gistfile1.py

from __future__ import division
from json import dumps

data = [
    'arostn a ard aoyd rad radrpd',
    'aodha a aoydydyo 43qlgdi',
    'ta <irhaotyqodl>og</irhaotyqodl>'
]

# dict of {words => [line nos]}
occurs = {}

for i in range(len(data)):
    line = data[i].split()

    for word in line:
        if word in occurs:
            occurs[word].append(i)
        else:
            occurs[word] = [i]

print dumps(occurs, indent=4)