WEBVTT

1
00:00:00.120 --> 00:00:03.480
<v Speaker 1>Welcome to our deep dive into ontologies using Python.

2
00:00:03.799 --> 00:00:04.599
<v Speaker 2>Ontologies.

3
00:00:04.719 --> 00:00:08.640
<v Speaker 1>Yeah, ontologies. You know, it's a way to describe knowledge

4
00:00:09.320 --> 00:00:11.400
<v Speaker 1>in a way that machines can actually work with it.

5
00:00:11.960 --> 00:00:14.919
<v Speaker 1>So we're using this book called ontologies with Python as

6
00:00:14.960 --> 00:00:18.120
<v Speaker 1>our main source. It focuses on this Python module called

7
00:00:18.199 --> 00:00:23.440
<v Speaker 1>owlready two hmm. And I think you'll be pretty surprised

8
00:00:23.440 --> 00:00:25.640
<v Speaker 1>by what you can actually do with ontologies, like what

9
00:00:25.839 --> 00:00:30.480
<v Speaker 1>you can build websites that like understand complex data wow,

10
00:00:30.800 --> 00:00:34.840
<v Speaker 1>or even power AI that can like reason like a human. Hmm.

11
00:00:35.479 --> 00:00:38.920
<v Speaker 2>That's that's pretty impressive, it is, right, So what is

12
00:00:38.960 --> 00:00:39.640
<v Speaker 2>an ontology?

13
00:00:39.840 --> 00:00:40.840
<v Speaker 1>Yeah? What is an ontology?

14
00:00:40.880 --> 00:00:43.119
<v Speaker 2>Like, how is it different from just regular programming?

15
00:00:43.240 --> 00:00:46.079
<v Speaker 1>Well, it's it's not just data, okay, like you would

16
00:00:46.079 --> 00:00:49.240
<v Speaker 1>find in this breadsheet or a database. It's it's about

17
00:00:49.359 --> 00:00:54.159
<v Speaker 1>meaning meaning okay, and relationships just between what between things? Ok?

18
00:00:54.399 --> 00:00:57.359
<v Speaker 1>It's about like, can give me like a simple analogy?

19
00:00:57.479 --> 00:01:00.759
<v Speaker 2>Sure, imagine a family tree. It's not just a list

20
00:01:00.799 --> 00:01:03.840
<v Speaker 2>of names, right right, It shows how those people are related.

21
00:01:04.400 --> 00:01:08.439
<v Speaker 2>So you have different categories like parents and children, which

22
00:01:08.480 --> 00:01:10.680
<v Speaker 2>we call classes in ontology terms.

23
00:01:10.719 --> 00:01:11.120
<v Speaker 1>Gotcha.

24
00:01:11.239 --> 00:01:15.000
<v Speaker 2>And then you have characteristics like age or hair color,

25
00:01:15.840 --> 00:01:19.200
<v Speaker 2>and we call those properties, and then each person in

26
00:01:19.239 --> 00:01:21.519
<v Speaker 2>the tree is an individual.

27
00:01:21.799 --> 00:01:26.640
<v Speaker 1>Okay, So an ontology lets us represent this kind of information, yes,

28
00:01:27.280 --> 00:01:30.359
<v Speaker 1>in a way that a computer can understand and reason

29
00:01:30.439 --> 00:01:34.159
<v Speaker 1>with and reason with. So instead of just seeing data points, yeah,

30
00:01:34.239 --> 00:01:37.920
<v Speaker 1>the computer can actually understand that grandparent exactly is a

31
00:01:37.959 --> 00:01:42.120
<v Speaker 1>meaningful category based on the relationships. That's pretty neat, Yeah

32
00:01:42.159 --> 00:01:42.439
<v Speaker 1>it is.

33
00:01:42.519 --> 00:01:46.120
<v Speaker 2>And this ability to capture meaning is really what makes

34
00:01:46.120 --> 00:01:47.280
<v Speaker 2>ontologies powerful.

35
00:01:47.439 --> 00:01:50.799
<v Speaker 1>Okay, so how does this owl ready to come into play?

36
00:01:50.879 --> 00:01:53.560
<v Speaker 2>So a already too is like our toolkit for working

37
00:01:53.599 --> 00:01:58.040
<v Speaker 2>with ontologies and Python, and it's surprisingly easy to use, really,

38
00:01:58.200 --> 00:01:59.920
<v Speaker 2>even if you're not a Python xper.

39
00:02:00.280 --> 00:02:00.400
<v Speaker 1>Ok.

40
00:02:00.799 --> 00:02:05.359
<v Speaker 2>Lets us load existing ontologies, modify them, and even create

41
00:02:05.560 --> 00:02:07.959
<v Speaker 2>entirely new ones from scratch.

42
00:02:08.199 --> 00:02:10.719
<v Speaker 1>So it's our way to actually put these ontologies into

43
00:02:10.759 --> 00:02:13.280
<v Speaker 1>action exact. Can you give me specific example?

44
00:02:13.400 --> 00:02:17.039
<v Speaker 2>Yeah? Sure, The book uses this example of a bacteria ontology.

45
00:02:17.280 --> 00:02:17.520
<v Speaker 1>Oka.

46
00:02:17.680 --> 00:02:21.520
<v Speaker 2>It might sound simple, but it's a good illustration of

47
00:02:21.520 --> 00:02:25.280
<v Speaker 2>how even a basic ontology can be powerful. So this

48
00:02:25.360 --> 00:02:30.199
<v Speaker 2>ontology has classes like bacteria shape and grouping, and then

49
00:02:30.240 --> 00:02:35.639
<v Speaker 2>it also has properties like gram positive to describe those bacteria,

50
00:02:36.080 --> 00:02:39.879
<v Speaker 2>and then each specific type of bacteria is represented as

51
00:02:39.919 --> 00:02:40.639
<v Speaker 2>an individual.

52
00:02:40.759 --> 00:02:43.919
<v Speaker 1>So it's like we're creating like a digital fingerprint for

53
00:02:44.000 --> 00:02:47.639
<v Speaker 1>each bacteria based on its characteristics exactly. Okay, And so

54
00:02:48.240 --> 00:02:50.719
<v Speaker 1>this setup allows us to do some pretty cool things.

55
00:02:50.960 --> 00:02:54.400
<v Speaker 2>Yeah, Like we could use this ontology to identify an

56
00:02:54.479 --> 00:02:57.759
<v Speaker 2>unknown bacteria oh wow, based on its characteristics.

57
00:02:57.759 --> 00:02:59.599
<v Speaker 1>So it's like we're giving a computer the ability to

58
00:02:59.639 --> 00:03:01.039
<v Speaker 1>play bacterial detective.

59
00:03:01.159 --> 00:03:02.159
<v Speaker 2>That's a great way to put it.

60
00:03:02.280 --> 00:03:05.039
<v Speaker 1>Yeah, if we tell it the bacteria is Gram positive

61
00:03:05.199 --> 00:03:09.159
<v Speaker 1>and Rod shaped, it can like use the ontology to

62
00:03:09.240 --> 00:03:13.080
<v Speaker 1>narrow down the possibility exactly. That's pretty practical, it is, But.

