Project translated automatically. Log in or register, to view the original Needed to: Working with the code: Determine the QueryID rating based on the user rating (the one who put the rating on that topic).Example: 20 users put ratings from 0 to 3 on the QueryID 1.The existing code has already defined the user’s assessment of their “general work” on all topics.Here is an example: Now we need to find the “best” rating: We need to take all of these 20 users and see their rating, whose rating is higher, that rating we take and display as the “true topic rating.”For example: As you can see from the file: In the end, a column called Level_base_on_rating should appear It shows a rating from 0 to 3 based on all users who have assessed QueryID.Likewise, from the first example we see that sometimes there is "no", for example, the topics have only 2 users with the same rating, then If we have 2 or more pollists with the same rating, then we have to compare their ratings.It wins the assessment that is most common.If we have all-level nothing, the best rating among these users wins.(for example, the user 1 and the user 2 have the same rating and the 1st gave a rating - 1, and the 2nd gave a rating - 3, then the column shows a rating of 3, as the best) The result: You need to build a new DataFrame, which will include a column with user ratings, as well as the QueryID rating, identified from user ratings and their ratings.and separately DataFrame, where there will be just a list of QueryIDs and some of their assessment obtained through the user comparison.Compared to: Also take an example from any 5 lines from DataFrame on the screen.I add the code, but it doesn’t need to change anything, it will only help to get user ratings.