BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.cienciaenmexico.com//igisc-2026//speaker//99GRTQ
BEGIN:VTIMEZONE
TZID:CST
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=5SU;BYMONTH=10;UNTIL=20001029T080000Z
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:STANDARD
DTSTART:20010930T030000
RRULE:FREQ=YEARLY;BYDAY=5SU;BYMONTH=9;UNTIL=20010930T080000Z
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:STANDARD
DTSTART:20021027T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000402T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4;UNTIL=20000402T090000Z
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20010506T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=5;UNTIL=20010506T090000Z
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20020407T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4;UNTIL=20220403T090000Z
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-igisc-2026-RJSRN7@pretalx.cienciaenmexico.com
DTSTART;TZID=CST:20261113T113000
DTEND;TZID=CST:20261113T114500
DESCRIPTION:Across much of Sub-Saharan Africa and South Asia\, formal land 
 registration covers only a small fraction of land parcels in rural and inf
 ormally held areas. Nigeria's own government places its national formal ti
 tling rate below 5 per cent (Federal Ministry of Housing and Urban Develop
 ment\, Nigeria\, 2025). The wealth is real. The records are not. This is t
 he condition under which AI systems are increasingly trained and deployed\
 , and it has received far less attention than the kind of bias the AI fair
 ness literature usually worries about.\n\nThat literature is good at catch
 ing unequal treatment. Demographic parity\, equalised odds\, calibration: 
 these tools ask whether identifiable groups are treated differently by an 
 algorithm. But a fairness audit built on these metrics would not detect th
 e pattern this paper is concerned with\, because the affected populations 
 are absent from the training records altogether\, not merely treated diffe
 rently within them. No model output exists to audit\, because no input dat
 a exists to audit it with.\n\nThis paper argues that geospatial data absen
 ce\, rather than discriminatory model design\, is a primary driver of AI r
 epresentativeness bias in low-data economies. To make this argument tracta
 ble\, the paper develops the Geospatial-AI Alignment (GAIA-M) framework an
 d operationalises it through the GAIA-M Data Infrastructure Quality Index 
 (GAIA-M DIQI)\, a composite score across five dimensions: cadastral covera
 ge\, data accessibility\, temporal stability\, GIS layer availability\, an
 d informal tenure recognition.\n\n**GAIA-M DIQI = (0.30 × CC) + (0.20 × 
 DA) + (0.20 × TS) + (0.15 × GIS) + (0.15 × ITR**)\n\nEach dimension is 
 scored from 1 to 10 against a rubric grounded in ISO 19157 (ISO\, 2013) an
 d the UN-Habitat Fit-for-Purpose Land Administration framework (Enemark\, 
 2009)\, sorting territories into four risk bands from Very High (0–39) t
 o Low (80–100). Most geospatial data quality work asks whether a map is 
 accurate\, complete\, and consistent. GAIA-M asks a different question on 
 top of that: whether the data is good enough to make an AI system trained 
 on it fair. The weighting structure reflects the centrality of cadastral v
 isibility to AI training data\, with cadastral coverage assigned the highe
 st weight\; sensitivity testing using equal weighting and three alternativ
 e weighting schemes produced the same risk classification for every countr
 y considered.\n\nNigeria\, Bangladesh\, and Indonesia were chosen delibera
 tely for how different their cadastral systems are from one another\, foll
 owing a most-different systems design (George and Bennett\, 2005). All thr
 ee share large Muslim populations and well-documented land administration 
 problems\, which is what makes the contrast between them meaningful. Rwand
 a is not a Muslim-majority country\, and was not chosen by the same logic\
 ; it serves a different purpose\, set out below.\n\nEach number used in th
 is comparison is drawn from a named\, publicly available source\, not esti
 mated or constructed by the authors. Nigeria sits below 5 per cent formal 
 registration\, a case of what the framework calls static absence\, records
  that were largely never created\, rather than records that have since gon
 e bad. Bangladesh sits at roughly 50 per cent\, with about half of all lan
 d parcels held through unregistered documents according to a 2020 househol
 d survey (Rahman and Hossain\, 2020). But that figure understates the deep
 er problem\, which is dynamic rather than static: land along the Brahmaput
 ra\, Ganges\, and Meghna shifts fast enough that a registered boundary can
  be substantially wrong within months. Indonesia has reached approximately
  71.5 per cent of total land area through more than a decade of systematic
  registration under its PTSL programme (Journal of Property Research\, 202
 5)\, the strongest of the three Muslim-majority cases. And then there is R
 wanda\, which has achieved near-universal registration\, more than 10 mill
 ion parcels titled nationwide between 2009 and 2013 (Gillingham and Buckle
 \, 2014)\, despite having the lowest GDP per capita of the four countries\
 , roughly 1\,070 US dollars (World Bank\, 2024).\n\nThat last fact is what
  makes Rwanda worth including. It breaks an obvious alternative explanatio
 n: that the whole pattern is really just about national income\, not land 
 administration at all. If income were driving the result\, Rwanda should b
 e the weakest case here. It is the strongest. One country is not proof of 
 anything\, but it is real\, sourced evidence against the simpler explanati
 on.\n\nNigeria scores 28\, Bangladesh 46\, Indonesia 68\, and Rwanda 72. T
 he full dimension-level breakdown behind each of those numbers is set out 
 in the accompanying description.\n\nRwanda's composite score is more cauti
 ous than its registration figure alone would suggest. Cadastral coverage i
 s well evidenced\; accessibility\, temporal stability\, and GIS infrastruc
 ture maturity are not\, so the score holds back where the evidence does. N
 o correlation coefficient\, significance test\, or spatial autocorrelation
  statistic appears anywhere in this paper. Four countries is not a sample 
 large enough to carry that kind of weight\, and the comparison is offered 
 as illustrative throughout\, not as something statistically tested.\n\nFor
  GIScience\, the contribution is this: spatial data quality assessment bec
 omes a direct input into algorithmic governance\, through a scoring instru
 ment that is transparent and recomputable rather than an opaque statistica
 l model. For AI governance\, the contribution is a pre-deployment index th
 at names the type of representativeness risk\, absence\, decay\, or concen
 tration\, before a model is trained rather than after it has already faile
 d. The next step is to test this properly: Pearson correlation and spatial
  autocorrelation analysis would be the right tools\, once sub-national dat
 a exists to run them on. Indonesia is the strongest candidate for that wor
 k\, since province-level religion data is already published by BPS-Statist
 ics Indonesia\, and province-level land registration progress is already t
 racked by the national land agency\, ATR/BPN. The other three countries in
  this paper do not yet have both halves of that data publicly available\; 
 Indonesia does.
DTSTAMP:20261122T123846Z
LOCATION:Video Room
SUMMARY:From Geospatial Data Gaps to AI Representativeness Bias: A Framewor
 k for Low-Data Economies - Ayobami Oluyemi Ayoola
URL:https://pretalx.cienciaenmexico.com/igisc-2026/talk/RJSRN7/
END:VEVENT
END:VCALENDAR