63
00:03:13.000 --> 00:03:14.240
<v Speaker 2>There's even more to it.

64
00:03:14.319 --> 00:03:14.840
<v Speaker 1>Okay.

65
00:03:14.879 --> 00:03:18.680
<v Speaker 2>One of the really amazing things is this concept of reasoners.

66
00:03:18.919 --> 00:03:19.520
<v Speaker 1>Reasoners.

67
00:03:19.639 --> 00:03:21.879
<v Speaker 2>Yeah, they're like superpowered logic engines.

68
00:03:21.960 --> 00:03:22.879
<v Speaker 1>Okay, how do they work?

69
00:03:23.199 --> 00:03:26.360
<v Speaker 2>So imagine you have a detective who's really good at

70
00:03:26.400 --> 00:03:29.520
<v Speaker 2>connecting the dots. Okay, that's what a reasoner does for

71
00:03:29.599 --> 00:03:35.800
<v Speaker 2>our ontology. It can automatically classify individuals, uncover hidden relationships

72
00:03:35.800 --> 00:03:40.960
<v Speaker 2>that weren't explicitly stated, wow, and even flag inconsistencies in

73
00:03:41.080 --> 00:03:41.960
<v Speaker 2>our knowledge base.

74
00:03:42.080 --> 00:03:45.439
<v Speaker 1>So it's like the reasoner takes everything we've defined exactly

75
00:03:45.560 --> 00:03:49.159
<v Speaker 1>and starts making logical deductions and revealing insights that we

76
00:03:49.240 --> 00:03:52.400
<v Speaker 1>might not even considered precisely. That's powerful.

77
00:03:52.439 --> 00:03:57.759
<v Speaker 2>It's what truly allows ontologies to go beyond simple data storage, okay,

78
00:03:57.800 --> 00:03:59.919
<v Speaker 2>and become tools for knowledge representation.

79
00:04:01.199 --> 00:04:03.759
<v Speaker 1>So are these reasoners built into OWL ready too?

80
00:04:04.319 --> 00:04:07.159
<v Speaker 2>That's the beauty of it. It has built in support

81
00:04:07.360 --> 00:04:10.879
<v Speaker 2>for popular reasoners like Hermity and Pellet, And.

82
00:04:10.879 --> 00:04:12.360
<v Speaker 1>To be clear, those are separate programs.

83
00:04:12.560 --> 00:04:16.519
<v Speaker 2>Yes, both Hermity and Pellet are Java programs, gotcha, So

84
00:04:16.600 --> 00:04:20.439
<v Speaker 2>you'll need to have Java installed on your system to

85
00:04:20.560 --> 00:04:23.560
<v Speaker 2>use them. But don't worry. Already two makes the setup

86
00:04:23.600 --> 00:04:24.639
<v Speaker 2>pretty straightforward.

87
00:04:24.759 --> 00:04:27.279
<v Speaker 1>Okay. So once we have those set up, how do

88
00:04:27.360 --> 00:04:27.920
<v Speaker 1>we use them?

89
00:04:28.120 --> 00:04:32.040
<v Speaker 2>It's surprisingly simple, okay. Already two has this function called

90
00:04:32.120 --> 00:04:35.560
<v Speaker 2>sink reasoner. You just run that and it's like hitting

91
00:04:35.600 --> 00:04:37.360
<v Speaker 2>the go button on our logic engine.

92
00:04:37.399 --> 00:04:42.639
<v Speaker 1>And by default, Alurity two uses the Hermite reasoner that's right, okay.

93
00:04:42.920 --> 00:04:45.879
<v Speaker 2>But if you want to use Pellet there are specific

94
00:04:45.959 --> 00:04:48.279
<v Speaker 2>functions for that as well, so it gives.

95
00:04:48.120 --> 00:04:48.920
<v Speaker 1>You the flexibility.

96
00:04:49.000 --> 00:04:49.199
<v Speaker 2>Yeah.

97
00:04:49.240 --> 00:04:53.879
<v Speaker 1>So, no matter which one we choose, Already two provides

98
00:04:53.920 --> 00:04:56.560
<v Speaker 1>a simple way to put them to work exactly. This

99
00:04:56.600 --> 00:04:57.720
<v Speaker 1>is all starting to come.

100
00:04:57.600 --> 00:04:59.879
<v Speaker 2>Together, it is, isn't it? Yeah, and as we dived,

101
00:05:00.720 --> 00:05:05.079
<v Speaker 2>you'll see just how powerful this combination of ontologies already

102
00:05:05.120 --> 00:05:06.720
<v Speaker 2>to and reasoners can be.

103
00:05:07.120 --> 00:05:09.839
<v Speaker 1>Now, you mentioned something earlier about an open world assumption.

104
00:05:10.120 --> 00:05:11.879
<v Speaker 1>Oh yeah, can you explain what that means?

105
00:05:12.279 --> 00:05:15.800
<v Speaker 2>Sure? The open world assumption basically means that we acknowledge

106
00:05:15.839 --> 00:05:18.199
<v Speaker 2>there might be information okay, that we don't know.

107
00:05:18.680 --> 00:05:21.399
<v Speaker 1>So it's like saying, hey, we might be missing some

108
00:05:21.639 --> 00:05:23.079
<v Speaker 1>pieces of a puzzle.

109
00:05:23.160 --> 00:05:26.680
<v Speaker 2>It's like admitting that our knowledge is never truly complete,

110
00:05:27.120 --> 00:05:30.600
<v Speaker 2>which contrasts with the closed world assumption, okay, where we

111
00:05:30.639 --> 00:05:33.199
<v Speaker 2>assume that we know everything.

112
00:05:32.800 --> 00:05:35.439
<v Speaker 1>God is to know. So this distinction is important, yes,

113
00:05:35.639 --> 00:05:39.199
<v Speaker 1>very because it affects how the reasoners work exactly. Okay,

114
00:05:39.240 --> 00:05:43.199
<v Speaker 1>So how does the open world assumption change the way

115
00:05:43.240 --> 00:05:43.800
<v Speaker 1>they operate?

116
00:05:43.879 --> 00:05:47.319
<v Speaker 2>Well? Under the open world assumption, a reasoner can't just

117
00:05:47.399 --> 00:05:51.480
<v Speaker 2>assume something is false because it's not explicitly stated in

118
00:05:51.519 --> 00:05:55.240
<v Speaker 2>the ontology. It has to consider the possibility that the

119
00:05:55.240 --> 00:05:56.879
<v Speaker 2>information is simply missing.

120
00:05:57.199 --> 00:06:01.560
<v Speaker 1>So it's a bit more cautious. Yes, it's knowing that

121
00:06:01.600 --> 00:06:05.800
<v Speaker 1>there might be unknown factors at play. Exactly, That makes sense.

122
00:06:06.199 --> 00:06:08.240
<v Speaker 1>So how does this work in practice?

