1
00:00:00,380 --> 00:00:03,830
So at this point, Armin, I could imagine
that these dialogs can get pretty complex,

2
00:00:03,830 --> 00:00:04,370
pretty quickly.

3
00:00:05,590 --> 00:00:08,980
>> Yes, they can,
as complex as you need them to.

4
00:00:08,980 --> 00:00:12,760
You can build lots of intents,
recognize lots of entities, and

5
00:00:12,760 --> 00:00:16,526
control the dialog with lots of nodes and
conditions.

6
00:00:16,526 --> 00:00:17,400
>> In this car chatbot,

7
00:00:17,400 --> 00:00:20,720
we have a lot more of those things
built out in a more detailed example.

8
00:00:20,720 --> 00:00:24,050
Let's take a look to see how
a more complex bot looks, and

9
00:00:24,050 --> 00:00:26,060
how we can copy some of it for
our own bot.

10
00:00:27,640 --> 00:00:30,800
So first you'll need to switch
back to your Workspaces page.

11
00:00:30,800 --> 00:00:34,960
We can expand this menu on the left,
and click Back to workspaces.

12
00:00:34,960 --> 00:00:37,910
And then this is the Car Dashboard
sample that was created for

13
00:00:37,910 --> 00:00:40,070
us when we created this workspace.

14
00:00:40,070 --> 00:00:45,560
So if we click on this, we get a little
note that it's being freshened up for us.

15
00:00:45,560 --> 00:00:48,750
And here we can see lots
of different intents.

16
00:00:48,750 --> 00:00:55,018
And you can see on the right that these
have lots of different samples built in.

17
00:00:55,018 --> 00:00:58,637
So this #capabilities one has
over a hundred different samples

18
00:00:58,637 --> 00:01:01,860
about asking what kind of
capabilities the chatbot has.

19
00:01:03,010 --> 00:01:05,230
Same thing if we switch to the entities.

20
00:01:05,230 --> 00:01:08,940
We can see there are a number of different
entities with lots of values and

21
00:01:08,940 --> 00:01:12,120
synonyms, and a more complex dialog.

22
00:01:12,120 --> 00:01:17,460
So here we have an initializer, much like
the welcome node that we saw in our own.

23
00:01:17,460 --> 00:01:22,570
And then there's #turn_off,
there's a #turn_up command here, and

24
00:01:22,570 --> 00:01:26,600
different nodes that respond to
different intents in this chatbot.

25
00:01:28,350 --> 00:01:30,610
Just like our own,
we can test this sample out.

26
00:01:30,610 --> 00:01:36,246
We can say things like hi,
what can you do?

27
00:01:36,246 --> 00:01:39,010
And the bot tells us that it's
trained to turn things on or

28
00:01:39,010 --> 00:01:40,640
off in the car dashboard, cool.

29
00:01:40,640 --> 00:01:44,350
So we can say yeah, find a nearby gas

30
00:01:44,350 --> 00:01:49,310
station, and
it says there's some gas stations nearby.

31
00:01:49,310 --> 00:01:52,860
This is very similar to the sample
that Arman walked us through earlier.

32
00:01:53,990 --> 00:01:56,860
So one very cool thing that we can
do is export these intents and

33
00:01:56,860 --> 00:02:00,100
entities from this workspace into another.

34
00:02:00,100 --> 00:02:02,170
Let's try it with the capabilities.

35
00:02:02,170 --> 00:02:07,910
So I'm going to switch back to Intents,
and I want to click on Export right here.

36
00:02:07,910 --> 00:02:11,200
Then I can select everything
that I want to export,

37
00:02:11,200 --> 00:02:14,520
I'll just do #capabilities,
give it a check box.

38
00:02:14,520 --> 00:02:17,530
I'm gonna close the tester over here,
and click Export.

39
00:02:18,940 --> 00:02:22,320
And it looks like that was exported
in comma separated values.

40
00:02:22,320 --> 00:02:23,885
Let's take a look at how that looks.

41
00:02:23,885 --> 00:02:27,980
And yeah,
it's just a huge list of different

42
00:02:27,980 --> 00:02:33,045
sample things that people might say to
understand capabilities, interesting.

