemic decline). Better handling of this can increase model performance. Lastly, hyperparameter tuning techniques such as grid search may also help improve performance.
Is there a correlation between the number of tourists and the average temperature of Baguio?
By plotting both the average monthly temperature and the monthly number of tourists in Baguio, the seasonality in both variables becomes more apparent. However, there appears to be no consistent relationship between them which is further supported by the correlation matrix on the right. At times, tourist peaks align with the lowest temperatures, while at other times, they coincide with the highest temperatures. A possible interpretation of this is that the number of tourists in Baguio is high during the summer when the temperature is at its highest.
Given the correlation matrix, there is a correlation of -0.09 between the temperature and the number of tourists in Baguio. This value denotes a very weak negative correlation.
Thus, as verified by our hypothesis testing, there is no correlation between the temperature of Baguio and the number of tourists.
What are the individual trends of the average monthly temperature of Baguio and the number of tourists?
As seen in the plot, the average monthly temperature in Baguio exhibits seasonality, with the yearly minimum during the start/end of the year and the yearly maximum around the second quarter of the year. This is expected as it is the hottest during the summer season and the coldest around December.
As seen in the plot, the number of tourists exhibits seasonality, with it peaking during the start of the second quarter (summer) and towards the end of the year (“Ber-months”). Furthermore, there was also a steady increase in the number of tourists each year up until the decline in 2020 due to the pandemic. The number however is steadily rising after 2021, exhibiting a recovery from the pandemic decline.
There is no strong correlation between the temperature in Baguio and the number of tourists. However, the number of tourists exhibits seasonality, with it peaking around the end/start of a year. There is also a steady increase in the number of tourists pre/post-pandemic. As for the temperature, there are no significant changes which must be maintained.
For further studies, the difference between the temperature in Baguio and other regions in the Philippines can be used instead of the temperature in Baguio itself. Doing so may better encapsulate the phenomenon of people choosing to go to cooler destinations to escape the heat.
Given the steady increase of tourists in Baguio, it is important that its local government ensures the protection of the environment amidst the growth in tourism through effective policies. An example is conducting rehabilitation projects during non-peak seasons. After all, good environmental conditions are beneficial in attracting tourists.
Rainbow greetings! I am Geri, a 4th year BS Computer Science student from UP Diliman. As someone who enjoys continuously learning and figuring out why and how things work, I hope to become a data scientist in the future. I also enjoy UI/UX design.
Aside from being a woman in CS, I am also a queer advocate. I devote my time away from academics to work for programs that help serve the LGBTQ+ community. I hope to one day be able to apply my learnings in the CS field to my advocacy.
Hello!!! I am igi, and I am a 4th year BS Computer Science student from UP Diliman. My interests in the field of CS include data science, machine learning, and software engineering. I plan to become a data analyst in the future.
My hobbies/interests outside CS include Dota 2, Pokemon, and running (away from my responsibilities, jk). Thank you for your interest in our project and I hope you enjoy reading our portfolio =).
Hello!!! I am Jim, and I am a 4th year BS Computer Science student from UP Diliman. I am currently working as a developer for an MMORPG called Dekaron Rising. I also moderate the forums for the game.
My hobbies outside CS are mostly gaming but I am also a speedcuber (speedsolving rubik's cubes).