123
00:06:08.560 --> 00:06:12.319
<v Speaker 2>So already two has a function called close world that

124
00:06:12.399 --> 00:06:16.040
<v Speaker 2>lets you temporarily close the world, either for specific parts

125
00:06:16.040 --> 00:06:18.560
<v Speaker 2>of the ontology or for the whole thing.

126
00:06:18.720 --> 00:06:21.639
<v Speaker 1>So we can basically tell the reasoner to focus on

127
00:06:21.680 --> 00:06:24.600
<v Speaker 1>a particular area and treat it as if we have

128
00:06:24.639 --> 00:06:27.639
<v Speaker 1>all the information. But why would we want to do that?

129
00:06:28.079 --> 00:06:32.360
<v Speaker 2>Well, sometimes closing the world can be beneficial, Okay. It

130
00:06:32.480 --> 00:06:37.240
<v Speaker 2>lets us make more specific deductions about a particular area

131
00:06:37.360 --> 00:06:37.959
<v Speaker 2>of knowledge.

132
00:06:37.959 --> 00:06:40.879
<v Speaker 1>So it's like saying, for this specific task, yeah, let's

133
00:06:40.879 --> 00:06:43.879
<v Speaker 1>assume we have all the pieces of exactly puzzle.

134
00:06:44.000 --> 00:06:47.079
<v Speaker 2>It can be helpful in situations where we need to

135
00:06:47.120 --> 00:06:50.839
<v Speaker 2>make decisions Okay, based on a limited set of facts. Gotcha,

136
00:06:50.920 --> 00:06:55.000
<v Speaker 2>But it's a trade off between precision and the ability

137
00:06:55.120 --> 00:06:56.519
<v Speaker 2>to handle potential unknowns.

138
00:06:56.920 --> 00:06:57.800
<v Speaker 1>Okay, that makes sense.

139
00:06:57.879 --> 00:07:00.240
<v Speaker 2>It's like choosing the right tool for the job exactly.

140
00:07:00.319 --> 00:07:03.439
<v Speaker 2>Now that we've laid the groundwork with reasoners and the

141
00:07:03.480 --> 00:07:06.000
<v Speaker 2>open world assumption, yeah, let's dive into some of the

142
00:07:06.040 --> 00:07:09.000
<v Speaker 2>things that we can actually do. Okay with these reasoners

143
00:07:09.079 --> 00:07:10.959
<v Speaker 2>in OWL, ready to what.

144
00:07:10.920 --> 00:07:12.680
<v Speaker 1>Kind of tasks can they help us with.

145
00:07:13.360 --> 00:07:17.560
<v Speaker 2>One of the most common and useful applications is consistency checking.

146
00:07:17.959 --> 00:07:19.240
<v Speaker 1>Okay, consistency check.

147
00:07:19.279 --> 00:07:21.800
<v Speaker 2>It's like having a built in quality control system.

148
00:07:21.920 --> 00:07:24.480
<v Speaker 1>Okay, but what exactly does consistency mean?

149
00:07:24.759 --> 00:07:27.759
<v Speaker 2>So a consistent ontology is one where there are no

150
00:07:28.000 --> 00:07:31.079
<v Speaker 2>logical contradictions. Okay, think of it this way. If you

151
00:07:31.160 --> 00:07:34.279
<v Speaker 2>told me that squares are a type of circle, that

152
00:07:34.319 --> 00:07:35.399
<v Speaker 2>would be a contradiction.

153
00:07:35.639 --> 00:07:35.839
<v Speaker 1>Right.

154
00:07:36.199 --> 00:07:39.639
<v Speaker 2>Reasoners can help us find these kind of errors in

155
00:07:39.680 --> 00:07:41.879
<v Speaker 2>our ontologies before they cause problems.

156
00:07:41.959 --> 00:07:44.519
<v Speaker 1>Ah. So it's like making sure all the pieces of

157
00:07:44.560 --> 00:07:48.839
<v Speaker 1>our knowledge why framework fit together logically. I can see

158
00:07:48.839 --> 00:07:51.279
<v Speaker 1>how that would be crucial. Yeah, especially if we're building

159
00:07:51.319 --> 00:07:53.160
<v Speaker 1>complex systems based on this.

160
00:07:53.519 --> 00:07:56.720
<v Speaker 2>Right, inconsistencies can lead to all sorts of unexpected and

161
00:07:56.759 --> 00:07:57.800
<v Speaker 2>incorrect results.

162
00:07:57.959 --> 00:08:00.439
<v Speaker 1>Okay, so we've got consistency checking. Else.

163
00:08:00.839 --> 00:08:07.439
<v Speaker 2>Another powerful application is automatic classification. Reasoners can automatically categorize

164
00:08:07.439 --> 00:08:12.160
<v Speaker 2>individuals into the correct classes based on their properties and

165
00:08:12.199 --> 00:08:14.680
<v Speaker 2>the relationships defined in the ontology.

166
00:08:14.759 --> 00:08:18.120
<v Speaker 1>So it's like having their reasoner act as an expert librarian.

167
00:08:18.399 --> 00:08:19.920
<v Speaker 2>Yeah, that's a good analogy.

168
00:08:19.519 --> 00:08:22.720
<v Speaker 1>Who can perfectly organize all the books based on their

169
00:08:22.759 --> 00:08:23.839
<v Speaker 1>content and genre.

170
00:08:24.040 --> 00:08:27.160
<v Speaker 2>Let's go back to our bacteria ontology. If we define

171
00:08:27.160 --> 00:08:31.319
<v Speaker 2>a new bacteria with certain characteristics, the reason I can

172
00:08:31.360 --> 00:08:34.519
<v Speaker 2>automatically figure out which class.

173
00:08:34.200 --> 00:08:37.159
<v Speaker 1>It belongs to, saving us a lot of manual.

174
00:08:36.879 --> 00:08:40.519
<v Speaker 2>Effort, exactly, And I imagine this becomes even more valuable

175
00:08:40.600 --> 00:08:43.399
<v Speaker 2>as our ontology grows and becomes more complex.

176
00:08:43.519 --> 00:08:48.320
<v Speaker 1>Okay, so we can use reasoners for consistency checking, automatic classification.

177
00:08:48.879 --> 00:08:50.080
<v Speaker 1>What other tricks do they have?

178
00:08:50.360 --> 00:08:54.440
<v Speaker 2>They can also help us uncover implicit relationships. Implicit relationships

179
00:08:54.519 --> 00:08:56.799
<v Speaker 2>are explicitly stated in the ontology.

180
00:08:56.799 --> 00:08:57.759
<v Speaker 1>Can you give me an example.

181
00:08:58.039 --> 00:09:02.519
<v Speaker 2>Sure. Let's say our ontology describes family relationships. Okay, we

182
00:09:02.600 --> 00:09:05.480
<v Speaker 2>know that John is the father of Mary and Mary

183
00:09:05.519 --> 00:09:06.320
<v Speaker 2>is the mother of Peter.

184
00:09:06.840 --> 00:09:09.039
<v Speaker 1>So we have two direct relationships.

