This is a cooperation result with Dr. May Yuan, and has presented in Yuan, M., & Liau, Y.T. (2015). From Spatial Analysis to Placial Analysis. In, the Evolving GIScience workshop in memory of Pete Fisher. University of Leicester, UK

excel files:

• tulsa_code: tulsa UCC codes to IBR codes
• join_table: aggregated into 12 crime types

python codes: ###data processing###

• 1tulsa: join UCC code with IBR codes
• 2con_tulsa: aggregate crime types and join IBR description
• 3shape: reproject and generate shape

###stability test###

• point_fre: random sampling approach

  • repeat: the number of repeating running the random sample approach (integer)
  • gridsize: cell size (integer)
  • table1: crime data (pandas dataframe)
  • outfile: output file of quadrat tif

• polyraster: polygonize raster

  • infile: quadrat tif
  • outfile: quadrat shapefile

• union: union polygons

  • infile: quadrat shapefile
  • outfile: union shapefile

• polysplit: split polygons with individual attributes

  • infile: union shapefile
  • outfile: individual polygon shapefile

• work_stab: the code to use the above four modules

###crime count###

• crimecount: count crime events within polygons

  • pointfile: crime events (dbf file)
  • polygonfile: polygons after the random sample approach (shapefile)

• SOMtable: organize 12 crime types in a table

  • dbf: dbf files of crime count
  • shp: shapefiles of crime count
  • outfile: output file (csv file)
  • pointfile: output for central point shapefiles

• work_SOM: the code to use the above two modules

###SOM group### -cluster3 -cluster4 -cluster_group: create sequences of crime types, incidence date - crimeshp: crime data (shapefile) - groupshp: after random sample approach (shapefile) - group(integer) - repeat: whether to count repeated sequences (0: non-repetitive; 1: repetitive) - outputcsv: output file (csv file)

-work_g: the code to use the cluster_group function