1
00:00:00,520 --> 00:00:04,980
Now let's talk about the things
our users want to turn on.

2
00:00:04,980 --> 00:00:09,000
An entity in Watson is
used to clarify an intent.

3
00:00:09,000 --> 00:00:15,415
So in this example, it could be radio,
headlights, basically appliances.

4
00:00:16,620 --> 00:00:18,760
Let's go right ahead, Ben,

5
00:00:18,760 --> 00:00:23,690
and let's create a new entity
as we click the Entities tab.

6
00:00:23,690 --> 00:00:26,290
>> Okay, Create new, here we go.

7
00:00:28,110 --> 00:00:29,750
>> I like those routed bits that show up.

8
00:00:29,750 --> 00:00:32,820
You can click Got it for
it to not pop up again.

9
00:00:32,820 --> 00:00:36,450
I usually do take my time and
I read those hints and tips.

10
00:00:36,450 --> 00:00:38,710
>> Cool, yeah,
a little extra context is always helpful.

11
00:00:39,900 --> 00:00:44,220
So you may ask yourself,
well now that we are turning on something,

12
00:00:44,220 --> 00:00:49,970
that's our intention, what are the type
of things that won't be turned on?

13
00:00:49,970 --> 00:00:51,590
Appliances is as such.

14
00:00:52,670 --> 00:00:55,220
>> All right, yeah,
let's call this appliances.

15
00:00:55,220 --> 00:00:58,894
Well, as you think about appliances,
let's give it some value.

16
00:00:58,894 --> 00:01:03,105
So for example, I would think ac,
that's an appliance.

17
00:01:03,105 --> 00:01:05,550
>> And that's a good one for
some synonyms as well.

18
00:01:05,550 --> 00:01:07,740
You can also say air conditioning.

19
00:01:10,080 --> 00:01:11,890
Yeah, fans is a good one.

20
00:01:11,890 --> 00:01:13,220
Maybe even air conditioner.

21
00:01:15,220 --> 00:01:16,050
>> I like it.

22
00:01:16,050 --> 00:01:16,940
>> Cool.

23
00:01:16,940 --> 00:01:21,150
>> Let's try a few more appliances.

24
00:01:21,150 --> 00:01:24,270
Perhaps lights.

25
00:01:24,270 --> 00:01:25,120
>> Lights good, yeah.

26
00:01:26,770 --> 00:01:29,880
>> What are funky ways
that you could say lights?

27
00:01:29,880 --> 00:01:30,900
>> Well, headlights.

28
00:01:30,900 --> 00:01:31,980
>> Headlights.

29
00:01:31,980 --> 00:01:32,890
>> Bulbs.

30
00:01:32,890 --> 00:01:33,510
>> Bulbs.

31
00:01:33,510 --> 00:01:34,048
>> Yeah, good.

32
00:01:34,048 --> 00:01:36,318
Tubelights maybe.

33
00:01:38,625 --> 00:01:39,395
Cool, sounds good?

34
00:01:39,395 --> 00:01:41,140
>> Very nice.

35
00:01:41,140 --> 00:01:43,140
Let's try wipers also.

36
00:01:43,140 --> 00:01:44,110
>> Okay, wipers.

37
00:01:45,600 --> 00:01:49,948
I've heard that as wiper blades or
maybe just blades itself.

38
00:01:49,948 --> 00:01:52,265
>> Absolutely.

39
00:01:52,265 --> 00:01:54,054
>> Yeah, how about a windshield wipers.

40
00:01:54,054 --> 00:01:57,820
>> Windshield wipers works very well.

41
00:01:57,820 --> 00:02:01,455
Just for the joy of it,
let's add one more, say music.

42
00:02:01,455 --> 00:02:02,511
>> Music's a good one, yeah.

43
00:02:06,115 --> 00:02:09,461
>> And the synonyms in which you could
issue commands to your car to turn on

44
00:02:09,461 --> 00:02:10,150
music maybe.

