digitalWestie

Using clustering to group Scotland's constituencies

A few days ago I briefly got involved in The Bureau Local's Voterpower Hack. Participants were provided with demographic and political data sourced from the Office of National Statistics, the Cabinet Office and the British Election Study.

To get started, I decided to play with some Python machine learning libraries, this was partly a learning exercise but I also wondered if I could surface any insights.

The first thing I did was run a k-means clustering algorithm on the constituency demographic & voting data.

You can see the results and the code below. In k-means clustering you input how many groups you want the algorithm to produce. Here I've provided 2,4, and 6. A '1' against a constituency means that the constituency belongs to the group in the column.

digitalWestie

var sortTable = function(tableId, columnNo, direction){
  var rows = $('#'+tableId+' tbody tr').get();

  rows.sort(function(a, b) {
    var A = $(a).children('td').eq(columnNo).text().toUpperCase();
    var B = $(b).children('td').eq(columnNo).text().toUpperCase();
    
    if (!isNaN(parseFloat(A))){ A = parseFloat(A) }
    if (!isNaN(parseFloat(B))){ B = parseFloat(B) }
    

digitalWestie

digitalWestie

https://www.youtube.com/watch?v=79Qb1tsoYZg&index=27&list=RDDfx9dUIVnFw

Savage - Don't Cry Tonight - too easy going ?

Riccardo Cioni - Fog (1984) - - less danceable

PATRIZIA PELLEGRINO MUSICA SPAZIALE

digitalWestie

def rebase(i,b):
  result = []
  while i > 0:
    result.insert(0, i % b)
    i = i // b
  return ''.join(str(x) for x in result)

def perform(n,b):
  k = len(n)
  y = ''.join(sorted(n))

digitalWestie

// Collect key: object pairs

var processData = function(data, collectionAddress, ){
  var collection = data;
  for (i in collectionAddress){
    collection = collection[collectionAddress[i]]
  }

  if (!Array.isArray(collection)){
    var arrayCollection = [];

digitalWestie

ECA Cafe

digitalWestie

Open Standards Reading

Resources to help define and evaluate open standards for data.

ODI - Open Standards for Data

Guidebook on creating and using open standards for data. http://standards.theodi.org/

Cabinet Office - Policy paper - Open Standards principles

Policy outlook on open data standards. Good outline of how to evaluate standards.

digitalWestie

import csv
from urllib.parse import urlparse, parse_qs

rows = []
with open('novjan_analytics.csv', newline='') as csvfile:
  spamreader = csv.reader(csvfile, delimiter=',', quotechar='"')
  for row in spamreader:
    rows.append(row)

params = ["postcode", "q", "category", "location_type", "keyword_sort", "page"]

digitalWestie

To help with a few projects that use Scottish place names, and their geographical centres, I've merged a few datasets.

• Scottish Postcode Index, provided by the NRS
• Geocoded data from Open Streetmap Nominatim service