185
00:09:08.399 --> 00:09:12.279
<v Speaker 2>There, right, But there's an implied relationship that's not explicitly stated.

186
00:09:12.679 --> 00:09:16.039
<v Speaker 2>John is the grandfather of Peter. Right. A reasoner can

187
00:09:16.039 --> 00:09:19.799
<v Speaker 2>figure this out based on the existing information about parent

188
00:09:19.879 --> 00:09:21.000
<v Speaker 2>child relationships.

189
00:09:21.320 --> 00:09:23.200
<v Speaker 1>I see. So it's like the reasoner is filling in

190
00:09:23.240 --> 00:09:27.320
<v Speaker 1>the gaps in our knowledge based on logical deductions.

191
00:09:27.399 --> 00:09:27.960
<v Speaker 2>Exactly.

192
00:09:28.039 --> 00:09:29.279
<v Speaker 1>That's pretty impressive, it.

193
00:09:29.240 --> 00:09:33.120
<v Speaker 2>Is, And this ability to uncover implicit relationships can be

194
00:09:33.200 --> 00:09:39.039
<v Speaker 2>incredibly valuable in many real world scenarios. What imagine using

195
00:09:39.080 --> 00:09:44.480
<v Speaker 2>this in a medical ontology to identify potential drug interactions, huh,

196
00:09:44.840 --> 00:09:50.360
<v Speaker 2>or predict disease risks based on complex relationships between gene

197
00:09:50.360 --> 00:09:51.679
<v Speaker 2>symptoms and treatments.

198
00:09:52.000 --> 00:09:55.639
<v Speaker 1>Okay, so we're giving the computer the ability to make connections. Yes,

199
00:09:55.720 --> 00:09:57.279
<v Speaker 1>that might not be obvious to us.

200
00:09:57.159 --> 00:10:00.480
<v Speaker 2>At first glance, precisely, and this is where the true

201
00:10:00.559 --> 00:10:04.639
<v Speaker 2>power of ontologies and reasoners really shines. They help us

202
00:10:04.720 --> 00:10:10.679
<v Speaker 2>move beyond simply storing data to actually understanding the knowledge

203
00:10:10.679 --> 00:10:12.240
<v Speaker 2>that's encoded in that data.

204
00:10:12.360 --> 00:10:15.080
<v Speaker 1>This is all fascinating, But before we get too carried away,

205
00:10:15.279 --> 00:10:18.240
<v Speaker 1>I want to go back to something you mentioned earlier, annotations.

206
00:10:18.279 --> 00:10:19.399
<v Speaker 1>Can you remind me what those are?

207
00:10:19.559 --> 00:10:24.200
<v Speaker 2>Sure? So, think of annotations as extra information that we

208
00:10:24.240 --> 00:10:28.200
<v Speaker 2>can attach to different parts of our ontology. They're like

209
00:10:28.240 --> 00:10:33.759
<v Speaker 2>sticky notes that provide additional context, explanations, or even links

210
00:10:33.799 --> 00:10:35.480
<v Speaker 2>to external resources.

211
00:10:35.759 --> 00:10:38.720
<v Speaker 1>So they're like adding helpful notes to our knowledge map.

212
00:10:39.000 --> 00:10:40.279
<v Speaker 1>Can you give me some examples.

213
00:10:40.440 --> 00:10:44.559
<v Speaker 2>Absolutely, we can use them to provide easy to understand

214
00:10:44.600 --> 00:10:50.279
<v Speaker 2>descriptions for complex terms, add comments or explanations for ourselves

215
00:10:50.399 --> 00:10:53.200
<v Speaker 2>or other developers, keep track of where we got the

216
00:10:53.240 --> 00:10:59.279
<v Speaker 2>information for a specific concept, link to relevant websites, research papers,

217
00:10:59.360 --> 00:11:03.639
<v Speaker 2>or database, or even track different versions of the ontology.

218
00:11:03.799 --> 00:11:06.360
<v Speaker 1>So it's all about adding more layers of meaning exactly,

219
00:11:06.480 --> 00:11:08.919
<v Speaker 1>and making our ontology richer and more informative.

220
00:11:09.080 --> 00:11:13.759
<v Speaker 2>Yeah, and already too, makes working with annotations really straightforward.

221
00:11:14.440 --> 00:11:18.200
<v Speaker 2>You can add, access, and modify annotations using a simple syntax,

222
00:11:18.600 --> 00:11:20.600
<v Speaker 2>just like you would with any other property.

223
00:11:20.759 --> 00:11:24.159
<v Speaker 1>Now, you mentioned something about annotations being helpful for full

224
00:11:24.240 --> 00:11:27.000
<v Speaker 1>text search. Yeah, can you explain how that works?

225
00:11:27.159 --> 00:11:32.320
<v Speaker 2>So imagine you're searching for information within a vast library. Okay,

226
00:11:32.399 --> 00:11:36.639
<v Speaker 2>the more detailed the index and cataloging system, the easier

227
00:11:36.679 --> 00:11:39.519
<v Speaker 2>it is to find what you're looking for. Right, That's

228
00:11:39.519 --> 00:11:41.080
<v Speaker 2>what annotations do for ontology.

229
00:11:41.240 --> 00:11:44.919
<v Speaker 1>So the more annotations we add, the easier becomes to

230
00:11:45.080 --> 00:11:48.679
<v Speaker 1>search exactly for specific information within our ontology.

231
00:11:48.759 --> 00:11:53.919
<v Speaker 2>Precisely, search engines can index the text content of our annotations,

232
00:11:54.000 --> 00:11:57.879
<v Speaker 2>making it much easier to find relevant concepts, even if

233
00:11:57.879 --> 00:11:59.960
<v Speaker 2>they're buried deep within the ontology.

234
00:12:00.080 --> 00:12:03.399
<v Speaker 1>So it's like providing extra keywords and clues to help

235
00:12:03.440 --> 00:12:06.480
<v Speaker 1>the search engine understand and navigate our ontology.

236
00:12:06.679 --> 00:12:10.320
<v Speaker 2>Yeah, and this becomes especially important as our ontologies grow

237
00:12:10.399 --> 00:12:15.120
<v Speaker 2>larger and more complex. Annotations help ensure that our knowledge

238
00:12:15.240 --> 00:12:17.639
<v Speaker 2>remains accessible and discoverable.

239
00:12:18.200 --> 00:12:20.759
<v Speaker 1>Now, shifting gears a bit, I'm curious about something you

240
00:12:20.799 --> 00:12:25.200
<v Speaker 1>mentioned earlier. Okay, worlds an owlready two. This sounds almost

241
00:12:25.240 --> 00:12:25.919
<v Speaker 1>like science fiction.

242
00:12:26.200 --> 00:12:29.679
<v Speaker 2>It's not quite science fiction, but it's a powerful concept.

243
00:12:29.759 --> 00:12:31.399
<v Speaker 1>Okay, So what are they think of.

