ULMS Discovery : DITF March 29, 2024 Meeting

Relevancy ranking and view settings

Table created by CH (See spreadsheet)

More than 150 cases

FRBR and Dedupe

People didn’t like the way it grouped things
People got some FRBR ‘fatigue’

We’ve all suffered the Google effect

Does it help the end user

People didn’t see “4 more versions” to click and then to click down to the right view

Resource and material type

Secondary resource types - is this coming over to this group?

Title/portfolio matching

CH: “We heard at this meeting that there is an internal setting that might help with exact title match searching (Ex Libris has to set it up, not us), but it might have to be done at the consortial level, and only one field could be ‘boosted’ in this option.”

Response: ExLibris wasn’t sure what this reference was, they told CH to put in a ticket

Check in with Nikki DeMoville - ask if she can share her background information, examples and solution/development requests

CL: it’s clear how things are blended - using the API - Can is experimenting with that. Active process of adjusting boosting 5 records with highest relevancy score. More than 50% use articles. They had a problem where too many books were showing up at the top.

Lens norm - shorter the field the more precise the match.

Microfilm as wanting to be lower that other mediums of “books”

Do Rapido results have ranking too? - CL hasn’t gotten to that stage yet in her API investigation. How do they treat these records differently from locally or CDI.

Why did people make these changes:

1. made a change because someone else said to

2. made a change because of a ticket recommendation from from ExLibris

3. made a change because of local request

boosting stuff we pay for?

open access resources showing up at the top - and the records are not great

More than 150 cases

Faceting

User testing from SJSU

OA tagging and unpaywall??

More than 150 cases