new PR for changes to prior blog posts and new frameworks post #923
Conversation
Ananya-Joshi
requested review from
carlynvandyke,
melange396 and
rlunde
as code owners
✅ Deploy Preview for cmu-delphi-main ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
|
@melange396 Would love to get your thoughts on the content! |
| blogdown::html_page: | ||
| toc: true | ||
| --- | ||
| Insights from public health data can keep communities safe. However, identifying these insights in large volumes of modern public health data can be laborious <sup>[1](#link1)</sup>. As a result, over the past few decades, public health agencies have built monitoring systems, like [ESSENCE](https://www.cdc.gov/nssp/new-users.html) (CDC), [EIOS](https://www.who.int/initiatives/eios) (WHO), and [DHIS2](https://dhis2.org/) (WHO), where users can set custom statistical alerts and then investigate these alerts using data visualizations <sup>[2](#link2)</sup>. These alerting systems largely follow the following formula <sup>[3](#link3)</sup> as shown in Fig 1.: |
There was a problem hiding this comment.
In the netlify preview, there is weird spacing after each of the links (including the footnote references) in this line/paragraph. I am not sure whats causing it... Are there tabs hidden in here instead of spaces? Is it related to your usage of the <sup> tags? Maybe try this carat notation instead?
There was a problem hiding this comment.
I think it should be taken care of now - seems like a sup issue!
| <center> | ||
| </center> | ||
|
|
||
| Now imagine each data point per signal was given an outlier score, and there is an arbitrary binary threshold over all the data. If we develop granular enough rankings, we can directly compare outlier points between different signals without arbitrary thresholds. Then, a data reviewer can then focus on data points in their order of importance and stop the review at any time. |
There was a problem hiding this comment.
nits (also im not sure what you mean by "arbitrary binary threshold") :
| Now imagine each data point per signal was given an outlier score, and there is an arbitrary binary threshold over all the data. If we develop granular enough rankings, we can directly compare outlier points between different signals without arbitrary thresholds. Then, a data reviewer can then focus on data points in their order of importance and stop the review at any time. | |
| Now imagine each data point was given an outlier score, and there is an arbitrary binary threshold over all the data. If we develop granular enough rankings, we can directly compare outlier points between different signals more objectively. Then, our data reviewer can focus on data points in their order of importance and stop the review at any time. |
There was a problem hiding this comment.
This is supposed to show that if you set a threshold like 0.9 above which everything is an outlier and below which everything is not an outlier, that's a bit of an arbitrary number (what about 0.89? why not 0.91?). So if you use rankings, you can combine these scores from different signals in a meaningful way because otherwise, there might just be a lot of things that are classified as outliers even if there is variation in those scores.
There was a problem hiding this comment.
Let me know if you think there's a clearer way to share this - I was hoping the picture would make it more clear, but if not this is the most important part of the blog, so maybe we can iterate on how to make it more understandable!
There was a problem hiding this comment.
@melange396
Just wanted to double check on this part - does it make more sense now?
There was a problem hiding this comment.
it makes sense. it was the word "binary" that threw me off; you mention thresholds a number of times before this, but this is the only time you call it a "binary threshold".
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
Co-authored-by: melange396 <[email protected]>
| <center> | ||
| </center> | ||
|
|
||
| Now imagine each data point per signal was given an outlier score, and there is an arbitrary threshold where scores higher than that threshold are considered outliers. If we develop granular enough rankings, we can directly compare outlier points between different signals without arbitrary thresholds. Then, our data reviewer can focus on data points in their order of importance and stop the review at any time. |
Co-authored-by: melange396 <[email protected]>
2 participants
No description provided.