Build Your Portfolio With These Data for Good Project IdeasFrom Data Visualization to Data Science, Get Inspired to Create Something.
Jasmine VasandaniBlockedUnblockFollowFollowingJul 5This article was inspired by a talk I gave titled “Data (Science) for Good” at Women in Data’s “Building Your Data Science Toolbox” symposium.
During my journey to becoming a data scientist, I built my portfolio with projects that showcased my hard skills while also contributing to social causes.
For instance, when I noticed that sports analytics didn’t have many analyses on women’s sports teams, I used an unsupervised machine learning algorithm to build a dream team of WNBA players.
Or when I kept hearing about bias in artificial intelligence, I decided to build a machine learning model that mitigated bias in AI.
In both of these projects, I showcased my hard skills as a data scientist while also contributing to the greater good and upholding my values.
To give back to the community of data professionals, I wanted to share some resources and tips to help you boost your portfolio with data for good projects.
Let Go of Fear and Just StartCredit: WOCinTech Chat.
The first tip I want to share with you is to let go of fear and start using the skills you already have to work on a data project.
Early in my journey to becoming a data scientist, I was most comfortable with using Excel to interpret data.
At the time, I didn’t know how to develop machine learning models in python.
Using just the skills I had back then, I could have started working on data-driven projects — but the only thing that held me back was fear.
If you’re in the early stages of learning specialized, hard skills to become a data professional, you can still work on projects using the existing skills you have.
I wish I could go back in time and tell myself this, but since I can’t I’m writing this article to tell you instead — just start!.I try to include resources here that everyone can be inspired by, no matter how novice or expert you are.
Create a Project WorkflowDownload the Data Science Workflow Canvas.
Whether you know what kind of data-driven projects you want to work on or you’re still not sure, start brainstorming your ideas.
Brainstorming, strategizing, ideating — the one thing that these words have in common is the tenant that there’s no such thing as a “wrong” idea.
For anyone that wants a structured way to brainstorm data science project workflows, I designed this canvas (pictured above) to help you get started.
If you need other tools to help you strategize a project workflow, you can also try hand-drawing or digitizing mind-maps, using design thinking tools, or simply opening up a blank word document (or a flipping to a blank page in your notebook if you’re an analog type of person) and start writing down your ideas.
Data for Good DatasetsNow that you have an idea for how to strategize your project workflow, you can start getting your hands dirty with these data-for-good datasets.
Or you can also scrape your data, or get out there and collect your own.
Most of the portals I’m sharing in the list below provide easy-to-download datasets.
Some of these resources are not purely focused on “data for good” but still have relevant datasets that you can download.
And lastly, these are just a few out of the many datasets out there!The Humanitarian Data ExchangeDiversity of Artists in Major U.
MuseumsCivil Rights Data CollectionGovernment Open Databases: U.
Open Data, U.
cities open dataClimate DataGender Data PortalHealth DataKaggle’s Data Science for Good: City of Los AngelesAcross the TimelineDiversity DataFiveThirtyEight: Our DataOr you can use Google’s Dataset Search to find something elseData VisualizationCredit: Giorgia LupiData visualization can be a powerful tool to convey the insights that you’ve gathered from your data analysis, especially if you’re working on a data-for-good project.
These are some of my favorite resources that showcase data visualizations for good.
Mona ChalabiGiorgia LupiPublications/BlogsHomepage of Gayta Science.
These are just a few blogs and/or publications that feature a collection of data-driven articles.
The PuddingGayta ScienceHer Hoop StatsCenter for Spatial ResearchData Feminism (hat tip Mitchell Bohman for sharing this with me!)Get InvolvedThere are so many ways to get involved and build communities around the notion of “data for good.
” Here are a few resources of ways you can get involved, whether it be through volunteering for an organization or attending a conference.
Data for Black LivesLatinX in AIDataKindWomen In DataBlack in AIWomen in Machine Learning and Data ScienceData For Good: CanadaConclusionI hope that these resources inspire you to create data-for-good projects!.Remember that these resources are just a snapshot of what’s available.
If you want to share more resources or projects that you’ve worked on, comment below!—Thanks to Anuva Kalawar for always sharing data-for-good resources with me.
Jasmine Vasandani is a data scientist and strategist.
You can learn more about her here: https://github.