BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ChamberMaster//Event Calendar 2.0//EN
METHOD:PUBLISH
X-PUBLISHED-TTL:P1H
REFRESH-INTERVAL:P1H
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
DTSTART:20070101T000000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:Eastern Daylight Time
END:DAYLIGHT
BEGIN:STANDARD
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
DTSTART:20070101T000000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:Eastern Standard Time
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20261202T150000
DTEND;TZID=America/New_York:20261202T160000
X-MICROSOFT-CDO-ALLDAYEVENT:FALSE
SUMMARY:ShopTalk w/Preservica: More access\, less process: natural language search and the description backlog
DESCRIPTION:Every state archive holds records that are safely preserved but barely described\, from agency email to Microsoft 365 content. Without a catalog entry\, those records are invisible to the public\, and full description can take years.This session explores how natural language search\, built directly into the archive\, can help open those collections to researchers sooner without compromising the authoritative record. Using a real preserved collection\, we'll show:How users find records by asking questions in plain language\, even where description is minimalHow every answer cites its source record\, with provenance and context intactHow access restrictions carry through so sensitive records stay protectedAttendees leave with a practical checklist for evaluating any AI search tool against archival standards.
X-ALT-DESC;FMTTYPE=text/html:<!DOCTYPE html><html><head><title></title></head><body aria-disabled="false"><div style="margin-top:1em\;margin-bottom:1em\;font-family:Poppins\;font-size:10pt\;color:rgb(0\,0\,0)\;">Every state archive holds records that are safely preserved but barely described\, from agency email to Microsoft 365 content. Without a catalog entry\, those records are invisible to the public\, and full description can take years.</div><div style="margin-top:1em\;margin-bottom:1em\;font-family:Poppins\;font-size:10pt\;color:rgb(0\,0\,0)\;">This session explores how natural language search\, built directly into the archive\, can help open those collections to researchers sooner without compromising the authoritative record. Using a real preserved collection\, we&rsquo\;ll show:</div><ul fr-original-style="" style="list-style-position: inside\;"><li style="font-family:Poppins\;font-size:10pt\;color:rgb(0\,0\,0)\;">How users find records by asking questions in plain language\, even where description is minimal</li><li style="font-family:Poppins\;font-size:10pt\;color:rgb(0\,0\,0)\;">How every answer cites its source record\, with provenance and context intact</li><li style="font-family:Poppins\;font-size:10pt\;color:rgb(0\,0\,0)\;">How access restrictions carry through so sensitive records stay protected</li></ul><div style="margin-top:1em\;margin-bottom:1em\;font-family:Poppins\;font-size:10pt\;color:rgb(0\,0\,0)\;">Attendees leave with a practical checklist for evaluating any AI search tool against archival standards.</div></body></html>
LOCATION:
UID:e.4188.1649078
SEQUENCE:3
DTSTAMP:20261113T183849Z
URL:https://business.statearchivists.org/upcoming-events/Details/shoptalk-w-preservica-more-access-less-process-natural-language-search-and-the-description-backlog-1940731?sourceTypeId=Hub
END:VEVENT
END:VCALENDAR