Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Clicks for item/user combinations that have no impressions #56

Open
hackebeil opened this issue May 10, 2017 · 3 comments
Open

Clicks for item/user combinations that have no impressions #56

hackebeil opened this issue May 10, 2017 · 3 comments

Comments

@hackebeil
Copy link

Hi guys.

When looking at the daily interactions, we noticed that there are some clicks by users for certain items where no impression has been recorded previously. Meaning, the interaction data suggests that some items are clicked by users before an impression has been made. The ratio of item/user combinations for which this is the case is not negligible (about 30% in our case). At this moment, this makes it very problematic to analyze the data with performance indicators, since the calculation of, for example, the conversion rate depends on the accuracy of the impression count.

Is this an error or can this be explained by some user behavior?

Cheers

@dkohlsdorf
Copy link
Contributor

Hey,
this can be explained since users can click on items in the job search.
The item appears in a users search result and the user clicks it.
Appearing in the results does not trigger an impression.

Daniel

@hackebeil
Copy link
Author

Hi Daniel.

A follow up question: The above mentioned "problem" (click but no impression) also happens for item/user combinations that we actually recommended. Meaning, we recommend item A to user B, user B clicks on item A, but there is no impression for A/B.

Is this due to the filtering you mentioned? And if so, why would you filter item/user combination that would have been successful? Is it just a random subsampling?

Cheers

@dkohlsdorf
Copy link
Contributor

No we filter if a user already received a push recommendation in an 8 hour window. That is our recommender in this case.

Daniel

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants