-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
317 lines (270 loc) · 13.6 KB
/
index.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
<html lang="en">
<head>
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-88966686-2"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag() { dataLayer.push(arguments); }
gtag('js', new Date());
gtag('config', 'UA-88966686-2');
</script>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>RecipeScape: An Interactive Tool for Analyzing Cooking Instructions at Scale</title>
<link rel="stylesheet" href="style.css">
<!-- bootstrap -->
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.2.0/css/bootstrap.min.css">
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.2.0/css/bootstrap-theme.min.css">
<!-- Google fonts -->
<link href="https://fonts.googleapis.com/css?family=Roboto:400,300" rel="stylesheet" type="text/css">
<link href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/css/bootstrap.min.css" rel="stylesheet">
<style type="text/css">
.normal {
font-size: 20px;
}
.center {
margin: auto;
text-align: left;
position: relative;
width: 800px;
}
.center_vid {
margin: 0 auto;
display: block;
}
button {
margin: auto;
margin-top: 10px;
margin-bottom: 10px;
display: block;
}
canvas {
margin: auto;
position: relative;
display: block;
}
</style>
</head>
<div id="header">
<h1>RecipeScape: An Interactive Tool for Analyzing Cooking Instructions at Scale</h1>
<div style="clear:both;"></div>
</div>
<div class="container sec" align="center">
<a href="https://www.kixlab.org">
<img src=logo.png style="height:30px; margin-top: 10px;margin-right: 20px;">
</a>
<a href="http://cs.kaist.ac.kr/">
<img src="img/kaist_logo.jpg" style="height:50px; margin-right: 20px;">
</a>
<a href="https://www.epfl.ch">
<img src="https://bernoulli.epfl.ch/images/website/logo-epfl-black.png" style="height:50px; margin-right: 20px;">
</a>
<a href="http://stanford.edu/">
<img src="https://cs.stanford.edu/people/karpathy/densecap/stanfordlogo.jpg" style="height:50px; margin-right: 20px;">
</a>
</div>
<div class="sechighlight">
<div class="container sec">
<h2>Abstract</h2>
<div id="coursedesc" style="font-size:18px">
For cooking professionals and culinary students, understanding cooking instructions is an essential yet
demanding task. Common tasks include categorizing different approaches to cooking a dish and identifying
usage patterns of particular ingredients or cooking methods, all of which require extensive browsing and
comparison. However, no existing system provides support for such in-depth and at-scale analysis. We
present RecipeScape, an interactive system for browsing and analyzing the hundreds of recipes of a single
dish available online. We also introduce a computational pipeline that extracts cooking processes from
recipe text and calculates a procedural similarity between them. To evaluate how RecipeScape supports
culinary analysis at scale, we conducted a user study with cooking professionals and culinary students with
500 recipes for two different dishes. Results show that RecipeScape clusters recipes into distinct
approaches, and captures notable usage patterns of ingredients and cooking actions.
