-
Notifications
You must be signed in to change notification settings - Fork 5
/
2013-09-oxford-real-data-scientist.html
403 lines (323 loc) · 14.1 KB
/
2013-09-oxford-real-data-scientist.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Will the real data scientist please stand up?</title>
<meta name="description" property="dc:description" content="An honest critique of data science and the data scientist"/>
<meta name="author" property="dc:creator" content="Tom Heath"/>
<meta name="apple-mobile-web-app-capable" content="yes"/>
<meta name="apple-mobile-web-app-status-bar-style" content="black-translucent"/>
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no">
<link rel="stylesheet" href="reveal.js/css/reveal.min.css">
<link rel="stylesheet" href="reveal.js/css/theme/sky.css" id="theme">
<!-- For syntax highlighting -->
<link rel="stylesheet" href="reveal.js/css/zenburn.css">
<!--[if lt IE 9]>
<script src="lib/js/html5shiv.js"></script>
<![endif]-->
<style type="text/css">
body #odi-presentation {
background-color: #FFFFFF;
}
#odi-presentation .logo {
border: none;
box-shadow: none;
}
#odi-presentation table {
width: 100%;
}
#odi-presentation td, #odi-presentation th {
text-align: center;
}
#odi-presentation th {
font-weight: bold;
}
#odi-presentation p {
margin: 0.8em 0 0.8em 0;
}
#odi-presentation li {
margin: 0.4em 0 0.4em 0;
}
#odi-presentation blockquote {
padding: 0.8em;
}
#odi-presentation .quote-source {
font-size: smaller;
}
#odi-presentation ul.centred-list {
list-style: none;
text-align: center;
}
#odi-presentation ul.centred-list li {
margin: 0.8em;
}
#odi-presentation .code {
font-family: courier;
border: 1px solid gray;
padding: 0.8em;
}
#odi-presentation .red-box {
border: 2px red dashed; padding: 4px;
}
</style>
</head>
<body id="odi-presentation">
<div class="reveal">
<!-- Any section element inside of this container is displayed as a slide -->
<div class="slides">
<section id="titlepage">
<h2>Will the Real<br/>Data Scientist<br/>Please Stand Up!</h2>
<p>Dr Tom Heath · Head of Research · Open Data Institute</p>
<p><a href="mailto:[email protected]">[email protected]</a> · <a href="https://twitter.com/tommyh">@tommyh</a></p>
<p>Digital Research 2013, Oxford, 10/09/2013</p>
<hr>
<span style="font-size: medium">
<p>You can access the slides at <a href="http://theodi.github.io/presentations/"> <code class="url">http://theodi.github.io/presentations/</code></a>.</p>
<p>Use arrows to navigate. Press 'f' for fullscreen. Press the <code>Escape</code> key to see all slides.</p>
</span>
</section>
<section>
<h2>Background</h2>
<ul class="centred-list">
<li class="fragment"><strong>PhD in Social Network-driven Recommender Systems</strong></li>
<li class="fragment"><strong>Heavily involved in Linked Data community</strong></li>
<li class="fragment"><strong>Senior Research/Data Scientist at Talis</strong></li>
<li class="fragment"><strong>Data Scientist at Open Data Institute</strong></li>
<li class="fragment"><strong>Head of Research at Open Data Institute</strong></li>
</ul>
</section>
<section>
<h2>The Open Data Institute</h2>
<ul class="centred-list">
<li class="fragment"><strong>non-profit, non-partisan</strong></li>
<li class="fragment"><strong><em>"helping others be successful<br/>with open data"</em></strong></li>
<li class="fragment"><strong>startups, training</strong></li>
<li class="fragment"><strong>tech services<br/></strong>(e.g. <a href="http://certificates.theodi.org/">http://certificates.theodi.org/</a>)</li>
<li class="fragment"><strong>research, policy</strong></li>
</ul>
</section>
<section>
<h2>Data Scien(ce|tist):<br/>An Honest Critique</h2>
</section>
<section>
<h2>Data Science:<br/>Why all the fuss?</h2>
</section>
<section>
<h2>Why all the Fuss?</h2>
<ul class="centred-list">
<li>
<strong>"the new oil"<br/>
"the new raw material"</strong>
</li>
<li class="fragment">
<strong>differences</strong><br/>
data is a non-rival good<br/>
marginal cost of distribution<br/>
falling cost of analysis
</li>
<li class="fragment">
<strong>a more level playing field</strong><br/>
more like gold than oil?
