Automatic Scoring.

[TwanD]

Hello everyone,

I'm beating my head with this and hoping someone can help me with this, I think it's simple but just can't think through it.

I want to use Alteryx to automatically assign a score on a scale of 0 - 10, including decimals for anything between 0 - 10.

The scoring is based on the last date a file was saved versus the current date calculated in seconds, for instance:

Last saved: 2020-09-08 15:54:47

Current date: DateTimeNow() function in Alteryx

Then the difference in seconds using the DateTimeDiff function.

All this is fine, but what I'm struggling next is to figure out the logic to assign on a scale 0 - 10, including decimals for anything between 0 - 10, for any file with lowest difference closer to 10 if not 10 exact, then it goes lower as the difference becomes greater.

Any help would be greatly appreciated.

Thanks!

[AbhilashR]

Hi @TwanD, I am thinking writing some form of formula and assigning a scale value based on the time difference range might be a potential approach, but the intervals depend on what you see in your sample set. If you had to do this manually, how would you assign the scale? Also feel free to share some sample dataset for everyone to better help.  

[TwanD]

@AbhilashR 

Please see the attached sample workflow. Currently the difference is calculated on days and the scoring is manually assigned based on a range. The result ultimately feeds into a Power BI visualization and I run into issues with each file overlaps each other in a quadrant chart (this scoring is one of the two I used for the chart).

I want to see each file as its own dot so thinking of calculating the differences on seconds instead of day but want to have the scoring automatically applied using a logic instead of manually putting in a number in the scale of 0 - 10.

Thanks.

[phottovy · Accepted solution]

Let me know if this is what you are looking for. I basically calculate to the oldest date, then calculate every other date difference as a percentage of the oldest date difference. After that, you just need to multiply the percentage by your scoring number, 10 in this case.

The attached workflow creates 100 random dates to go with your provided date and then scores them between 0 and 10.

[Qiu]

@TwanD 

By any chance, Multi Field Banning is what are looking for?

[TwanD]

@phottovy This is what I'm looking for, thank you so much! I definitely could not think it through like you did. Will test it out with my own data and will reach back out if I still have issues.

Thanks!!!

[TwanD]

@Qiu 

Unfortunately, the Multi Field Banning doesn't give me the granular detail for closely saved file and would create overlaps in scoring as the scoring via this method is whole number.

But I appreciate you showing another route. Thanks!!!