</div>
</div>
</div>
<div class="container sec">
<div align="center">
<iframe width="640" height="360" align="middle" src="https://www.youtube.com/embed/b1TccM1kN4U?rel=0"
frameborder="0" gesture="media" allow="encrypted-media" allowfullscreen></iframe>
</div>
</div>
<div class="sechighlight" align="center">
<div>
<div class="instructor" style="width:150px">
<a href="https://minsukchang.com">
<div class="instructorphoto"><img src=unnamed.jpg></div>
<div>Minsuk Chang</div>
</a>
<div>KAIST</div>
</div>
<div class="instructor" style="width:150px">
<div class="instructorphoto"><img src="https://avatars3.githubusercontent.com/u/17401551?s=460&v=4"></div>
<div>Leonore Guillain</div>
<div>EPFL</div>
</div>
<div class="instructor" style="width:150px">
<a href="http://www.hyeungshikjung.com/">
<div class="instructorphoto"><img src="https://avatars1.githubusercontent.com/u/5417330?s=460&v=4"></div>
<div>Hyeungshik Jung</div>
</a>
<div>KAIST</div>
</div>
<div class="instructor" style="width:150px">
<div class="instructorphoto"><img src="img/vivian_hare.jpg"></div>
<div>Vivian Hare</div>
<div>Stanford University</div>
</div>
<div class="instructor" style="width:150px">
<a href="http://juhokim.com">
<div class="instructorphoto"><img src="img/juho.jpg"></div>
<div style="width:150px">Juho Kim</div>
</a>
<div>KAIST</div>
</div>
<div class="instructor" style="width:150px">
<a href="https://graphics.stanford.edu/~maneesh/">
<div class="instructorphoto"><img src="img/maneesh.jpg"></div>
<div style="width:150px">Maneesh Agrawala</div>
</a>
<div>Stanford University</div>
</div>
</div>
</div>
<div class="container sec" style="font-size:22px">
<h2>Live Interface Dashboard</h2>
<a href="https://kixlab.github.io/recipescape-dashboard/">Link to our analytics dashboard interface
</a><br>
<div align="center">
<img src="ss_dashboard.png" width="80%" />
</div>
<br>
<p>RecipeScape is an interface for analyzing cooking processes at scale with three main visualization components:
(a) RecipeMap provides clusters of recipes with respect to their structural similarities
(b) RecipeDeck provides in-depth view and pairwise comparisons of recipes
(c) RecipeStat provides usage patterns of cooking actions and ingredients. <p>
</div>
<div class="sechighlight">
<div class="container sec" style="font-size:18px">
<h2>Computational Pipeline</h2>
<div class="row">
<div class="col-md-5">
<img src="ss_pipeilne.png" width="90%" />
</div>
<div class="col-md-7">
<p>Computational Pipeline that uses Part-of-Speech tagger and human annotation to convert recipe text
into a tree representation,
and calculates pairwise distance to visualize the similarities. <p>
<p>
<b>Data Gathering:</b> In the data gathering step, we crawl all search results for a
queried dish, like chocolate chip cookie and tomato pasta, from recipe websites that use
the schema.org’s Recipe scheme. <br>
<b>Parsing:</b> We use off the shelf POS tagger and human annotation to parse tokens of the
crawled recipes. More detail is provided in the section below on annotation interface.<br>
<b>Similarity Comparison:</b> In order to obtain similarities between the recipes, we use a
tree edit distance, a commonly used technique for comparing tree structures. However,
to incorporate the semantic difference between individual cooking actions and ingredients
in capturing the structural difference, we dynamically adjust the weights associated with
the relabel operations.
These weights are calculated using the cosine similarities of words from a pre-trained word
embedding model.<br>
<b>Distance Matrix: </b> This similarity information is stored in a pairwise distance
matrix, where each element is the tree edit distance between the corresponding recipes. The
distance matrix is then converted into x,y coordinates using the Gram matrix. <br>
<b>Hierarchical Clustering:</b> We used hierarchical clustering to group recipes with
procedural similarities.
</p>
</div>
</div>
</div>
</div>
<div class="container sec" style="font-size:18px">
<h2>Recipe Annotation Interface</h2>
<div class="row">
<div class="col-md-5">
<img src="ss_annotation.png" width="90%" />
</div>
<div class="col-md-7">
We use <a href="https://stanfordnlp.github.io/CoreNLP/">Stanford CoreNLP</a>’s Part- of-Speech (POS) tagger
to label verbs and nouns in recipe instructions crawled in the data gathering step.
Most state-of-the-art POS taggers are statistically trained using mostly declarative sentences.
In result, the performance is rather limited with imperative sentences in recipe instructions like “Whisk
in chocolate hazelnut spread until combined and re- move from heat.”,
because parsers recognize “Whisk” as a noun, “spread” as a verb, but they are a verb and a noun in this
sentence respectively.
