search

linked-conservation-data / board-pilot-data

Processed datasets for the board reattachment pilot of the Linked Conservation Data project
1 stars 0 forks source link

Boards not linked to any events in LC data #13

Closed natuk closed 3 years ago

natuk commented 3 years ago

It appears that the boards in the LC records are not linked to any modification event and therefore there are no techniques or dates associated with them. I have checked records 2028 and 1466_027 - models below: 1466_027 gv 2028 gv

acampagnolo commented 3 years ago

This was due to the selection of records. I went through the data and selected a different set of records with more targeted data. I'll regenerate the RDFs once I have finished updating the scripts.

acampagnolo commented 3 years ago

In generating the new dataset, I've encountered 4 terms that weren't in the vocabulary. They are added and highlighted in green in the spreadsheet of inferred terms: https://docs.google.com/spreadsheets/d/1eDVvIKZGwt1h8RoLaC5rED_3P0ZpYh9OB025BhC4vRU/edit#gid=0

natuk commented 3 years ago

Hm, I have looked at a few records and again I cannot find any modification events linked to boards. I checked: 3929, 2687, 2561, 2028.

acampagnolo commented 3 years ago

you managed to pick items without board info! Try 1697, 2247, 2473, 2474, 2605 See this list for possible candidates LOC_FileSelection_20201210.xlsx

natuk commented 3 years ago

No, I am afraid that in all of these records, the boards are not connected to any modification events. See for example 1697: 1697 gv Also this example shows double modification events as noted at #19 - possibly related?

acampagnolo commented 3 years ago

Caught an xpath issue. Regenerated the RDFs (this only applied to the LoC data). The double modification is not a problem if it has a technique, the bug was duplicating one instance without adding the technique.

natuk commented 3 years ago

Thank you - these look good now.