45
00:02:10,150 --> 00:02:13,615
>> Sure, radio, songs.

46
00:02:15,415 --> 00:02:18,525
Should we even do song itself, or does
it matter between singular and plurally?

47
00:02:18,525 --> 00:02:20,375
>> In fact let's do song itself.

48
00:02:20,375 --> 00:02:21,860
>> Okay.

49
00:02:21,860 --> 00:02:28,680
>> There we're Ben, an interesting place
to explore would be the system entities.

50
00:02:28,680 --> 00:02:30,560
Now this is fantastic.

51
00:02:30,560 --> 00:02:35,180
IBM has provided out of the box
a number of system entities that

52
00:02:35,180 --> 00:02:38,442
spares us from going through
the trouble building these.

53
00:02:38,442 --> 00:02:39,650
>> Yeah, that's very helpful.

54
00:02:39,650 --> 00:02:42,260
>> Location, time and date.

55
00:02:42,260 --> 00:02:46,750
We might find in some of our applications
that we may want to turn on a certain

56
00:02:46,750 --> 00:02:50,690
system entity, especially,
or perhaps with location.

57
00:02:50,690 --> 00:02:55,715
If we want to ask how is the weather,
and your browser is aware,

58
00:02:55,715 --> 00:03:03,160
geospatially aware, we would probably
turn on the location system entity.

59
00:03:03,160 --> 00:03:05,940
>> Cool, and we can turn those on,
it looks like individually, or

60
00:03:05,940 --> 00:03:07,630
we can turn all of them on if we want to.

61
00:03:07,630 --> 00:03:10,540
>> Correct.
>> Now before we try this out, Armen,

62
00:03:10,540 --> 00:03:15,330
can I ask, back in the entity we just
created, there was this checkbox,

63
00:03:15,330 --> 00:03:18,630
I got it here, for fuzzy matching, right?

64
00:03:18,630 --> 00:03:21,500
So here it says that it's in beta and
it's currently off.

65
00:03:21,500 --> 00:03:23,775
Is this something that we should turn on?

66
00:03:23,775 --> 00:03:26,265
>> I've always had great fun
with the fuzzy matching.

67
00:03:26,265 --> 00:03:31,030
[COUGH] Frankly I think it
helps more than it hinders,

68
00:03:31,030 --> 00:03:35,220
although it might take a moment for
the system to build.

69
00:03:35,220 --> 00:03:36,920
Let's do a fun experiment.

70
00:03:36,920 --> 00:03:38,270
Let's turn that on.

71
00:03:38,270 --> 00:03:39,490
>> Okay.

72
00:03:39,490 --> 00:03:45,390
>> And let's add another synonym
to say any one of these.

73
00:03:45,390 --> 00:03:51,250
To music, perhaps or-
>> Okay, let's call it a music player.

74
00:03:51,250 --> 00:03:51,950
>> Music player.

75
00:03:54,870 --> 00:03:58,110
Now then do this, click the Chat icon.

76
00:03:59,390 --> 00:04:01,230
Notice what's in this tray.

77
00:04:01,230 --> 00:04:02,740
>> Okay, here we go.

78
00:04:02,740 --> 00:04:05,340
>> This will take just a few seconds but

79
00:04:05,340 --> 00:04:10,880
a lot of activities taking place
on quite powerful servers.

80
00:04:10,880 --> 00:04:15,580
Machine learning technologies right now
are under way at this very moment, and

81
00:04:15,580 --> 00:04:19,386
they are building new patterns
based on these intents and

82
00:04:19,386 --> 00:04:21,500
entities that we introduced to the system.

83
00:04:21,500 --> 00:04:24,140
>> And I'm glad you brought that up
because this is the kind of thing

84
00:04:24,140 --> 00:04:26,980
whenever we make changes when
we're adding new entities or

85
00:04:26,980 --> 00:04:30,840
editing existing ones, we wanna make
sure that Watson has a few moments

86
00:04:30,840 --> 00:04:33,930
to train up on the changes before
we try and test anything out.

87
00:04:33,930 --> 00:04:36,470
Because otherwise,
we might not see the changes right away.

88
00:04:36,470 --> 00:04:38,790
>> I'm so glad you mentioned that,
Ben, absolutely true.

89
00:04:38,790 --> 00:04:42,475
Yes, that bit of a patience goes
a long way with cognitive systems.

90
00:04:44,058 --> 00:04:47,380
>> Okay cool, it finished,
which means we can test it out.

91
00:04:47,380 --> 00:04:51,320
All right, so I'm gonna clear my chat log
again with a refresh,although it is live.

92
00:04:51,320 --> 00:04:53,570
You can start hitting it right away.

93
00:04:53,570 --> 00:04:56,333
So let's try it with turn on the radio.

94
00:04:58,302 --> 00:05:01,928
And over here, if we scroll down,
we have matched the intent, once again,

95
00:05:01,928 --> 00:05:03,670
turn on with a high confidence.

96
00:05:03,670 --> 00:05:06,660
And here in the entities,
we have matched the appliances and

97
00:05:06,660 --> 00:05:08,898
it has gotten a value here,
the music value.

98
00:05:08,898 --> 00:05:11,540
All right, cool, so it did match it.

99
00:05:11,540 --> 00:05:13,030
Let's try one of our synonyms.

100
00:05:13,030 --> 00:05:14,879
Let's say play rock songs.

101
00:05:17,226 --> 00:05:22,927
And the same thing, we have matched
the intent, here is the appliance,

102
00:05:22,927 --> 00:05:27,136
and it has the input text
here as play rock songs.

103
00:05:27,136 --> 00:05:30,760
But it's mapped the value to
music because again it knows that

104
00:05:30,760 --> 00:05:32,470
songs is a synonym for music.

105
00:05:32,470 --> 00:05:32,970
Really cool.

106
00:05:34,105 --> 00:05:36,580
Armen, I'd like to try
another experiment here.

107
00:05:36,580 --> 00:05:38,330
I never had a card with a faucet, so

108
00:05:38,330 --> 00:05:42,730
I'm gonna see what happens if
I say switch on the faucet.

109
00:05:42,730 --> 00:05:45,000
>> I love this experiment, sure.

110
00:05:45,000 --> 00:05:48,830
>> All right, so
in this case there is no entity,

111
00:05:48,830 --> 00:05:50,250
it has no idea what we're asking about.

112
00:05:50,250 --> 00:05:52,400
So the entities is blank and

113
00:05:52,400 --> 00:05:55,590
it's worth investigating here that
we get some warnings in the output.

114
00:05:55,590 --> 00:05:58,160
It says no dialogue node matched for
the input.

115
00:05:58,160 --> 00:06:00,520
And that's because we haven't
taken that third step yet.

116
00:06:00,520 --> 00:06:01,720
We haven't created the dialogue.

117
00:06:01,720 --> 00:06:03,040
>> Very true.

118
00:06:03,040 --> 00:06:06,930
>> Cool, so at this point our Chatbot is
able to understand what we want to do in

119
00:06:06,930 --> 00:06:08,195
relation to turning things on.

120
00:06:08,195 --> 00:06:12,617
Now we could take some kind of action if
our Chatbot system here were hooked up to

121
00:06:12,617 --> 00:06:16,168
an actual automobile,
we could interact with the hardware and

122
00:06:16,168 --> 00:06:17,718
turn these appliances on.

123
00:06:17,718 --> 00:06:20,860
In this web based example,
we won't turn anything on and off.

124
00:06:20,860 --> 00:06:24,290
But we still need to have our
Chatbot reply to the user.

125
00:06:24,290 --> 00:06:28,360
The user will want to know if their intent
was understood, and if anything happened.

126
00:06:28,360 --> 00:06:29,200
Let's take a short break and

127
00:06:29,200 --> 00:06:32,120
then we'll add the final artifact
to the service, a dialogue