18F / 10x-MLaaS

Repository for machine learning tool, MeL, that assist in providing insights for open text data. This tool is part of the 10x Machine Learning as a Service project (formerly known as Qualitative Data Management).
https://10x.gsa.gov
Other
13 stars 8 forks source link

PAB Review Prep #44

Closed csmcallister closed 5 years ago

csmcallister commented 5 years ago

10x Qualitative Data User story

As a person that cares about this project, I'd like us to think about the following things to include/be aware of going into the PAB meeting.

10x/Qualitative Data Management Run Through Follow Up Questions

csmcallister commented 5 years ago
screen shot 2018-09-27 at 4 08 58 pm screen shot 2018-09-27 at 4 08 07 pm screen shot 2018-09-27 at 4 07 36 pm
csmcallister commented 5 years ago

This is also interesting if we want to speak to the number federal website users:

screenshot_20180928-171332

DK1999 commented 5 years ago

2.64 billion visits is only pages with DAP code. They don't cover the whole gov.

DK1999 commented 5 years ago

Here is a before image of comments. Which of these are helpful to our writers for improving or understanding how people use our content? image

DK1999 commented 5 years ago

Here is after the HSM has evaluated them. image Now the writers don't have to read the one's marked spam. It's not just putting aside the real spammy junk. It even understands which comments are helpful to us.

DK1999 commented 5 years ago

Agencies / Projects with survey comments that are interested in our work: HHS USDA / Centers of Excellence Customer Experience CAP Goals Working Group GSA via the Office of Customer Experience.

DK1999 commented 5 years ago

Non-survey usage: PBS and Fleet Management Contact Center intake https://www.ftccomplaintassistant.gov/?utm_source=takeaction#crnt&panel1-1 https://consumercomplaints.fcc.gov/hc/en-us Search.gov

DK1999 commented 5 years ago

https://www.regulations.gov/

DK1999 commented 5 years ago

Every single member of congress has a comment box form. White House, too. They must get SO MUCH SPAM.

amoose commented 5 years ago

❤️