</li>
<li class="fragment"><strong>everyone wants a piece of the action</strong></li>
</ul>
</section>
<section>
<h2>The Rise of the<br/>Data Scientist</h2>
<ul class="centred-list">
<li class="fragment"><strong><em>"<a href="http://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century/">Sexiest job title of the 21st Century</a>"</em></strong></li>
<li class="fragment"><strong>recruitment bun-fight</strong></li>
<li class="fragment"><strong><em>"i gotta get me one of them there<br/>data scientists"</em></strong></li>
<li class="fragment"><strong>intentions are good</strong></li>
</ul>
</section>
<section>
<h2>Pros of the Concept <em>"Data Science"</em></h2>
<ul class="centred-list">
<li class="fragment"><strong>shines the spotlight on data</strong></li>
<li class="fragment"><strong>highlights the potential for value</strong></li>
<li class="fragment"><strong>a unifying label for what we already do</strong></li>
<li class="fragment"><strong>a focal point for our aspirations</strong></li>
</ul>
</section>
<section>
<h2>Cons of the Concept "Data Science"</h2>
<ul class="centred-list">
<li class="fragment"><strong>a catch-all label for what we already do</strong></li>
<li class="fragment"><strong>a focal point for our confusion!</strong></li>
</ul>
</section>
<section>
<h2>What is Data Science?</h2>
<ul class="centred-list">
<li class="fragment"><strong>often defined in terms of attributes of <br/>the data scientist</strong></li>
</ul>
<span class="fragment">
<blockquote>Data science isn't just about the existence of data, or making guesses about what that data might mean; it's about testing hypotheses and making sure that the conclusions you're drawing from the data are valid.</blockquote>
<p class="quote-source">— <a href="http://radar.oreilly.com/2010/06/what-is-data-science.html" title="">http://radar.oreilly.com/2010/06/what-is-data-science.html</a></p>
</span>
<ul class="centred-list">
<li class="fragment"><strong>just sounds like science, right?</strong></li>
</ul>
</section>
<section>
<h2>The Naming Problem</h2>
<ul class="centred-list">
<li class="fragment"><strong>biological science, computer science, <br/>political science, sports science...</strong></li>
<li class="fragment"><strong>the discipline is the focus, not the tool</strong></li>
<li class="fragment"><strong>where does this leave data science?</strong></li>
</ul>
</section>
<section>
<h2>Two Interpretations of<br/>Data Science</h2>
<ul class="centred-list">
<li class="fragment"><strong>the science of data?</strong></li>
<li class="fragment"><strong>science with data?</strong></li>
</ul>
</section>
<section>
<h2>Interpretation 1:<br/>The Science of Data</h2>
</section>
<section>
<h2>(Open) Data<br/>Research Topics</h2>
<ul class="centred-list">
<li><strong>Organising and Publishing</strong></li>
<li class="fragment">what is a meaningful definition of a collection/data set?</li>
<li class="fragment">how should data sets be described on the Web?</li>
<li class="fragment">how can we best extract (and aggregate)<br/>these descriptions?</li>
<li class="fragment">how do licensing choices affect<br/>the data ecosystem?</li>
</ul>
</section>
<section>
<h2>(Open) Data<br/>Research Topics</h2>
<ul class="centred-list">
<li><strong>Discovery, Comprehension and Use</strong></li>
<li class="fragment">how do we prioritise data discovery<br/>(i.e. crawler optimisation)?</li>
<li class="fragment">can we meaningfully summarise large data sets?</li>
<li class="fragment">what are the optimal indexing schemes for data?</li>
<li class="fragment">how linked can data be?</li>
<li class="fragment">are there novel cognitive architectures that can underpin our interactions with data?</li>
</ul>
</section>
<section>
<h2>Interpretation 2:<br/>Science with Data</h2>
<ul class="centred-list">
<li class="fragment"><strong>duh!</strong></li>
</ul>
</section>
<section>
<h2>Give the People what they Want!</h2>
<ul class="centred-list">
<li class="fragment">actionable, data-driven insights and answers...</li>
<li class="fragment">...to significant problems</li>
<li class="fragment">could be business/organisational/societal</li>
