[NEW_FEATURE][CLIENT_REQUEST]Multi-selection on duplicate

[lduf]

I want to :

Suggest an idea

My suggestion is about :

SCAN website

My suggestion is about : (other choice)

Describe your suggestion

Duplicates: once the algorithmic comparison is done we receive a list of cases with different scores and an option "mark as not duplicate", it is necessary to be able to carry out multiple selections and/or set the scores that can be excluded (accepted as not duplicates), for example those <70, otherwise we have a long list that we need to verify and tick one by one

First name

JP

Last name

Baraybar

Email

jpbaraybar@icrc.org

Company

ICRC

[github-actions[bot]]

AI-Assisted Response:

Issue Summary: Enhancement for Bulk Selection and Filtering of Duplicate Cases on SCAN Website

Context:

The user has proposed a feature for the SCAN website that deals with managing duplicates identified by an algorithm. After an algorithmic comparison, the user is presented with a list of cases, each with a score indicating the likelihood of being a duplicate. The current process requires manual review to mark items as "not duplicate", which is tedious, especially when dealing with long lists.

Suggestion:

JP Baraybar suggests implementing a feature that would allow bulk actions on the list of potential duplicates, such as being able to perform multiple selections at once. Additionally, the user requests an option to automatically exclude cases below a certain score threshold (e.g., <70) from the list of duplicates, thus reducing manual effort in the review process.

Issue Type:

Feature Request

Questions for Clarification:

1. @lduf, could you specify if the score threshold should be a fixed value (e.g., <70) or adjustable by the user?
2. Does the current system allow any form of bulk selection, or is it entirely individual case selection?
3. Are there any specific criteria or fields by which you would like to filter the cases apart from the score?
4. Are you envisioning this feature as an addition to an existing UI on the SCAN website, or would this require designing a new interface component?
5. Can you provide any information about the existing technological stack or frameworks used on the SCAN website to assess the feasibility of your request?

[github-actions[bot]]

AI-Assisted Response:

Issue Summary: Feature Request for Bulk Selection and Score-Based Exclusion in Duplicate Management

A feature request has been submitted for the SCAN website suggesting an enhancement to the duplicate handling functionality. The user wants the system to support multi-select actions and to filter out potential duplicates by setting a score threshold, such as scores below 70, to be automatically considered as non-duplicates. This would streamline the process of managing duplicates by reducing the manual effort of reviewing and marking each individual case when dealing with long lists of algorithmically determined duplicates.

Contextual Information:

Organization: International Committee of the Red Cross (ICRC) Submitter: JP Baraybar (jpbaraybar@icrc.org)

Issue Type:

Feature Request

Questions for Clarification:

• @lduf, could you clarify if you need the multi-select feature for operating within a list or across multiple lists/pages?
• Would you prefer a configurable score threshold that you can set and modify, or should there be fixed thresholds?
• Apart from marking cases as not duplicates, are there other bulk actions you anticipate needing for these records?
• How often do you encounter the necessity to process potential duplicates in this manner?
• Are there any specific workflows or permission levels that should be considered while implementing this feature to ensure proper usage and control?

[github-actions[bot]]

AI-Assisted Response:

Issue Summary: Feature Request for Batch Selection on Duplicate Cases

Context: JP Baraybar from ICRC has suggested an enhancement for the SCAN website's duplicate handling functionality.

Rephrased Suggestion: The user proposes the introduction of a feature that allows batch processing of potential duplicate cases after an algorithmic comparison has been made. Currently, a list presents each case with a score, and users have to individually mark cases as "not duplicate." The user requests the capability to perform multiple selections at once and/or to define a threshold score so that all cases falling below this score (e.g., less than 70) will automatically be considered as "not duplicates," thus reducing the manual effort needed to review and tick each case individually.

Issue Classification: This should be categorized as a feature request, as it involves adding new functionality to the existing system.

Tagging the Creator: Hello @lduf, to ensure that we fully understand your feature request and can assess its feasibility, could you please provide additional details on the following:

1. Is batch deselection currently supported in any form, or is the process completely manual at the moment?
2. What is the average number of cases that a user typically has to review when considering duplicates?
3. Are there any existing user permissions or roles that we should consider when implementing this feature (e.g., only certain users should have the ability to batch select)?
4. Could you clarify whether users require an option to review the cases that are automatically marked as "not duplicates" by the system for additional quality control?
5. Are there potential concerns regarding the accuracy of the algorithm used to score the cases that might affect the implementation of automated deselection?

Understanding these aspects will help us ensure that the proposed feature aligns with user needs and system capabilities.

[lduf]

Duplicate #339