To overcome this drawback and to more accurately identify tokens for cooking actions and cooking
ingredients,
we recruited 12 annotators on campus to use a custom built <b>web-based interface (left) </b> for
annotating recipe instructions.
After an iteration with the POS tagger, the crowd annotator fixes the tags that are labelled incorrectly.
</div>
</div>
<br>
</div>
<div class="sechighlight">
<div class="container sec" style="font-size:18px">
<div class="row">
<div class="col-md-5">
<h2>Annotated Recipe Dataset</h2>
<p> Please cite our work if you use our annotated data! </p>
<p> <a href="ccc_trees.json"> recipes from "Chocolate Chip Cookie" query</a> </p>
<p> <a href="tomatopasta_trees.json"> recipes from "Tomato Pasta" query</a> </p>
<h2>Source Code</h2>
<p> <a href="https://github.com/kixlab/recipescape-dashboard"> Dashboard </a></p>
<p> <a href="https://github.com/kixlab/recipescape-pipeline"> Pipeline </a></p>
<h2>CHI 2018 Paper (Camera-Ready)</h2>
<p> <a href="https://dl.acm.org/citation.cfm?doid=3173574.3174025"> ACM DL Link</a> </p>
<p> <a href="RecipeScape_CHI_2018__Camera_Ready_.pdf"> Link to paper </a> </p>
<p> <a href="RecipeScape_CHI2018_HCIK_Poster_v2.pdf">Poster </a> for <a href="http://hci.kaist.ac.kr">HCI
@ KAIST 2018</a> Winter Workshop </p>
<h2>CHI 2018 Slides </h2>
<p> <a href="20180426_RecipeScape_CHI2018.pdf"> Slides </a></p>
</div>
<div class="col-md-7">
<h2>Bibtex</h2>
<pre style="font-size:12px;">
@inproceedings{chang2018recipescape,
title={RecipeScape: An Interactive Tool for Analyzing Cooking Instructions at Scale},
author={Chang, Minsuk and Guillain, L{\'e}onore V and Jung, Hyeungshik and Hare, Vivian M and Kim, Juho and Agrawala, Maneesh},
booktitle={Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems},
pages={451},
year={2018},
organization={ACM}
}
</pre>
</div>
</div>
<hr class="style1">
<div class="row">
<div class="col-md-5">
<h2>CHI 2017 LBW Paper</h2>
<p>While our most recent paper is getting ready for publication, you can access our previous
exploration <a href="ea1524-chang.pdf">here - Link to LBW Paper</a>. The interface and the
computational details have changed, but the overall approach is similar. </p>
<p><a href="https://dl.acm.org/citation.cfm?id=3053118"> ACM DL Link</a></p>
<p><a href="RecipeScape_LBW_poster_print.pdf"> Poster</a></p>
<p><a href="20170202_Minsuk_RecipeScape_LBW.pdf"> Slides </a> for <a href="http://hci.kaist.ac.kr">
HCI@KAIST Workshop </a> - 2017.02.02</p>
</div>
<div class="col-md-7">
<h2>Bibtex</h2>
<pre style="font-size:12px;">
@inproceedings{chang2017recipescape,
title={Recipescape: Mining and analyzing diverse processes in cooking recipes},
author={Chang, Minsuk and Hare, Vivian M and Kim, Juho and Agrawala, Maneesh},
booktitle={Proceedings of the 2017 CHI Conference Extended Abstracts on Human Factors in Computing Systems},
pages={1524--1531},
year={2017},
organization={ACM}
}
</pre>
</div>
</div>
</div>
</div>
<div class="container sec">
<div id="footer">
This research is supported by HRHRP at <a href="http://www.kaist.ac.kr"> KAIST</a>, and the <a href="https://brown.stanford.edu/">
Brown Institute </a> at Stanford University.
</div>
</div>
</body>
</html>