Use Data Analysis Tools to Solve Real-life “Eating” Problems
As a student who lives in Tai Wai, I appreciate the convenience brought about by abundant food options very much. But sometimes it’s hard to find the one that suits your taste, especially when you need to balance the food and price.
In this assignment, to solve these problems, I summarized the top 10 welcomed restaurants that are suitable for students, and the top 5 welcomed cuisine styles which represent the food style of Tai Wai based on openrice data scraping, cleaning, analysis and presentation. All the data processing tools we’ve learned in class are super helpful in getting some insights in the performance of restaurants in Tai Wai, which truly gives me an idea of “what to eat” everyday. By the way, I have updated (kind of redesigned..) my webpage based on the comments given last week to reserve more spaces for the content.
However, the data processing is a lot more difficult than what I previously thought, especially the data cleaning part, where I had to manually look up for the GREL functions used in Openrefine to see how to clean the messy data I scraped. This is crazily struggling, because you need to explore the exact function you need instead of simply following the general rules. But thank god I’ve passed it through after several long long afternoons. I have also encountered some problems in SQL queries, e.g., the incorrect decimal of numbers which makes it hard to present meaningful data after calculating. But to be optimistic, these all helped me to further my understanding and real-life application on SQL.
Turning back to the data, there are mainly two objectives: 1) to find out what restaurants are both delicious and suitable to students; 2) what kinds of cuisine style can represent the food in Tai Wai. Firstly, I broke down these objectives into smaller indicators. For the first one, I decided to calculate the “favorable rate (number of likes/number of total reviews)” to determine whether the food is good or not, and I also chose to use “highest price <= 200HKD" to screen out the restaurants that meet student's purchasing power. For the second goal, I again used the indicator "favorable rate" but refined them into "average favorable rate" to better represent the overall situation. Then I grouped the results by "cuisine style" to see which one is the most welcomed in Tai Wai and added the "average price" in the table for viewer's further reference.
What's worth mentioning is that after all the calculation and sorting, I got the top cuisine styles descendently ordered by "average favorable rate". But I found the cuisine style of the first entry is "international", which I think is very ambiguous and lacks value for people to get useful information. So I dropped this result and get the subsequent 5 entries as the final ones.
Just like what we've done in this assignment, data scraping and analysis are truly being helpful in making decisions and solving real-life problems. There's no reason for us to ignore these skills. Many thanks to this opportunity to exercise.
You can access my webpage here.