244
00:12:31.360 --> 00:12:35.440
<v Speaker 2>A world as a self contained environment within owl ready to? Okay,

245
00:12:35.519 --> 00:12:37.320
<v Speaker 2>it's like having a separate sandbox.

246
00:12:37.480 --> 00:12:40.360
<v Speaker 1>Okay, I like this out of a sandbox. Yeah. It

247
00:12:40.399 --> 00:12:45.080
<v Speaker 1>makes me feel like I can explore and experiment without

248
00:12:45.120 --> 00:12:45.879
<v Speaker 1>breaking anything.

249
00:12:46.080 --> 00:12:46.440
<v Speaker 2>Yeah.

250
00:12:46.480 --> 00:12:48.600
<v Speaker 1>So why would we want to use these separate worlds?

251
00:12:48.840 --> 00:12:53.279
<v Speaker 2>There are several reasons why worlds can be incredibly useful experimentation.

252
00:12:53.879 --> 00:12:56.799
<v Speaker 2>You can test out different changes or additions to your

253
00:12:56.799 --> 00:13:00.159
<v Speaker 2>ontology without worrying about messing up the main version and

254
00:13:00.519 --> 00:13:04.840
<v Speaker 2>scenario analysis. You can create different worlds to model various

255
00:13:04.919 --> 00:13:09.919
<v Speaker 2>hypothetical situations or explore different perspectives on the same data. Gotcha,

256
00:13:10.120 --> 00:13:14.039
<v Speaker 2>data separation. If you're working with data from multiple sources

257
00:13:14.559 --> 00:13:17.720
<v Speaker 2>or domains, you can keep them separate in different worlds

258
00:13:18.080 --> 00:13:19.279
<v Speaker 2>for easier management.

259
00:13:19.399 --> 00:13:22.200
<v Speaker 1>So it's like having different versions of our ontology exactly,

260
00:13:22.279 --> 00:13:25.480
<v Speaker 1>and we can switch between or even merge together if needed.

261
00:13:25.600 --> 00:13:30.879
<v Speaker 2>Precisely. It's a powerful way to manage complexity and explore

262
00:13:30.960 --> 00:13:32.720
<v Speaker 2>different possibilities without risk.

263
00:13:33.200 --> 00:13:37.080
<v Speaker 1>So how do we actually create and work with these worlds?

264
00:13:37.360 --> 00:13:39.360
<v Speaker 2>Already too makes it quite intuitive.

265
00:13:39.559 --> 00:13:39.919
<v Speaker 1>Okay.

266
00:13:40.279 --> 00:13:43.720
<v Speaker 2>You use the world class to create and manage your world.

267
00:13:44.240 --> 00:13:46.320
<v Speaker 2>It's like having a world building toolkit.

268
00:13:46.480 --> 00:13:49.159
<v Speaker 1>Okay, So we can create these separate worlds. Yeah, but

269
00:13:49.240 --> 00:13:51.799
<v Speaker 1>can we also interact with them? That's a great question,

270
00:13:52.320 --> 00:13:54.000
<v Speaker 1>or even share information between them?

271
00:13:54.320 --> 00:13:58.120
<v Speaker 2>So, while worlds are designed to be isolated environments okay,

272
00:13:58.279 --> 00:14:02.919
<v Speaker 2>already too, does provide some mechanisms for communication between them.

273
00:14:03.279 --> 00:14:06.240
<v Speaker 1>How would we go about transferring information between them?

274
00:14:07.120 --> 00:14:11.000
<v Speaker 2>One way is to use the import from function. You

275
00:14:11.000 --> 00:14:13.600
<v Speaker 2>can think of it like copying and pasting elements okay,

276
00:14:13.639 --> 00:14:15.399
<v Speaker 2>from one world into another.

277
00:14:15.279 --> 00:14:19.159
<v Speaker 1>So we can reuse parts of our ontology in different

278
00:14:19.200 --> 00:14:22.440
<v Speaker 1>contexts or scenarios. Yeah, that's pretty handy. Are there any

279
00:14:22.440 --> 00:14:23.039
<v Speaker 1>other ways?

280
00:14:23.279 --> 00:14:27.720
<v Speaker 2>You can also work with RDF triples directly? Remember those

281
00:14:27.720 --> 00:14:30.679
<v Speaker 2>are like the building blocks of ontologies already too. Lets

282
00:14:30.720 --> 00:14:33.960
<v Speaker 2>you manipulate these triples which can be used to transfer

283
00:14:34.039 --> 00:14:38.120
<v Speaker 2>information between worlds if you need more fine grain control.

284
00:14:38.360 --> 00:14:42.120
<v Speaker 1>Okay, so we have a few different options for interworld communication.

285
00:14:43.360 --> 00:14:46.279
<v Speaker 1>But is there anything we need to be cautious about.

286
00:14:46.519 --> 00:14:49.559
<v Speaker 2>Yeah, it's important to remember that changes made in one

287
00:14:49.600 --> 00:14:52.159
<v Speaker 2>world do not automatically affect the others.

288
00:14:52.440 --> 00:14:55.759
<v Speaker 1>Ah, so if I make a change in my experimental world,

289
00:14:56.480 --> 00:14:59.799
<v Speaker 1>it won't mess up the main world unless I explicitly

290
00:15:00.279 --> 00:15:01.159
<v Speaker 1>those changes.

291
00:15:00.879 --> 00:15:04.639
<v Speaker 2>Over precisely, each world maintains its own independent state.

292
00:15:05.080 --> 00:15:05.399
<v Speaker 1>Okay.

293
00:15:05.840 --> 00:15:09.639
<v Speaker 2>It's a powerful feature, but it's important to be aware

294
00:15:09.759 --> 00:15:13.080
<v Speaker 2>of the separation when working with multiple worlds.

295
00:15:13.320 --> 00:15:17.240
<v Speaker 1>This has been an incredible exploration of the capabilities of OWL.

296
00:15:17.279 --> 00:15:19.440
<v Speaker 1>Ready to uh huh, I'm starting to feel like I

297
00:15:19.440 --> 00:15:23.279
<v Speaker 1>can build and manipulate. That's great these ontologies with a

298
00:15:23.320 --> 00:15:24.279
<v Speaker 1>lot more confidence.

299
00:15:24.399 --> 00:15:25.960
<v Speaker 2>That's great to hear. And now that we have a

300
00:15:26.000 --> 00:15:29.399
<v Speaker 2>good grasp of the fundamentals, I think it's time to

301
00:15:29.559 --> 00:15:32.840
<v Speaker 2>move on to some real world applications.

302
00:15:32.960 --> 00:15:34.919
<v Speaker 1>Okay, I'm ready to see how these concepts are being

303
00:15:35.039 --> 00:15:35.799
<v Speaker 1>used in practice.

304
00:15:35.919 --> 00:15:38.679
<v Speaker 2>Great, let's start with a field that's near and dear

305
00:15:38.720 --> 00:15:40.840
<v Speaker 2>to my heart, bioinformatics.

306
00:15:40.840 --> 00:15:44.639
<v Speaker 1>Bioinformatics, Now that's a field where I can imagine ontologies.