43
00:02:33,045 --> 00:02:34,890
Right, I'm gonna close this.

44
00:02:34,890 --> 00:02:38,950
And now we also want to export
the corresponding entities.

45
00:02:38,950 --> 00:02:45,520
So we have @capabilities_subject,
I guess let's take that one.

46
00:02:45,520 --> 00:02:50,798
Export @capabilities_subject,
we'll get another CSV.

47
00:02:50,798 --> 00:02:53,300
And now we can switch back to our own app.

48
00:02:53,300 --> 00:02:58,430
So I'm gonna go back to Workspaces,
click on the Treehouse Car Chatbot, and

49
00:02:58,430 --> 00:02:59,390
we can just import the data.

50
00:02:59,390 --> 00:03:03,300
Again, this is just a way to show
how you might wanna do data entry

51
00:03:03,300 --> 00:03:04,360
in a different format.

52
00:03:04,360 --> 00:03:08,750
You could have somebody type a bunch
of samples in a comma separated value,

53
00:03:08,750 --> 00:03:12,500
some kind of text document, and
then import them just like this.

54
00:03:12,500 --> 00:03:14,600
So if I click Import, I can choose a file.

55
00:03:15,970 --> 00:03:19,240
And I'll go to Downloads and
I'll get the most recent file.

56
00:03:20,800 --> 00:03:24,691
And you can see I've got multiple
copies because I tried this before, but

57
00:03:24,691 --> 00:03:26,260
I want the intent first.

58
00:03:26,260 --> 00:03:31,660
So I'm gonna grab the intents file,
click Open and Import.

59
00:03:31,660 --> 00:03:34,750
After a moment we get
a successful window here,

60
00:03:34,750 --> 00:03:37,320
one new intent with 120 examples added.

61
00:03:37,320 --> 00:03:39,160
And if I click on Done,

62
00:03:39,160 --> 00:03:43,270
you can see all the different capabilities
similar to what we saw in the sample.

63
00:03:44,410 --> 00:03:47,300
And for completeness,
let's do the same thing for entities.

64
00:03:47,300 --> 00:03:51,720
We will Import, Choose a file,
grab those entities,

65
00:03:51,720 --> 00:03:53,880
which you can see in the file name here.

66
00:03:53,880 --> 00:03:57,666
Click on Open, Import,
get the success message, and

67
00:03:57,666 --> 00:04:00,558
we get those @capabilities_subject.

68
00:04:00,558 --> 00:04:03,850
Now, if we wanted to do dialog as well,
then it's a bit further.

69
00:04:03,850 --> 00:04:07,110
We need to export and
import the whole workspace, right Armin?

70
00:04:07,110 --> 00:04:13,280
>> Very true, at that point,
we import a JSON file that would contain,

71
00:04:13,280 --> 00:04:18,100
most likely, intent entities and
a dialog format.

72
00:04:18,100 --> 00:04:22,010
And that's a good point,
we can do it for the whole workspace.

73
00:04:22,010 --> 00:04:25,490
We could also create a node,
I'm just gonna do this for a sample here.

74
00:04:25,490 --> 00:04:29,449
And here in the response,
we haven't looked at this yet, but

75
00:04:29,449 --> 00:04:32,203
we can switch over to
the JSON editor view.

76
00:04:32,203 --> 00:04:36,112
And we could pick and choose Individual
snippets of JSON, if we wanted to, for

77
00:04:36,112 --> 00:04:38,320
a specific node, is that correct?

78
00:04:38,320 --> 00:04:39,390
>> Very true, very true.

79
00:04:39,390 --> 00:04:43,843
In fact, we use Span language
in this JSON format to

80
00:04:43,843 --> 00:04:47,860
build rather complex dialog structures.

81
00:04:47,860 --> 00:04:48,470
>> Cool, very cool.

82
00:04:50,450 --> 00:04:54,614
Okay, so I'm going to get rid of this
node, because it's not doing anything,

83
00:04:54,614 --> 00:04:55,370
Delete here.

84
00:04:55,370 --> 00:04:58,400
[LAUGH] It gives me a nice warning,
yes, I do wanna delete it.

