<h1>Data Maketh (Hu)man</h1>
<p>Is our data really an <b>accurate</b> representation of ourselves? Enough to determine if we are <b>trustworthy</b> enough to own a credit card? </p>
<div style="margin-top: 30px;">
   · Clustered based on a combination of the individual's ethnicity and whether they were approved. <br/>
   · Sized based on the amount of debt the individual has. <br />
   · Colored based on what industry the individual works in.<br />
</div>
<svg id="base" width="1500" height="950">
<g id="layer1"></g>
</svg>
<style>
.node {fill:none;}
.node text {font-size: 12px; fill: white; text-anchor: middle}
.CommunicationServices {fill:#CD5C5C; background-color:#CD5C5C; color: #dddddd;}
.ConsumerDiscretionary {fill:#FF7F50; background-color:#FF7F50;}
.ConsumerStaples {fill:#C71585; background-color:#C71585; color: #dddddd;}
.Education {fill:#F0E68C; background-color:#F0E68C;}
.Energy {fill:#BA55D3; background-color:#BA55D3;}
.Financials {fill:#3CB371; background-color:#3CB371;}
.Healthcare {fill:#AFEEEE; background-color:#AFEEEE;}
.Industrials {fill:#000080; background-color:#000080;}
.InformationTechnology {fill:#FFEBCD; background-color:#FFEBCD;}
.Materials {fill:#CD853F; background-color:#CD853F;}
.RealEstate {fill: #FAEBD7; background-color: #FAEBD7;}
.Research {fill:#DCDCDC; background-color:#DCDCDC;}
.Transport {fill:#4169E1; background-color:#4169E1;}
.Utilities {fill:#800000; background-color:#800000;}
.fociLabel {font-size: 20px; fill: #000000; text-anchor: middle;}
</style>
data = FileAttachment("credit_card_approvals_dataset.csv").csv()
{
for (let i = 0; i < data.length; i++) {
if (data[i]["Approved"] == "1" && data[i]["Ethnicity"] == "White") {
data[i]["Approved_Ethnicity"] = "white_approved"
}
else if (data[i]["Approved"] == "0" && data[i]["Ethnicity"] == "White") {
data[i]["Approved_Ethnicity"] = "white_rejected"
}
else if (data[i]["Approved"] == "1" && data[i]["Ethnicity"] == "Black") {
data[i]["Approved_Ethnicity"] = "black_approved"
}
else if (data[i]["Approved"] == "0" && data[i]["Ethnicity"] == "Black") {
data[i]["Approved_Ethnicity"] = "black_rejected"
}
else if (data[i]["Approved"] == "1" && data[i]["Ethnicity"] == "Asian") {
data[i]["Approved_Ethnicity"] = "asian_approved"
}
else if (data[i]["Approved"] == "0" && data[i]["Ethnicity"] == "Asian") {
data[i]["Approved_Ethnicity"] = "asian_rejected"
}
else if (data[i]["Approved"] == "1" && data[i]["Ethnicity"] == "Latino") {
data[i]["Approved_Ethnicity"] = "latino_approved"
}
else if (data[i]["Approved"] == "0" && data[i]["Ethnicity"] == "Latino") {
data[i]["Approved_Ethnicity"] = "latino_rejected"
}
else if (data[i]["Approved"] == "1" && data[i]["Ethnicity"] == "Other") {
data[i]["Approved_Ethnicity"] = "other_approved"
}
else if (data[i]["Approved"] == "0" && data[i]["Ethnicity"] == "Other") {
data[i]["Approved_Ethnicity"] = "other_rejected"
}
// if (data[i]["PriorDefault"] == "0") {
// data[i]["PriorDefault"] = "NoPriorDefault"
// } else {
// data[i]["PriorDefault"] = "YesPriorDefault"
// }
if (data[i]["Employed"] == "0") {
data[i]["Employed"] = "Unemployed"
} else {
data[i]["Employed"] = "Employed"
}
if (data[i]["Industry"] == "Real Estate") {
data[i]["Industry"] = "RealEstate"
}
}
}
fociFeatures = ({
"white_approved":{x: 200, y: 200},"white_rejected":{x: 450, y: 200},"black_approved":{x: 700, y: 200},"black_rejected":{x: 950, y: 200},
"asian_approved":{x: 200, y: 450},"asian_rejected":{x: 450, y: 450},"latino_approved":{x: 700, y: 450},"latino_rejected":{x: 950, y: 450},
"other_approved":{x: 200, y: 700},"other_rejected":{x: 450, y: 700}
})
clElem = d3.select(svgContainer).select("#layer1")
options = ({
"startFoci": fociFeatures,
"fociField": "Approved_Ethnicity",
"classField": "Industry", // to set their class/color
"dotScaleFactor": 5,
"strength": -1,
"alphaTarget": 0.09,
"collisionFactor": 1.5
})
clusters = dvCluster(clElem,data,800,800,"Debt",options)
labelLayer = clElem.append("g").attr("id","labels")
dvClusterFociLabels(labelLayer,fociFeatures,{offsetY:155})
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.