307
00:15:44.720 --> 00:15:48.759
<v Speaker 1>You're absolutely right being incredibly useful. There's so much complex

308
00:15:48.879 --> 00:15:51.639
<v Speaker 1>data to organize and make sense of.

309
00:15:51.799 --> 00:15:55.240
<v Speaker 2>And one of the most widely used ontologies in bioinformatics

310
00:15:55.919 --> 00:15:57.080
<v Speaker 2>is the geneontology.

311
00:15:57.240 --> 00:16:01.159
<v Speaker 1>Geneontology or go for sure go. I've definitely heard that

312
00:16:01.320 --> 00:16:03.559
<v Speaker 1>term before, and I'm not entirely sure what it is

313
00:16:03.840 --> 00:16:04.240
<v Speaker 1>of course.

314
00:16:04.279 --> 00:16:09.799
<v Speaker 2>So the gene ontology is a hierarchical classification system that

315
00:16:09.840 --> 00:16:12.480
<v Speaker 2>describes the functions of genes and proteins.

316
00:16:12.799 --> 00:16:15.840
<v Speaker 1>So it's like a giant dictionary or encyclopedia. That's a

317
00:16:15.919 --> 00:16:18.000
<v Speaker 1>great analogy for genes and proteins.

318
00:16:18.120 --> 00:16:22.440
<v Speaker 2>GGO provides a standardized vocabulary and structure that helps us

319
00:16:22.440 --> 00:16:27.240
<v Speaker 2>describe what genes and proteins do in a consistent way.

320
00:16:27.360 --> 00:16:30.919
<v Speaker 1>But WEO is more than just a list of definitions.

321
00:16:30.320 --> 00:16:34.240
<v Speaker 2>Right, How is it actually structured? GGO is organized into

322
00:16:34.279 --> 00:16:40.159
<v Speaker 2>three main branches or some ontologies. Molecular function This branch

323
00:16:40.279 --> 00:16:44.559
<v Speaker 2>focuses on what a gene product actually does at the

324
00:16:44.600 --> 00:16:49.000
<v Speaker 2>molecular level, think of things like catalytic activity or binding.

325
00:16:49.080 --> 00:16:54.159
<v Speaker 2>Biological process. This branch describes the larger processes or pathways

326
00:16:54.600 --> 00:16:57.480
<v Speaker 2>that a gene product is involved in. Examples would be

327
00:16:57.519 --> 00:17:02.679
<v Speaker 2>things like cell growth or signal transduction. And then cellular component. Okay,

328
00:17:02.919 --> 00:17:06.880
<v Speaker 2>this branch describes where a gene product is located within

329
00:17:06.960 --> 00:17:10.119
<v Speaker 2>a cell, like the nucleus or membrane.

330
00:17:10.160 --> 00:17:11.599
<v Speaker 1>So it's like a three dimensional map.

331
00:17:11.720 --> 00:17:12.200
<v Speaker 2>You got it.

332
00:17:12.240 --> 00:17:13.480
<v Speaker 1>A gene and protein function.

333
00:17:13.759 --> 00:17:18.400
<v Speaker 2>Joe provides a comprehensive framework for understanding how genes and

334
00:17:18.480 --> 00:17:21.880
<v Speaker 2>proteins contribute to the complexity of life.

335
00:17:22.039 --> 00:17:24.960
<v Speaker 1>This is fascinating, but how is this actually used in research?

336
00:17:25.799 --> 00:17:28.119
<v Speaker 2>So, show go is used in a wide range of

337
00:17:28.160 --> 00:17:32.680
<v Speaker 2>bioinformatics applications. For example, it can help us predict the

338
00:17:32.759 --> 00:17:37.559
<v Speaker 2>function of genes that haven't been fully studied, analyze groups

339
00:17:37.559 --> 00:17:41.240
<v Speaker 2>of genes that are activated in certain diseases or conditions,

340
00:17:41.960 --> 00:17:47.240
<v Speaker 2>map genes and proteins to known biological pathways, or integrate

341
00:17:47.359 --> 00:17:48.799
<v Speaker 2>data from various sources.

342
00:17:48.920 --> 00:17:51.119
<v Speaker 1>So it sounds like JOE is a fundamental tool. It

343
00:17:51.160 --> 00:17:53.799
<v Speaker 1>is for anyone working with gene and protein data. Yes,

344
00:17:53.960 --> 00:17:56.160
<v Speaker 1>but where does already two fit into all of this?

345
00:17:56.400 --> 00:17:59.880
<v Speaker 2>Already two has excellent support for JOE. You can easily

346
00:18:00.119 --> 00:18:03.960
<v Speaker 2>load the entire gene ontology into already two as an

347
00:18:04.000 --> 00:18:07.400
<v Speaker 2>ontology and then use all of the features we've discussed

348
00:18:07.839 --> 00:18:09.279
<v Speaker 2>to analyze and explore the data.

349
00:18:09.359 --> 00:18:12.400
<v Speaker 1>So we could use owl ready too to navigate the

350
00:18:12.519 --> 00:18:17.799
<v Speaker 1>GROW hierarchy, yes, search for specific terms and even infer

351
00:18:17.920 --> 00:18:20.920
<v Speaker 1>relationships between different GROW concepts.

352
00:18:20.960 --> 00:18:24.359
<v Speaker 2>Exactly, it's like having a powerful search engine and analysis

353
00:18:24.359 --> 00:18:28.599
<v Speaker 2>toolk specifically designed for working with GROW data.

354
00:18:29.119 --> 00:18:31.559
<v Speaker 1>Can you give me a specific example of how they

355
00:18:31.640 --> 00:18:33.119
<v Speaker 1>might be used together in research?

356
00:18:33.599 --> 00:18:36.440
<v Speaker 2>Sure, let's talk about protein interactions.

357
00:18:37.200 --> 00:18:40.400
<v Speaker 1>Proteins. They're like the tiny machinescisely that make.

358
00:18:40.319 --> 00:18:44.279
<v Speaker 2>Life happen, and proteins rarely work in isolation. They interact

359
00:18:44.319 --> 00:18:47.920
<v Speaker 2>with each other in complex ways to carry out their functions.

360
00:18:48.160 --> 00:18:52.319
<v Speaker 1>So understanding these interactions is key to understanding how life.

361
00:18:52.039 --> 00:18:54.480
<v Speaker 2>Works exactly, and GROW can help us with that.

362
00:18:54.640 --> 00:18:54.920
<v Speaker 1>Okay.

363
00:18:55.079 --> 00:18:56.920
<v Speaker 2>Let's say we have a list of proteins that are

364
00:18:56.960 --> 00:18:59.640
<v Speaker 2>known to interact. We can use ow already two to

365
00:18:59.720 --> 00:19:04.240
<v Speaker 2>reach retrieve the grog terms associated with each protein, essentially

366
00:19:04.279 --> 00:19:06.759
<v Speaker 2>annotating them with their functional roles.

