1
00:00:00,720 --> 00:00:04,020
- For our last example,
we're going to introduce

2
00:00:04,020 --> 00:00:07,320
a combination of geocoding and mapping

3
00:00:07,320 --> 00:00:10,720
and use that information to help us take

4
00:00:10,720 --> 00:00:14,470
a live tweet stream and plot the resulting

5
00:00:14,470 --> 00:00:18,000
tweets that we get on an interactive map.

6
00:00:18,000 --> 00:00:20,850
And as you'll see, the tweets will be

7
00:00:20,850 --> 00:00:24,435
plotted as map markers which the user

8
00:00:24,435 --> 00:00:28,820
of the map can click on
to pop open the map marker

9
00:00:28,820 --> 00:00:32,560
and to look at the sender of the tweet

10
00:00:32,560 --> 00:00:35,210
and the tweet's text as well.

11
00:00:35,210 --> 00:00:37,540
So, we'll once again be collecting

12
00:00:37,540 --> 00:00:40,080
streaming tweets using the same techniques

13
00:00:40,080 --> 00:00:43,190
that we used in the last two examples.

14
00:00:43,190 --> 00:00:47,730
Now, as far as location
information in tweets is concerned

15
00:00:47,730 --> 00:00:51,520
by default, for privacy
reasons, twitter disables

16
00:00:51,520 --> 00:00:54,640
precise location
information that would give

17
00:00:54,640 --> 00:00:56,650
latitude and longitude coordinates

18
00:00:56,650 --> 00:00:59,140
from which a tweet was sent.

19
00:00:59,140 --> 00:01:03,440
If you want, as a user of
twitter, you can opt in

20
00:01:03,440 --> 00:01:06,280
to allowing twitter to
track your locations

21
00:01:06,280 --> 00:01:08,910
and some people do choose to do that.

22
00:01:08,910 --> 00:01:11,950
However, even though most tweets don't

23
00:01:11,950 --> 00:01:14,850
contain precise location information

24
00:01:14,850 --> 00:01:19,190
they do typically include
the user's home location

25
00:01:19,190 --> 00:01:22,710
however, that may be a fake location

26
00:01:22,710 --> 00:01:26,080
or a fictitious location,
it may be invalid

27
00:01:26,080 --> 00:01:27,840
depending on what they typed in

28
00:01:27,840 --> 00:01:30,690
when they set up their user account.

29
00:01:30,690 --> 00:01:34,870
So, what we're going to
do is grab those locations

30
00:01:34,870 --> 00:01:37,210
the user's home locations, and we're going

31
00:01:37,210 --> 00:01:40,400
to take those and translate them into

32
00:01:40,400 --> 00:01:44,160
actual latitudes and longitudes if we can

33
00:01:44,160 --> 00:01:45,950
for any valid locations

34
00:01:45,950 --> 00:01:50,190
and then we'll use those
to plot our map markers.

35
00:01:50,190 --> 00:01:52,480
Now to do part of this
work, we're going to be

36
00:01:52,480 --> 00:01:55,410
taking advantage of the geopy library

37
00:01:55,410 --> 00:01:59,520
which we had you install
in early on in this lesson

38
00:01:59,520 --> 00:02:02,820
at the same time that you
were installing Tweepy

39
00:02:02,820 --> 00:02:06,930
and in the geopy library,
it provides geocoding

40
00:02:06,930 --> 00:02:08,820
capabilities which is the ability to

41
00:02:08,820 --> 00:02:11,780
take a location and get the corresponding

42
00:02:11,780 --> 00:02:14,840
latitude and longitude in the world.

43
00:02:14,840 --> 00:02:18,430
Geopy has support for dozens of different

44
00:02:18,430 --> 00:02:22,240
geocoding services, many
of those geocoding services

45
00:02:22,240 --> 00:02:25,890
have either free tiers or lite tiers

46
00:02:25,890 --> 00:02:28,050
with rate limits on them

47
00:02:28,050 --> 00:02:31,200
and for our purposes, we chose to use

48
00:02:31,200 --> 00:02:36,200
the OpenMapQuest geocoding
service via the geopy library.

49
00:02:37,790 --> 00:02:40,040
Now, as you'll see,
we're going to be able to

50
00:02:40,040 --> 00:02:42,020
take a location such as Boston, MA

51
00:02:43,070 --> 00:02:45,360
turn it into a latitude and longitude

52
00:02:45,360 --> 00:02:48,040
which are numeric values that we will use

53
00:02:48,040 --> 00:02:50,080
for plotting on the maps.

54
00:02:50,080 --> 00:02:53,340
The OpenMapQuest geocoding
API gives you the ability

55
00:02:53,340 --> 00:02:56,970
to make up to 15 thousand
transactions per month

56
00:02:56,970 --> 00:03:00,040
on their free tier, and you can sign up

57
00:03:00,040 --> 00:03:03,470
for OpenMapQuest at this location.

58
00:03:03,470 --> 00:03:06,260
So, once you go to that website

59
00:03:06,260 --> 00:03:09,230
in the upper right corner of
the page at the moment, anyway

60
00:03:09,230 --> 00:03:11,760
is where you'll see the sign up link

61
00:03:11,760 --> 00:03:14,700
so you'll wanna go ahead and sign up

62
00:03:14,700 --> 00:03:17,190
and you'll have to fill
out your information

63
00:03:17,190 --> 00:03:19,960
and once you're signed
up, it will log you in

64
00:03:19,960 --> 00:03:23,500
and it will take you to this web page

65
00:03:23,500 --> 00:03:25,260
and If it doesn't you can enter

66
00:03:25,260 --> 00:03:29,470
this web page on your
own and on that web page

67
00:03:29,470 --> 00:03:34,470
you'll be able to create new
keys for accessing this API.

