You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
IDOT publishes traffic data (2023). This data could potentially be very valuable for the commercial valuations team and for the residential model.
Let's explore opportunities. The Commercial Valuations team would be interested in learning our findings and in having the data (as, say, traffic counts per PIN), though it's also possible this could be useful for the res model.
Complication: IDOT's data doesn't seem to have a geographic column.
Phase 1 goal (est. 1 week)
Jean found a REST service. Test requests to the REST service to see if we can get data on a local leaflet map. (Or, if you can find a different way to assign geographies, great!)
Description of the data. What is it? Any spatial variance?
Phase 2 goal (est. 1-2 weeks)
Add raw data to s3
Adding to the lake: A high-level design doc of what this might look like to add to our data lake. Were we to construct average traffic counts per parcel (per year), what transformation steps are needed? Where might this data live in our data lake? Would this ingest be one-time, or annual (like some of our other ingest scripts)?
Add to lake, such that we can eventually get estimated traffic counts per PIN. Please anticipate heavily consulting with Jean and Billy.
Phase 3 goal (est 1 week)
Featurize this for the model pipeline, test adding it to the model, and produce a model feature report.
Phase 4 goal:
Prepare to provide a flat file extract, per non-class-2 PIN, to the Commercial Valuations team.
Update docs as necessary for how to maintain/update this feature.
Stretch goal:
Data impacts: Spend some time digging into this data, comparing traffic counts to OSM maps of primary and secondary roads. Do traffic counts basically correlate with OSM primary and secondary roads (e.g., primary road = more traffic, secondary road = less traffic), such that adding traffic count wouldn't give meaningful new information? Or, are there instances where adding traffic counts does provide information not already conveyed in primary/secondary roads? Draft a memo with your findings.
The text was updated successfully, but these errors were encountered:
IDOT publishes traffic data (2023). This data could potentially be very valuable for the commercial valuations team and for the residential model.
Let's explore opportunities. The Commercial Valuations team would be interested in learning our findings and in having the data (as, say, traffic counts per PIN), though it's also possible this could be useful for the res model.
Complication: IDOT's data doesn't seem to have a geographic column.
Phase 1 goal (est. 1 week)
Phase 2 goal (est. 1-2 weeks)
Phase 3 goal (est 1 week)
Phase 4 goal:
Stretch goal:
The text was updated successfully, but these errors were encountered: