1
00:00:01,010 --> 00:00:02,530
- [Instructor] Now there are far more

2
00:00:02,530 --> 00:00:06,240
additional tools and
capabilities in IBM Watson

3
00:00:06,240 --> 00:00:09,460
and in the IBM Cloud than
what I have talked about

4
00:00:09,460 --> 00:00:13,270
here so far and beyond
what I'll demonstrate

5
00:00:13,270 --> 00:00:15,110
in our case study that's coming up

6
00:00:15,110 --> 00:00:18,140
a couple of videos from now,
but I did want to take a moment

7
00:00:18,140 --> 00:00:20,620
to point out a few
additional tools and services

8
00:00:20,620 --> 00:00:23,490
that you may find handy if you decide

9
00:00:23,490 --> 00:00:26,560
to work with IBM Watson
for developing apps

10
00:00:26,560 --> 00:00:28,240
for your organization.

11
00:00:28,240 --> 00:00:30,920
So, a relatively new
tool that they've added

12
00:00:30,920 --> 00:00:34,270
is the Watson Studio,
which is a convenient way

13
00:00:34,270 --> 00:00:36,380
to both create and manage the projects

14
00:00:36,380 --> 00:00:39,647
that you're working on using
the Watson capabilities

15
00:00:39,647 --> 00:00:42,040
and the IBM Cloud capabilities.

16
00:00:42,040 --> 00:00:44,180
There's a whole bunch of
different things you can do

17
00:00:44,180 --> 00:00:46,350
with Watson Studio such as

18
00:00:46,350 --> 00:00:49,380
getting your data ready for
analysis in the first place,

19
00:00:49,380 --> 00:00:51,470
which is one of the key aspects

20
00:00:51,470 --> 00:00:54,300
of performing a data science study.

21
00:00:54,300 --> 00:00:56,070
You can create Jupyter Notebooks,

22
00:00:56,070 --> 00:01:00,250
which are a really convenient
way to manage both code

23
00:01:00,250 --> 00:01:04,830
and text and images and audio and data,

24
00:01:04,830 --> 00:01:06,450
all that get wrapped together

25
00:01:06,450 --> 00:01:09,800
to perform a complete data science study.

26
00:01:09,800 --> 00:01:12,980
You also have the ability
to create and train models

27
00:01:12,980 --> 00:01:15,270
and use them with the various machine

28
00:01:15,270 --> 00:01:18,490
and deep-learning capabilities
that Watson provides,

29
00:01:18,490 --> 00:01:20,460
and in the next couple of lessons,

30
00:01:20,460 --> 00:01:22,800
we're going to be talking
about machine learning

31
00:01:22,800 --> 00:01:25,320
and deep learning and
we're actually going to be

32
00:01:25,320 --> 00:01:28,600
creating and training models using various

33
00:01:28,600 --> 00:01:32,470
key Python AI libraries as well.

34
00:01:32,470 --> 00:01:35,380
So, one of the things that
Watson Studio provides

35
00:01:35,380 --> 00:01:37,510
is a bunch of preconfigured projects

36
00:01:37,510 --> 00:01:39,780
for typical Watson workflows.

37
00:01:39,780 --> 00:01:42,120
So if you're interested
in checking that out,

38
00:01:42,120 --> 00:01:43,320
take a look at this URL.

39
00:01:44,520 --> 00:01:47,990
Separately, they have the
Knowledge Studio tool,

40
00:01:47,990 --> 00:01:52,110
and most of the web services
that I've shown you so far,

41
00:01:52,110 --> 00:01:54,680
in fact all of the web services
that I've shown you so far

42
00:01:54,680 --> 00:01:57,180
through the demos and
the ones that we'll use

43
00:01:57,180 --> 00:02:01,000
later in this lesson,
have predefined models

44
00:02:01,000 --> 00:02:04,670
that allow us to perform
incredibly powerful tasks

45
00:02:04,670 --> 00:02:07,340
with minimal numbers of lines of code.

46
00:02:07,340 --> 00:02:11,320
So as you are going to see in
our case study in this lesson,

47
00:02:11,320 --> 00:02:13,600
we're going to be able to take advantage

48
00:02:13,600 --> 00:02:17,450
of predefined models that
IBM has already trained

49
00:02:17,450 --> 00:02:22,210
to do things like take
speech and turn it into text

50
00:02:22,210 --> 00:02:25,540
and we're going to do that
both in English and Spanish.

51
00:02:25,540 --> 00:02:28,960
We're going to demonstrate
translating between languages,

52
00:02:28,960 --> 00:02:32,330
going from English to Spanish
and from Spanish to English,

53
00:02:32,330 --> 00:02:34,680
and they have pretrained
models for doing that.

54
00:02:34,680 --> 00:02:38,160
And we're going to take text
in both Spanish and English

55
00:02:38,160 --> 00:02:41,040
and convert it back into speech as well.

56
00:02:41,040 --> 00:02:44,080
And again, they have predefined
models for doing that.

57
00:02:44,080 --> 00:02:45,810
So if you had to go and create

58
00:02:45,810 --> 00:02:47,920
those models from scratch yourself,

59
00:02:47,920 --> 00:02:51,440
that could take enormous
amounts of time and data

60
00:02:51,440 --> 00:02:55,190
to get it right, but IBM
already has that for you.

61
00:02:55,190 --> 00:02:59,440
Now separately, they don't
know in advance every single

62
00:02:59,440 --> 00:03:02,570
industry and application,
so they also provide this

63
00:03:02,570 --> 00:03:05,860
Knowledge Studio where
you can train Watson

64
00:03:05,860 --> 00:03:09,900
and create models that are
specific to the applications

65
00:03:09,900 --> 00:03:11,670
you want to create.

66
00:03:11,670 --> 00:03:14,010
They also have a more powerful tool

67
00:03:14,010 --> 00:03:17,190
called the Knowledge Catalog
which is for enterprises.

68
00:03:17,190 --> 00:03:20,900
And this gives you the ability
to manage all of the data

69
00:03:20,900 --> 00:03:22,550
across your application,

70
00:03:22,550 --> 00:03:25,210
provides various security capabilities,

71
00:03:25,210 --> 00:03:27,550
but most importantly they claim

72
00:03:27,550 --> 00:03:30,660
to support over 100
different data cleaning

73
00:03:30,660 --> 00:03:32,870
and data wrangling operations,

74
00:03:32,870 --> 00:03:34,490
which are two of the key steps

75
00:03:34,490 --> 00:03:37,180
in preparing data for analysis.

76
00:03:37,180 --> 00:03:40,760
So, if you have a massive
amount of organizational data

77
00:03:40,760 --> 00:03:43,260
and information that you
would like to analyze,

78
00:03:43,260 --> 00:03:46,110
you might take a look at
the Knowledge Catalog.

79
00:03:46,110 --> 00:03:49,710
And finally, and this one
is kind of interesting to me

80
00:03:49,710 --> 00:03:51,880
because of what it provides,

81
00:03:51,880 --> 00:03:55,360
Cognos Analytics is a
relatively new tool as well.

82
00:03:55,360 --> 00:03:58,760
It came out right as we
were completing our book,

83
00:03:58,760 --> 00:04:01,010
so we didn't get a chance to study it

84
00:04:01,010 --> 00:04:02,820
and put it into our books.

85
00:04:02,820 --> 00:04:05,970
We intend to do that
for the next editions.

86
00:04:05,970 --> 00:04:09,580
But it provides both
artificial intelligence

87
00:04:09,580 --> 00:04:11,860
and machine-learning capabilities

88
00:04:11,860 --> 00:04:13,800
that enable you to look at your data

89
00:04:13,800 --> 00:04:17,040
and discover information
and visualize information

90
00:04:17,040 --> 00:04:19,510
without you doing any programming at all.

91
00:04:19,510 --> 00:04:23,440
And as we've followed up
on the writing of our books

92
00:04:23,440 --> 00:04:25,380
and started to record these videos,

93
00:04:25,380 --> 00:04:29,150
almost on a daily basis
now we're seeing articles

94
00:04:29,150 --> 00:04:32,820
about capabilities in
artificial intelligence

95
00:04:32,820 --> 00:04:36,570
and machine learning
that are fully automated

96
00:04:36,570 --> 00:04:40,330
and taking advantage of the
power of the computing systems

97
00:04:40,330 --> 00:04:42,620
that are out there and the
massive amounts of data

98
00:04:42,620 --> 00:04:45,350
that are out there to help you perform

99
00:04:45,350 --> 00:04:47,990
incredible tasks with no code at all,

100
00:04:47,990 --> 00:04:52,650
or very, very minimal code
in some cases as well.

101
00:04:52,650 --> 00:04:54,590
So one of the things that
I thought was interesting

102
00:04:54,590 --> 00:04:56,210
about the Cognos Analytics is

103
00:04:56,210 --> 00:04:58,210
that they have a
natural-language interface

104
00:04:58,210 --> 00:05:02,620
where you can talk to it, and
it will answer your questions

105
00:05:02,620 --> 00:05:06,673
based on what it has learned
from your organization's data.