course-dprep / airbnb-effect-of-price-on-reviews Goto Github PK

Maandag 11 september af

The research question is clearly articulated and important. The choice for the research method (e.g., regression analysis) is motivated well. The way of deployment (e.g., PDF report, dashboard, …) is useful and accessible to potential knowledge users, and clearly communicates the conclusions of the analysis. The automated and reproducible workflow is of potential use to other students and the larger scientific community.

Take a deep dive into the Make pipeline and learn how it works so we can apply this to our project in order to automatize the code

this is a practice issue

Deliverable

Come up with a dataset to use

• all issues still relevant?
• file names all in lower caps
• makefiles incorrect: need file names as target, not a self-chosen name (unless it's your first rule)
• dependency structure incorrect (i.e., prerequisites need to be added)
• final rule (all) only needs to have the "final" data set to build, not all
• make needs to realize it's "done".
• clean legacy files - they are stored in your file version history anyways
• explicitly choose which files to keep (e.g., R vs. Rmd)
• output needs to be stored in gen, not in a subfolder of src.
• remove legacy files

To what extent do the top 25% most expensive single rooms get more reviews than the bottom 25% of single rooms compared across (5?) European capitals on a quarterly time base? An analysis of multiple cities and time-frames.

[src_download_data.Rmd], this dataset is done in a way that it contains all the data of the 100 most cheap and 100 most expensive private rooms in 5 capital cities. we need to merge this with the reviews (amount of reviews) in R. please write code for this

Try to make to code more structured and comprehensive.

type of analysis
conclusion analysis
how to run the project

Maandag 11 september af

1.2 Repository structure and documentation (10%)
The end-to-end workflow, kickstarted with one of the workflow templates available at Tilburg Science Hub, is made publicly available on GitHub. The repository contains a readme.md (in markdown format, so that it renders well on GitHub), which clearly explains the project’s goal, and provides instructions to potential contributors/replicators on how to run the project. The project has a concise and accurate name, enticing the potential user to explore the workflow. An appropriate short name for the repository’s location is chosen (e.g., github.com/yourusername/investigating-airbnb). Additional metadata on GitHub is provided (e.g., a short project description), so that the repository feels and looks professional and complete.

Tip: Structure your README.md

The readme is not a copy of your way of deployment (e.g., your report), but links to it. Therefore, only summarize the main results in your readme (e.g., in bullet point format), or include screenshots (e.g., to your app, to your “best” figure).

More information on how a proper readme should look like can be found at Tilburg Science Hub.

save the barcharts somewhere as output of research

We have downloaded the data, and now it needs to be transformed in data that is relevant for our research. we are halfway done, but this still needs to be completed

trying to make a loop to improve the "raw"code

What can we submit to our research?

change the target variable bc its better for analysis

in folder "src/data-preperation/src_download_data.rmd". this is the file that contains the code to download all the data for the listings. also make such a code for the download of review data and merge these two datasets.

that file contains data for the top 100 and bottom 100 private rooms of all the cities!

Deliverable

A clear (short) story for chapter 2 & 3

add clear descriptions to your commits when solving issues

Hey guys,

I've figured out you've versioned your DATA in your source code / data prep directory. Please ensure to only version source code. Also note I saw many files that we really don't need (e.g., legacy files you've cloned from the template).

Please properly clean the directory.

Some help is here: https://tilburgsciencehub.com/audit/workflow-checklist/?utm_campaign=referral-short

The About should entail a short summary of the project

the README has to be updated, with more details than how it previous was.

The idea is to create a makefile where all the work merges. The endgoal is a makefile that operates the project properly

divide the code into chunks according to the ITO methods, so we can run these code chunks seperately through make.

in our advanced code, we need to make use of data packages in rmarkdown to show our capabilities

Deliverable

cleaned up master branch

Purpose

Decide a topic to use for a RQ