68
00:03:34,990 --> 00:03:38,460
So I'm gonna go ahead and
pull up my logged in account

69
00:03:38,460 --> 00:03:40,278
and you can see I've already created

70
00:03:40,278 --> 00:03:43,620
a DeitelTest app, a similar concept to

71
00:03:43,620 --> 00:03:45,820
what we did with creating a Twitter app

72
00:03:45,820 --> 00:03:47,890
I won't click on this because then it will

73
00:03:47,890 --> 00:03:50,750
reveal my private key,
but what you're going to

74
00:03:50,750 --> 00:03:53,490
do here is create a new key

75
00:03:53,490 --> 00:03:56,120
and so I'll pull up that page briefly

76
00:03:56,120 --> 00:03:59,010
all you need to specify is the app name

77
00:03:59,010 --> 00:04:02,010
we don't use this
concept of a callback URL

78
00:04:02,010 --> 00:04:03,350
in any of our apps.

79
00:04:03,350 --> 00:04:05,120
So, you'll specify the name and click

80
00:04:05,120 --> 00:04:07,940
create app, once you do
that, you'll be taken

81
00:04:07,940 --> 00:04:09,810
to a page where you can manage

82
00:04:09,810 --> 00:04:12,410
the app's keys and in our case

83
00:04:12,410 --> 00:04:15,170
I could click on this to
show you what that looks like

84
00:04:15,170 --> 00:04:18,380
but, basically, what
you're going to need to do

85
00:04:18,380 --> 00:04:22,270
is grab the consumer key value

86
00:04:22,270 --> 00:04:25,800
and you're going to
replace in our keys.py file

87
00:04:25,800 --> 00:04:30,800
the YourKeyHere value for
the mapquest_key variable.

88
00:04:31,400 --> 00:04:34,650
So you'll wanna go ahead and open keys.py

89
00:04:34,650 --> 00:04:37,150
like you did earlier today or earlier

90
00:04:37,150 --> 00:04:39,010
whenever you were watching the videos

91
00:04:39,010 --> 00:04:43,440
and you'll replace your
key in the keys.py file

92
00:04:43,440 --> 00:04:47,230
or you'll put your key in the keys.py file

93
00:04:47,230 --> 00:04:50,840
make sure you save the
file before you import it

94
00:04:50,840 --> 00:04:55,840
into the IPython session or
into any code that requires it.

95
00:04:57,410 --> 00:05:00,410
Now, in addition, we're
going to be using another

96
00:05:00,410 --> 00:05:02,780
library which you have not installed yet.

97
00:05:02,780 --> 00:05:05,580
It's the Folium library,
let me pull up its

98
00:05:05,580 --> 00:05:08,120
GitHub page here, so on this page

99
00:05:08,120 --> 00:05:11,120
you'll find the GitHub information

100
00:05:11,120 --> 00:05:14,010
and installation instructions

101
00:05:14,010 --> 00:05:17,160
but basically the folium library is

102
00:05:17,160 --> 00:05:21,727
a python library that takes
advantage of Leaflet.js

103
00:05:23,510 --> 00:05:25,540
which is a JavaScript mapping library

104
00:05:25,540 --> 00:05:28,650
for creating maps in
the context of webpages

105
00:05:29,670 --> 00:05:33,050
so we'll be using this to create our maps.

106
00:05:33,050 --> 00:05:36,020
As you'll see we will output an HTML file

107
00:05:36,020 --> 00:05:38,800
that actually contains the mapping code

108
00:05:38,800 --> 00:05:42,010
and when you open that HTML
file in your web browser

109
00:05:42,010 --> 00:05:44,230
you'll be presented with the interactive

110
00:05:44,230 --> 00:05:46,470
map that we're going to create.

111
00:05:46,470 --> 00:05:49,400
So, before you execute any
of the code in this example

112
00:05:49,400 --> 00:05:52,020
you will want to make
sure that you install

113
00:05:52,020 --> 00:05:55,210
the folium library and
it will also install

114
00:05:55,210 --> 00:06:00,170
as part of that, the support
for Leaflet.js as well.

115
00:06:00,170 --> 00:06:04,970
Now, Leaflet.js uses open
source maps by default

116
00:06:04,970 --> 00:06:09,210
specifically, it uses maps
from the OpenStreetMap.org

117
00:06:09,210 --> 00:06:12,500
website, and they do require us to display

118
00:06:12,500 --> 00:06:16,170
the copyright notice that
you see here on the screen

119
00:06:16,170 --> 00:06:20,350
so all of the maps are
copyright to the contributors

120
00:06:20,350 --> 00:06:24,303
that created those maps
so, we're just making it-

121
00:06:25,670 --> 00:06:27,060
making you - excuse me

122
00:06:27,060 --> 00:06:30,710
aware of the fact that these
maps are under copyright

123
00:06:30,710 --> 00:06:33,010
so please be careful when using

124
00:06:33,010 --> 00:06:35,950
them in your own applications.

125
00:06:35,950 --> 00:06:38,380
So with that said, the next thing that

126
00:06:38,380 --> 00:06:40,030
we're going to do in the next video

127
00:06:40,030 --> 00:06:44,820
is start the setup for
creating this application

128
00:06:44,820 --> 00:06:47,040
that's going to stream a bunch of tweets

129
00:06:47,040 --> 00:06:50,860
and translate their
locations into latitudes

130
00:06:50,860 --> 00:06:54,213
and longitudes so that we
can plot them on a map.