settingsLogin | Registersettings

Extracted entities numbers not matching results for that entity

0 votes

Hi -

We're experiencing unusual behavior with the entities tab in the Appraisal Module. I've attached a couple of screen caps. Basically, the Person entities window reports that there are x number of messages matched, but when we click through to view those messages, the application finds more than the reported number.

Just wondering if this is a known issue and what the cause may be.

Thanks,
farrell


asked May 4, 2016 by laissezfarrell (470 points)

2 Answers

+1 vote
 
Best answer

Hi Farrell and Jessica,

Thanks for your feedback. We agree that the discrepancy can be confusing and are discussing presenting the information better to allay confusion. Basically, the numbers provided on the browse screens align with the entity indexing performed during ingest. In contrast, the number provided on the message review screen is based on a keyword search ePADD performs using the extracted entity selected as a search term. This may result in a higher number since it will return additional results that would not be caught by the NER, e.g. lower case spellings.

This has led to some understandable confusion, which we aim to address in future releases. I hope this helps clarify.

Best,

Josh

answered May 4, 2016 by Josh_Schneider (5,040 points)
selected May 4, 2016 by Josh_Schneider

That is very helpful. Thanks!

0 votes

I'm also curious about this as we've recently experienced the same behaviour.

answered May 4, 2016 by Jessica
ePADD is a software package developed by Stanford University's Special Collections & University Archives that supports archival processes around the appraisal, ingest, processing, discovery, and delivery of email archives.
...