<li class="fragment">amounts to evidence-based everything</li>
<li class="fragment">c.f. the <em>"growth hacker"</em></li>
</ul>
</section>
<section>
<h2>How?</h2>
<ul class="centred-list">
<li class="fragment">identify the problem/question</li>
<li class="fragment">design research/analysis protocol</li>
<li class="fragment">identify the required data</li>
<li class="fragment">operational change to ensure the data is collected!</li>
<li class="fragment">perform the analysis</li>
<li class="fragment">report back to 'client'</li>
<li class="fragment">close the loop through organisational change</li>
<!-- <li class="fragment">this is about more than data, it's about organisational transformation!</li> -->
</ul>
</section>
<section>
<h2>An Example</h2>
<p><img src="2013-06-bangalore-yahoo-ss-data-and-the-web/prescribinganalytics.png" alt="Prescribing Analytics screenshot"/></p>
<ul class="centred-list">
<li>potential £200 million saving / year</li>
</ul>
<p class="quote-source">— <a href="http://prescribinganalytics.com/" title="Prescribing Analytics">http://prescribinganalytics.com/</a></p>
</section>
<section>
<h2>Caveats</h2>
<ul class="centred-list">
<li class="fragment"><strong>there has to be a research question</strong><br/>(otherwise it's 'just' engineering)</li>
<li class="fragment"><strong>careful of the obsession with e.g. machine learning</strong></li>
</ul>
</section>
<section>
<h2>Training Priorities</h2>
<ul class="centred-list">
<li><strong>statistics, experimental design, communication, business nous</strong></li>
</ul>
</section>
<section>
<h2>Lessons for the<br/>Research Community</h2>
<ul class="centred-list">
<li class="fragment"><strong>a different style of large scale data crunching</strong></li>
<li class="fragment"><strong>nurturing communities</strong></li>
</ul>
</section>
<section>
<h2>Key Challenges for<br/>Data Scientists</h2>
</section>
<section>
<h2>Access to Data</h2>
<ul class="centred-list">
<li class="fragment"><strong>this is not a lab environment</strong></li>
<li class="fragment">work with the data you can scrounge, and/or petition effectively within the business to start collecting what you need</li>
</ul>
</section>
<section>
<h2>Data Quality</h2>
<ul class="centred-list">
<li class="fragment"><strong>again, not a lab environment</strong></li>
<li class="fragment"><strong>e.g. how testable/tested is your cleansing code?</strong></li>
</ul>
</section>
<section>
<h2>Domain Specialism</h2>
<ul class="centred-list">
<li class="fragment"><strong>meaningful questions come from domain problems, not the existence of data</strong></li>
</ul>
</section>
<section>
<h2>Questions?</h2>
<p><a href="http://theodi.github.io/presentations/2013-09-oxford-real-data-scientist.html">http://theodi.github.io/presentations/2013-09-oxford-real-data-scientist.html</a></p>
</section>
<section>
<p>Dr Tom Heath · Head of Research · Open Data Institute</p>
<p><img src="brand/odi-logo-lg.png" alt="ODI" style="border:none;"></a></p>
<p><a href="mailto:[email protected]">[email protected]</a> · <a href="https://twitter.com/tommyh">@tommyh</a></p>
</section>
</div>
<div id="brand">
<a class="image" href="http://theodi.org"><img height="40" style="position:fixed; left:10px; bottom:50px; z-index:30;" src="brand/odi_logo.svg" alt="ODI"></a>
<br>
<a class="image" href="http://creativecommons.org/licenses/by-sa/3.0/"><img width="97" height="34" style="position:fixed; left:10px; bottom:10px; z-index:30;" src="brand/licence.png" alt="Creative Commons"></a>
</div>
</div>
<script src="reveal.js/js/head.min.js"></script>
<script src="reveal.js/js/reveal.min.js"></script>
<script>
Reveal.initialize({
controls: false,
progress: true,
history: true,
center: true,
theme: Reveal.getQueryHash().theme, // available themes are in /css/theme
transition: Reveal.getQueryHash().transition || 'default', // default/cube/page/concave/zoom/linear/fade/none
dependencies: [
]
});
</script>
<script type="text/javascript">
var _gaq = _gaq || [];
_gaq.push(['_setAccount', 'UA-34573394-10']);
_gaq.push(['_trackPageview']);
(function() {
var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
})();
</script>
</body>
</html>