Legacy Dev Forum Posts

 View Only

Sign Up

Back to discussions
Expand all | Collapse all

Quality API - Evaluations Aggregates Query

  • 1.  Quality API - Evaluations Aggregates Query

    Posted 06-05-2025 18:18

    wchavez | 2020-10-05 21:13:15 UTC | #1

    Good afternoon Forum:

    I started to use the Quality API - Evaluations Aggregates. I have the following questions, and I appreciate your help if you can guide to me in order to use it well:

    1. If I use the POST /api/v2/analytics/evaluations/aggregates/query, the interval that I need to specify should cover the evaluations created at that time? Or the finished (scored) evaluations within that time? Or the evaluations that were associated to the conversations within that time?
    2. Should this endpoint retrieves only the finished or already scored linked evaluations for conversations?
    3. If a conversation has two evaluators with the same evaluation and only one evaluator has scored it, then should the aggregations only return the metrics resume by considering only the one that was scored? How I can retrieve the non-scored evaluation in order to persist its state in my database using the same endpoint?
    4. If it is not possible, What do you recommend to us in order to get a bulk of evaluations grouped by conversationId with its scored and non-scored linked evaluations?

    Thank you in advance.

    anon33938915 | 2020-10-07 17:12:30 UTC | #2

    Hello,

    Thanks for your questions!

    1. Your first thought is correct: you should specify a time interval that covers when the evaluations were created.
    2. Yes, each of the aggregate metrics will include data only from submitted/scored evaluations.
    3. Yes, only the score from the evaluator who has completed the evaluation will be included in the aggregations.
    4. Do you think the bulk conversation query endpoint could work for your use-case? It is /api/v2/analytics/conversations/details/jobs, and you can see it here. You could use the EvaluationDetailQueryPredicate to search for conversations that have formId present. That should return conversations that have evaluations, even if those evaluations are not submitted/scored yet. For this endpoint, you should use a time interval based on the conversation start date. Please feel free to post here again if you'd like any help forming that query, or if you don't think that will work for you.

    system | 2020-11-07 17:23:44 UTC | #3

    This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.

    This post was migrated from the old Developer Forum.

    ref: 8976



Related Content