367
00:19:06.839 --> 00:19:09.400
<v Speaker 1>So we're giving each protein a label that tells us

368
00:19:09.440 --> 00:19:09.960
<v Speaker 1>what it does.

369
00:19:10.400 --> 00:19:14.240
<v Speaker 2>Right. Then we can use already two's reasoning capabilities okay

370
00:19:14.240 --> 00:19:18.640
<v Speaker 2>to analyze those labels, looking for patterns or connections between

371
00:19:18.880 --> 00:19:21.240
<v Speaker 2>the functions of the interacting proteins.

372
00:19:21.400 --> 00:19:23.799
<v Speaker 1>Oh, so we might find that a group of interacting

373
00:19:23.839 --> 00:19:27.839
<v Speaker 1>proteins is involved in the same biological process exactly, even

374
00:19:27.880 --> 00:19:28.960
<v Speaker 1>if we didn't know that before.

375
00:19:29.039 --> 00:19:35.240
<v Speaker 2>This kind of analysis can be incredibly valuable for generating hypotheses,

376
00:19:36.079 --> 00:19:41.000
<v Speaker 2>understanding disease mechanisms, and even identifying potential drug targets.

377
00:19:41.240 --> 00:19:44.079
<v Speaker 1>It shows how already two and Joe can be used

378
00:19:44.119 --> 00:19:46.400
<v Speaker 1>to make sense of complex biological data.

379
00:19:47.000 --> 00:19:50.279
<v Speaker 2>It's just one example of how ontologies are being used

380
00:19:50.319 --> 00:19:53.480
<v Speaker 2>to advance our understanding of the complexities of life.

381
00:19:53.519 --> 00:19:56.440
<v Speaker 1>That's right, we've only just scratched the surface of real

382
00:19:56.480 --> 00:19:57.480
<v Speaker 1>world applications.

383
00:19:57.480 --> 00:20:02.000
<v Speaker 2>Where should we explore next gears and talk about healthcare? Healthcare?

384
00:20:02.119 --> 00:20:05.480
<v Speaker 1>I imagine ontologies could be really valuable for organizing and

385
00:20:05.519 --> 00:20:07.240
<v Speaker 1>making sense of all the medical knowledge.

386
00:20:07.359 --> 00:20:11.680
<v Speaker 2>Absolutely, healthcare is a field that's ripe for ontology driven solutions.

387
00:20:12.079 --> 00:20:15.039
<v Speaker 2>Think about all the different terms and concepts and procedures

388
00:20:15.079 --> 00:20:19.440
<v Speaker 2>in medicine. It's a complex web of information, and.

389
00:20:19.319 --> 00:20:22.440
<v Speaker 1>With electronic health records becoming more common. Uh huh, there's

390
00:20:22.480 --> 00:20:25.680
<v Speaker 1>an even greater need to standardize and make sense of

391
00:20:25.720 --> 00:20:27.440
<v Speaker 1>all that data exactly.

392
00:20:27.480 --> 00:20:31.039
<v Speaker 2>Ontologies can help us do just that. They can be

393
00:20:31.200 --> 00:20:35.160
<v Speaker 2>used to create a structured representation of medical knowledge that

394
00:20:35.200 --> 00:20:37.960
<v Speaker 2>can be understood by both humans and computers.

395
00:20:38.200 --> 00:20:41.240
<v Speaker 1>So how are ontologies actually being used in healthcare today?

396
00:20:41.920 --> 00:20:45.279
<v Speaker 2>So one area where ontologies are making a big impact

397
00:20:45.599 --> 00:20:48.599
<v Speaker 2>is in electronic health records or EHRs.

398
00:20:48.759 --> 00:20:49.240
<v Speaker 1>EHRs.

399
00:20:49.279 --> 00:20:52.960
<v Speaker 2>Okay, they can help standardize the way patient information is recorded,

400
00:20:53.240 --> 00:20:57.359
<v Speaker 2>making it easier to share data between different healthcare providers.

401
00:20:57.640 --> 00:20:59.960
<v Speaker 1>That makes sense. It's like having a common language precise

402
00:21:00.160 --> 00:21:02.920
<v Speaker 1>for health care data, so everyone's on the same page.

403
00:21:03.000 --> 00:21:05.559
<v Speaker 2>And this can have a huge impact on patient care.

404
00:21:06.119 --> 00:21:08.799
<v Speaker 2>For example, if a patient goes to a new doctor, Yeah,

405
00:21:08.960 --> 00:21:12.519
<v Speaker 2>the doctor can quickly access their complete medical history from

406
00:21:12.559 --> 00:21:15.240
<v Speaker 2>their EHR, even if it was created by a different

407
00:21:15.279 --> 00:21:16.200
<v Speaker 2>healthcare provider.

408
00:21:16.319 --> 00:21:20.200
<v Speaker 1>That could definitely help prevent misdiagnoses and ensure that patients

409
00:21:20.200 --> 00:21:21.960
<v Speaker 1>get the right treatment exactly.

410
00:21:22.160 --> 00:21:25.440
<v Speaker 2>Another area where ontologies are being used is in clinical

411
00:21:25.440 --> 00:21:31.319
<v Speaker 2>Decisions support systems or cdss's. These systems help clinicians make

412
00:21:31.359 --> 00:21:33.839
<v Speaker 2>more informed decisions about patient care.

413
00:21:34.400 --> 00:21:36.799
<v Speaker 1>So how do ontologies fit into these systems.

414
00:21:36.920 --> 00:21:41.119
<v Speaker 2>Ontologies can be used to encode medical guidelines and best practices.

415
00:21:41.720 --> 00:21:46.160
<v Speaker 2>This allows cdssses to analyze patient data and provide clinicians

416
00:21:46.200 --> 00:21:48.079
<v Speaker 2>with viance based recommendations.

417
00:21:48.119 --> 00:21:50.400
<v Speaker 1>So it's like having a team of expert advisors.

418
00:21:50.759 --> 00:21:52.160
<v Speaker 2>That's a great way to think about it.

419
00:21:52.039 --> 00:21:54.519
<v Speaker 1>Working behind the scenes, Yeah, to help doctors make the

420
00:21:54.519 --> 00:21:55.880
<v Speaker 1>best decisions.

421
00:21:55.359 --> 00:21:58.599
<v Speaker 2>And this can be incredibly valuable for improving patient safety

422
00:21:58.640 --> 00:22:00.000
<v Speaker 2>and outcomes.

423
00:21:59.799 --> 00:22:02.799
<v Speaker 1>It seems like ontologies are already making a tangible difference

424
00:22:02.839 --> 00:22:03.400
<v Speaker 1>in healthcare.

425
00:22:03.759 --> 00:22:07.400
<v Speaker 2>They are, but healthcare isn't the only field where ontologies

426
00:22:07.400 --> 00:22:08.279
<v Speaker 2>are making a difference.

427
00:22:08.319 --> 00:22:10.799
<v Speaker 1>Are there any other fields where they're being used effectively?

