Drunk Already | By Coffee

Hey guys,

I just finished the second assignment of data scraping, cleaning and analysis. My focus is to study the [cafes] in [Causeway Bay], and I have adopted three aspects that people have showed interests on:

A. Zero-unsatisfaction

Personally speaking, seeing a cafe who has received even one “dislike” annoys me. As one of the basic attributes people use to know whether a cafe is good or not, the number of dislikes does matter a lot. Therefore, by applying the “0-dislike” filter, I managed to find the cafes who never receives a bad reputation on OpenRice, reducing people’s concerns on trying these cafes.

B. Cost-effectiveness (students-only)

Me myself as a student know the consideration about price, especially in Hong Kong. It is crucial to know whether the place I am attracted to costs 500HKD for just one cup of coffee. As OpenRice has in its price classification, there are four in total: below 50HKD, 51-100HKD, 101-200HKD, 201-400HKD, I used this as my filtering condition, generating the results of the top 5 choices for coffee lovers whose cost are under 100, as well as receiving more than 500 in both likes and review. This conclusion may be helpful to students like me to spend affordable money on a good coffee experience.

C. Socially-active

OpenRice has its nature as a social platform that displays all the relevant info for people to search for a nice place to go. A helpful function is the “bookmark” which ensures people with not only the visibility to see how many people are also interested in this cafe, but also the possibility of predicting how many repeat customers among them. In this sense, I used the number of bookmarks (over 20K+) to have 10 cafes selected for people who value this social attribute.

You can have a look here.