Connecting Python with DHIS2 and Applying ML algorithms

Implementation - Implémentation
1

I would love to share my work with community on connecting python with DHIS2 and applying ML algorithms on available data.
Its about finding low performing health facilities based on key indicators.
Visit the post (https://www.linkedin.com/posts/lutfullah-shifaa-a325091b5_dhis-2-demo-sierra-leone-activity-7255636352455589888-trox?utm_source=share&utm_medium=member_android) here and feel free to share your comments and thoughts. I love to see other community members’ work around this topic and collaborate.
Good Luck :+1:

3 Likes
2

Thank you for sharing your innovation with the DHIS2 community!

It’s great to see you’ve been able to use the Play instances API to make this work. Have you tested this on any real instances?

Thanks!

1 Like
3

Hello @Gassim
Yes, I am using it in our quarterly performance reviews to pull data from different sources including from Afghanistan DHIS2.
Thanks

2 Likes
4

Nice! It’s great to hear it’s already being used in the field.

1 Like
5

thank you @LutfullahShifaa for sharing your results from a very novel case study in Afghanistan of triangulating data from various sources, including dhis2, and processing via ML for performance reviews.

I thought you might be interested in this analysis from the HIV sector in Kenya, which took a similar k means clustering approach, but the focus is here on reporting consistency from counties (completeness and timeliness).

Another paper here on sharing these insights from those analysis via dashboards

If other CoP members have similar experiences in processing dhis2 data with machine learning algorithms, this might be a good thread to build on and share :slight_smile:

2 Likes
6

Thank you @brian for sharing the papers and proposing to CoP members to share their experience also :+1::+1::+1:I am totally in favor of it.

2 Likes