Inexplicable search filtering result: wrong county persists at top of the list.

LegacyUser

Juli said: https://www.familysearch.org/search/r...



The town of Kunbaja is in south-central Hungary, very near Szabadka, which is now in Serbia. Why are the top seven results from Borsod county, which is partly now in Slovakia, i.e. exactly the opposite side of the country? And why do these results persist despite filtering the list to just Bács-Bodrog county?

It's as if the search filters are being completely ignored.

I checked the places database, and neither of the Borsod places has anything incorrect in it that I can see, and nothing with the same names is coming up in Bács-Bodrog, either. The only error or problem is that the search algorithm fails to turn up "Csaba, Borsod", because it's actually Hejőcsaba (also written as Heő-Csaba and other variations). (The village has now been absorbed into the city of Miskolc.) But nothing like that applies to Balajt, which is the result at the very top of the list. What on Earth is going on here?

LegacyUser

Tom Huber said: It looks to be some kind of problem parsing the Bács-Bodrog on my Win 10 with Chrome. My results using your link shows only Borsod places.

I checked to see if the problem was with the Place standard, but Bács-Bodrog is clearly identified as a county in the system.

So it may be a parsing problem or some other kind of bug. That's my guess.

LegacyUser

Cousin David said: It is regrettable that not every single “town” can be cross-referenced with every single name by which it has been known throughout the centuries. Surely, researchers are aware of this. It simply cannot be done to everyone’s satisfaction. Make sure the alternate spellings are in the information. A time frame might also be identified. US researchers find similar issues with the formation of Counties and where the records are kept. But it’s simply not possible at this stage. Even getting researchers to use proper diacritical marks is a training exercise. Formatting Hispanic names; German and Eastern European municipalities. And the exchange of an I for an E or an E for O. Patience.

LegacyUser

Lundgren said: Juli,

Thanks for letting us know.

We'll look at it.

LegacyUser

Juli said: Wow, even better:
https://www.familysearch.org/search/r...



This is with the collection restricted to Hungary Catholic baptisms, but otherwise the same filter settings as the search and screenshot above. 14 results, every single one of them from Mezőtárkány, with the county in this case mis-identified as Borsod: Mezőtárkány is and near as I can tell always has been on the Heves county side of the county line. In any case, it's a 3 hour 40 minute drive on modern highways from Kunbaja.

LegacyUser

Lundgren said: Juli,

I haven't had a chance to dig deeply. (It's a holiday here.)

I expect this is what is happening though:

The birth place for the first record in the results is: "Balajt, Borsod, Hungary": https://www.familysearch.org/search/r...

If you put: "Balajt, Borsod, Hungary" into the places research tool, there is no exact match. However, the fifth one on the list of places that may match is: "Bácsborsód, Bács-Bodrog, Hungary"

We will confirm this is what is going on, but I expect that is why you are seeing these results.

LegacyUser

Lundgren said: I believe you are saying the place: "
Mezőtárkány, Borsod, Hungary" does not, and has not ever existed?

LegacyUser

Juli said: I checked the 1808 "Repertorium Locorum" (https://digitalia.lib.pte.hu/hu/pub/l...) and it also has Mező-Tárkány in "comitatus Hevesiensis", so no, "Mezőtárkány, Borsod" has never existed.

LegacyUser

Lundgren said: I reported the problem to the team that is responsible for the data.