428
00:22:11.160 --> 00:22:15.039
<v Speaker 2>Absolutely, ontologies are finding applications in a wide range of domains,

429
00:22:15.759 --> 00:22:19.920
<v Speaker 2>engineering and manufacturing to business and finance.

430
00:22:20.640 --> 00:22:23.319
<v Speaker 1>Wow, that's a pretty diverse range. Could you give me

431
00:22:23.359 --> 00:22:27.160
<v Speaker 1>an example of how ontologies might be used in engineering? Sure?

432
00:22:27.359 --> 00:22:30.039
<v Speaker 2>Think about the design and development of a complex product

433
00:22:30.160 --> 00:22:34.960
<v Speaker 2>like an airplane. Okay, there are so many different components,

434
00:22:35.079 --> 00:22:37.759
<v Speaker 2>systems and subsystems involved.

435
00:22:37.519 --> 00:22:41.400
<v Speaker 1>Right, and it's critical that all those pieces work together flawlessly.

436
00:22:41.000 --> 00:22:45.559
<v Speaker 2>Exactly, and ontologies can help engineers manage this complexity. They

437
00:22:45.599 --> 00:22:48.799
<v Speaker 2>can be used to create a detailed model of the airplane,

438
00:22:49.160 --> 00:22:51.359
<v Speaker 2>representing all its parts and how they interact.

439
00:22:51.559 --> 00:22:53.920
<v Speaker 1>So it's like having a digital blueprint of the airplane

440
00:22:54.000 --> 00:22:57.480
<v Speaker 1>decisely that captures all the engineering knowledge and relationships.

441
00:22:57.680 --> 00:23:00.279
<v Speaker 2>And this model can be used for a variety of persons,

442
00:23:00.960 --> 00:23:04.960
<v Speaker 2>from simulation and testing to maintenance and troubleshooting.

443
00:23:05.519 --> 00:23:07.960
<v Speaker 1>I can see how that would be incredibly valuable for

444
00:23:08.119 --> 00:23:10.640
<v Speaker 1>ensuring the safety and reliability of such a complex system.

445
00:23:11.000 --> 00:23:16.359
<v Speaker 2>Absolutely, ontologies are also being used in manufacturing to optimize

446
00:23:16.400 --> 00:23:20.720
<v Speaker 2>production processes, manage supply chains, and improve product quality.

447
00:23:21.160 --> 00:23:26.160
<v Speaker 1>So they're helping to make engineering and manufacturing more efficient.

448
00:23:25.720 --> 00:23:28.319
<v Speaker 2>And effective exactly. And these are just a few examples

449
00:23:28.359 --> 00:23:30.920
<v Speaker 2>of how ontologies are being used in the real world.

450
00:23:31.440 --> 00:23:36.960
<v Speaker 2>The possibilities are truly vast, and as the field of

451
00:23:37.000 --> 00:23:40.720
<v Speaker 2>ontology engineering continues to evolve, we can expect to see

452
00:23:40.799 --> 00:23:45.119
<v Speaker 2>even more innovative applications emerge in the years to come.

453
00:23:45.359 --> 00:23:47.279
<v Speaker 1>This has been an incredible journey through the world of

454
00:23:47.279 --> 00:23:49.799
<v Speaker 1>ontology has I feel like I've gained a whole new

455
00:23:49.880 --> 00:23:52.000
<v Speaker 1>understanding of their power and potential.

456
00:23:52.160 --> 00:23:54.200
<v Speaker 2>It's been a pleasure exploring these concepts with you.

457
00:23:54.319 --> 00:23:55.880
<v Speaker 1>The pleasure's all might and I hope.

458
00:23:55.680 --> 00:23:59.039
<v Speaker 2>You've found this dive into ontologies as fascinating as I have.

459
00:23:59.319 --> 00:24:02.680
<v Speaker 1>Absolutely, so as we wrap up, what are some key

460
00:24:02.720 --> 00:24:04.720
<v Speaker 1>takeaways that our listeners should keep in mind.

461
00:24:04.759 --> 00:24:07.039
<v Speaker 2>Well, I think the most important takeaway is that ontologies

462
00:24:07.079 --> 00:24:10.759
<v Speaker 2>are not just abstract concepts. They're practical tools that are

463
00:24:10.799 --> 00:24:14.039
<v Speaker 2>already being used to solve real world problems and create

464
00:24:14.160 --> 00:24:17.119
<v Speaker 2>innovative solutions in a wide range of domains.

465
00:24:17.440 --> 00:24:21.440
<v Speaker 1>And we've seen examples of that in healthcare, bioinformatics, engineering,

466
00:24:21.720 --> 00:24:22.000
<v Speaker 1>just to.

467
00:24:22.000 --> 00:24:25.279
<v Speaker 2>Name a few exactly. Another key takeaway is that Already

468
00:24:25.319 --> 00:24:28.960
<v Speaker 2>two is a powerful and accessible library. Yeah, that makes

469
00:24:28.960 --> 00:24:31.720
<v Speaker 2>it easy to work with ontologies and Python.

470
00:24:32.039 --> 00:24:32.400
<v Speaker 1>Okay.

471
00:24:32.640 --> 00:24:36.119
<v Speaker 2>Whether you're a researcher, developer, or just someone who's curious

472
00:24:36.119 --> 00:24:39.920
<v Speaker 2>about ontologies, Already too provides a great starting point for

473
00:24:40.119 --> 00:24:42.200
<v Speaker 2>exploring this fascinating field.

474
00:24:42.519 --> 00:24:46.079
<v Speaker 1>So if you're intrigued by the possibilities of ontologies, don't

475
00:24:46.079 --> 00:24:49.000
<v Speaker 1>be intimidated. Yeah, there are plenty of resources available to

476
00:24:49.000 --> 00:24:49.799
<v Speaker 1>help you get started.

477
00:24:49.960 --> 00:24:53.920
<v Speaker 2>I remember, ontologies are all about capturing and organizing knowledge, right,

478
00:24:54.000 --> 00:24:57.000
<v Speaker 2>So whether you're working with medical data, engineering, designs, or

479
00:24:57.039 --> 00:24:59.200
<v Speaker 2>even just trying to make sense of your own personal

480
00:24:59.319 --> 00:25:03.839
<v Speaker 2>knowledge base, ontologies can provide a valuable framework for structuring

481
00:25:03.880 --> 00:25:05.200
<v Speaker 2>and understanding information.

482
00:25:05.480 --> 00:25:07.559
<v Speaker 1>This has been an incredible deep dive, and I want

483
00:25:07.559 --> 00:25:09.480
<v Speaker 1>to thank you for sharing your expertise with us.

484
00:25:09.559 --> 00:25:12.119
<v Speaker 2>It's been my pleasure, and to our listeners.

485
00:25:11.759 --> 00:25:15.119
<v Speaker 1>I encourage you to keep exploring the world of ontologies. Yes,

486
00:25:15.799 --> 00:25:18.640
<v Speaker 1>who knows, you might just discover the next groundbreaking application.

487
00:25:18.839 --> 00:25:19.319
<v Speaker 2>You never know