You signed in with another tab or window.Reload to refresh your session.You signed out in another tab or window.Reload to refresh your session.You switched accounts on another tab or window.Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: pgml-dashboard/content/blog/how-to-improve-search-results-with-machine-learning.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -216,7 +216,7 @@ This is considered a Supervised Learning problem, because we have a labeled data
216
216
217
217
###Training Data
218
218
219
-
First things first, we need to record some user clicks on our search results. We'll create a new table to store our training data, which are the observed inputs and output of our new relevance function. In a real system, we'd probably have separate tables to record**sessions**,**searches**,**results**,**clicks** and other events, but for simplicity in this example, we'll just record the exact information we need to train our model in a single table. Everytime we perform a search, we'll record the`ts_rank` fortheboth the**title** and**body**, and whether the user**clicked** on the result.
219
+
First things first, we need to record some user clicks on our search results. We'll create a new table to store our training data, which are the observed inputs and output of our new relevance function. In a real system, we'd probably have separate tables to record**sessions**,**searches**,**results**,**clicks** and other events, but for simplicity in this example, we'll just record the exact information we need to train our model in a single table. Everytime we perform a search, we'll record the`ts_rank` for both the**title** and**body**, and whether the user**clicked** on the result.