Understanding the spread of Extinct and Endangered languages across India

Visualizing the spread of low resource languages across India.

Data Description

The dataset of extinct and endangered languages around the world is created by The Guardian and is available here.

Variables: The name of language, longitude, latitude, degree of endangerment and the number of speakers.

Data Visualisation

Languages by latitude/longitude and Population are plotted.

The interactive visualisation of Languages by Latitude/longitude inform us, where all the extinct and endangered languages are located across India.

Similarly, Languages by number of speakers is plotted.

#collapse-hide
import numpy as np
import pandas as pd
pd.options.mode.chained_assignment = None

from IPython.display import HTML
from chart_studio import plotly
import plotly.graph_objs as go
from plotly import tools
from plotly.offline import iplot, init_notebook_mode
init_notebook_mode()

language_data = pd.read_csv('data.csv', usecols=[0, 1, 5, 7, 10, 12, 13])
language_data = language_data.rename(
    columns={'Name in English':'language', 'Country codes alpha 3':'locations',
             'Degree of endangerment':'risk', 'Number of speakers':'population'})
language_data.columns = language_data.columns.str.lower()
language_data['risk'] = language_data['risk'].str.title()
language_data['population'] = language_data['population'].fillna(-1)

# endangered or extinct languages in India
language_ind = language_data[language_data['locations'].str.contains('IND') == True]

language_ind['risk'] = language_ind['risk'].replace(
    ['Vulnerable', 'Definitely Endangered', 'Severely Endangered',
     'Critically Endangered', 'Extinct'], [1, 2, 3, 4, 5])

language_ind = language_ind[['language', 'risk', 'population', 'latitude', 'longitude']]

Languages by Latitude/Longitude

#collapse-hide
labels = ['Isolated', 'Threatened', 'Endangered', 'Abandoned', 'Extinct']
colors = ['rgb(0, 157, 220)', 'rgb(128, 206, 237)', 'rgb(255, 182, 128)',
          'rgb(255, 115, 13)', 'rgb(242, 23, 13)']

traces = []
for i in range(1, 6):
    traces.append(dict(
        type = 'scattergeo',
        lon = language_ind[language_ind.risk == i]['longitude'],
        lat = language_ind[language_ind.risk == i]['latitude'],
        text = language_ind[language_ind.risk == i]['language'],
        mode = 'markers',
        name = labels[i-1],
        marker = dict( 
            size = 12,
            opacity = 0.85,
            color = colors[i-1],
            line = dict(color = 'rgb(255, 255, 255)', width = 1)
        )
    ))
# print(traces)
layout = dict(
         title = 'Languages by Latitude/Longitude in India (2016)<br>'
                 '<sub>Click Legend to Display or Hide Categories</sub>',
         showlegend = True,
         legend = dict(
             x = 0.85, y = 0.4
         ),
        geo = dict(
                 scope = 'asia',
                 showland = True,
                 landcolor = 'rgb(250, 250, 250)',
                 subunitwidth = 1,
                 subunitcolor = 'rgb(217, 217, 217)',
                 countrywidth = 1,
                 countrycolor = 'rgb(217, 217, 217)',
                 showlakes = True,
                 lakecolor = 'rgb(255, 255, 255)')
        )


figure = dict(data = traces, layout = layout)

#collapse-hide
import plotly.io as pio

pio.show(figure)

Languages by Population

#collapse-hide
language_ind = language_ind.sort_values('population', ascending = False)
language_ind['text'] = language_ind['language'] + '<br>' + 'Population ' + language_ind[
                                                                 'population'].astype(str)

new_traces = []
for i in range(1, 6):
    new_traces.append(dict(
        type = 'scattergeo',
        lon = language_ind[language_ind.risk == i]['longitude'],
        lat = language_ind[language_ind.risk == i]['latitude'],
        text = language_ind[language_ind.risk == i]['text'],
        mode = 'markers',
        name = labels[i-1],
        hoverinfo = 'text+name',
        marker = dict( 
            size = (language_ind[language_ind.risk == i]['population'] + 1) ** 0.18 * 6,
            opacity = 0.85,
            color = colors[i-1],
            line = dict(color = 'rgb(255, 255, 255)', width = 1)
        )
    ))

new_layout = dict(
         title = 'Languages by Population in India (2016)<br>'
                 '<sub>Click Legend to Display or Hide Categories</sub>',
         showlegend = True,
         legend = dict(
             x = 0.85, y = 0.4
         ),
         geo = dict(
             scope = 'asia',
             showland = True,
             landcolor = 'rgb(250, 250, 250)',
             subunitwidth = 1,
             subunitcolor = 'rgb(217, 217, 217)',
             countrywidth = 1,
             countrycolor = 'rgb(217, 217, 217)',
             showlakes = True,
             lakecolor = 'rgb(255, 255, 255)')
        )

new_figure = dict(data = new_traces, layout = new_layout)

#collapse-hide
import plotly.io as pio

pio.show(new_figure)