85
00:04:58,400 --> 00:05:00,570
And I'm gonna do the same
double test as before.

86
00:05:00,570 --> 00:05:05,040
First, I'm gonna open up and
Watson looks like it's done training.

87
00:05:05,040 --> 00:05:08,492
So I can ask this bot, what can you do?

88
00:05:08,492 --> 00:05:09,840
And I should get a response.

89
00:05:11,180 --> 00:05:14,122
This is because I didn't setup
an actual dialog for it.

90
00:05:14,122 --> 00:05:15,348
>> Very true.
>> I would need to go ahead and

91
00:05:15,348 --> 00:05:16,590
manually do it.

92
00:05:16,590 --> 00:05:21,820
So let's do that real quick, actually,
just to make this example complete.

93
00:05:21,820 --> 00:05:26,880
We'll name this #capabilities, just
like we did before to match the intent.

94
00:05:28,150 --> 00:05:32,880
If the bot matches the new #capabilities
intent, there it is in auto complete,

95
00:05:32,880 --> 00:05:39,516
we'll enter a response like,
I can turn on things like the radio,

96
00:05:39,516 --> 00:05:45,190
heater, AC, and lights.

97
00:05:45,190 --> 00:05:47,150
Okay, now I'm gonna close this.

98
00:05:49,050 --> 00:05:52,010
And let's see, does Watson need
to train itself on this one or

99
00:05:52,010 --> 00:05:53,190
is that a pretty quick change?

100
00:05:53,190 --> 00:05:56,080
Looks like it was a pretty quick change,
so let's try it again.

101
00:05:57,250 --> 00:06:01,367
What can you do?

102
00:06:01,367 --> 00:06:03,583
And now, sure enough,
it tells us it can turn things on and

103
00:06:03,583 --> 00:06:05,480
off like the radio, etc.

104
00:06:05,480 --> 00:06:07,460
>> Interesting to note, Ben,

105
00:06:07,460 --> 00:06:12,955
the Watson training actually takes place
when we build intent and entities.

106
00:06:12,955 --> 00:06:16,170
>> Okay.
>> Those are the two constructs that tap

107
00:06:16,170 --> 00:06:18,230
into machine learning technologies.

108
00:06:18,230 --> 00:06:21,249
The dialog really is a Boolean structure,
a yes or

109
00:06:21,249 --> 00:06:25,480
no that we can navigate through the tree.

110
00:06:25,480 --> 00:06:26,080
>> Okay, that's nice.

111
00:06:26,080 --> 00:06:28,910
So then when we were importing,
it triggered some training,

112
00:06:28,910 --> 00:06:30,100
which happened behind the scenes.

113
00:06:30,100 --> 00:06:32,880
And it was already done before we
even cared about it, which is great.

114
00:06:32,880 --> 00:06:34,350
>> Indeed.

115
00:06:34,350 --> 00:06:37,606
>> This is fantastic, Armin, thank you for
walking us through the setup.

116
00:06:37,606 --> 00:06:40,742
What do you say we take a little break,
and then we make this chatbot even more

117
00:06:40,742 --> 00:06:44,060
useful by adding in another
Watson service called Discovery?

118
00:06:44,060 --> 00:06:45,150
>> Sounds great.

119
00:06:45,150 --> 00:06:48,695
There are so many meaningful ways
to use these services together.

120
00:06:48,695 --> 00:06:51,488
But discovery is particularly useful,

121
00:06:51,488 --> 00:06:57,210
because it allows our bot to discover
new answers that we don't program.

122
00:06:57,210 --> 00:07:02,110
In fact, I often like to think of
the discovery service as a area where

123
00:07:02,110 --> 00:07:06,540
you find out things that you
don't know that you don't know.

124
00:07:06,540 --> 00:07:10,610
And I can trust that with
a conversation service, that is where

125
00:07:10,610 --> 00:07:14,890
at least you know what it is that you
don't know, and then you may ask it.

126
00:07:14,890 --> 00:07:18,530
We shall see more as
the course progresses.

127
00:07:18,530 --> 00:07:19,220
>> Sounds great.

128
00:07:19,220 --> 00:07:20,210
All right, see you shortly.