forked from xinlongj/xinlongj.github.io
-
Notifications
You must be signed in to change notification settings - Fork 0
/
fintext.html
331 lines (245 loc) · 16.2 KB
/
fintext.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
<!DOCTYPE html>
<html >
<head>
<!-- Site made with Mobirise Website Builder v5.3.0, https://mobirise.com -->
<meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="generator" content="Mobirise v5.3.0, mobirise.com">
<meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
<link rel="shortcut icon" href="assets/images/site-1-128x128.png" type="image/x-icon">
<meta name="description" content="The details of the study on financial text analysis.">
<title>Financial Text Analysis</title>
<link rel="stylesheet" href="assets/bootstrap/css/bootstrap.min.css">
<link rel="stylesheet" href="assets/bootstrap/css/bootstrap-grid.min.css">
<link rel="stylesheet" href="assets/bootstrap/css/bootstrap-reboot.min.css">
<link rel="stylesheet" href="assets/tether/tether.min.css">
<link rel="stylesheet" href="assets/dropdown/css/style.css">
<link rel="stylesheet" href="assets/socicon/css/styles.css">
<link rel="stylesheet" href="assets/theme/css/style.css">
<link rel="preload" as="style" href="assets/mobirise/css/mbr-additional.css"><link rel="stylesheet" href="assets/mobirise/css/mbr-additional.css" type="text/css">
</head>
<body>
<!-- Analytics -->
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-10509316-3"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-10509316-3');
</script>
<!-- /Analytics -->
<section class="menu cid-rpQOWImaEh" once="menu" id="menu2-1c">
<nav class="navbar navbar-expand beta-menu navbar-dropdown align-items-center navbar-fixed-top navbar-toggleable-sm">
<button class="navbar-toggler navbar-toggler-right" type="button" data-toggle="collapse" data-target="#navbarSupportedContent" aria-controls="navbarSupportedContent" aria-expanded="false" aria-label="Toggle navigation">
<div class="hamburger">
<span></span>
<span></span>
<span></span>
<span></span>
</div>
</button>
<div class="menu-logo">
<div class="navbar-brand">
<span class="navbar-caption-wrap"><a class="navbar-caption text-black display-4" href="https://mobirise.co">@ICT, CAS</a></span>
</div>
</div>
<div class="collapse navbar-collapse" id="navbarSupportedContent">
<ul class="navbar-nav nav-dropdown" data-app-modern-menu="true"><li class="nav-item">
<a class="nav-link link text-black display-4" href="index.html">
Home
</a>
</li>
<li class="nav-item">
<a class="nav-link link text-black display-4" href="calendar.html">
Calendar
</a>
</li>
<li class="nav-item">
<a class="nav-link link text-black display-4" href="resch.html">Research</a>
</li><li class="nav-item"><a class="nav-link link text-black display-4" href="pubs.html">Publications</a></li><li class="nav-item"><a class="nav-link link text-black display-4" href="teach.html">Teaching</a></li><li class="nav-item"><a class="nav-link link text-black display-4" href="stds.html">Students</a></li></ul>
<div class="navbar-buttons mbr-section-btn"><a class="btn btn-sm btn-primary display-4" href="https://scholar.google.com/citations?user=W8wrWfMAAAAJ&hl=en" target="_blank"><span class="socicon socicon-googlescholar mbr-iconfont mbr-iconfont-btn" style="font-size: 16px;"></span></a> <a class="btn btn-sm btn-primary display-4" href="mailto:[email protected]"><span class="socicon socicon-mail mbr-iconfont mbr-iconfont-btn" style="font-size: 16px;"></span></a></div>
</div>
</nav>
</section>
<section once="footers" class="cid-rpRZOXFy6T" id="footer6-1d">
<div class="container">
<div class="media-container-row align-center mbr-white">
<div class="col-12">
<p class="mbr-text mb-0 mbr-fonts-style display-4">
© Last modified on Jan. 2021 by
</p>
</div>
</div>
</div>
</section>
<section class="mbr-section article content10 cid-rSkkW9U2IC" id="content10-1r">
<div class="container">
<div class="inner-container" style="width: 100%;">
<hr class="line" style="width: 32%;">
<div class="section-text align-center mbr-white mbr-fonts-style display-5">Enlightenments, like accidents, happen only to prepared minds.
<div>--Herbert Simon</div></div>
<hr class="line" style="width: 32%;">
</div>
</div>
</section>
<section class="mbr-section content4 cid-rSkePDOHOf" id="content4-1e">
<div class="container">
<div class="media-container-row">
<div class="title col-12 col-md-8">
<h2 class="align-center pb-3 mbr-fonts-style display-2">Financial Text Analysis</h2>
<h3 class="mbr-section-subtitle align-center mbr-light mbr-fonts-style display-5">DL/RL models for understanding financial related text,
<div>e.g. financial news, reports or social media.</div></h3>
</div>
</div>
</div>
</section>
<section class="mbr-section article content9 cid-rSkeRqGufp" id="content9-1f">
<div class="container">
<div class="inner-container" style="width: 100%;">
<hr class="line" style="width: 25%;">
<div class="section-text align-center mbr-fonts-style display-7">We mainly investigate DL/RL methods in understanding financial related text, such as document/sentence/aspect level sentiment analysis, extractive/abstractive summarization of financial documents. Due to the reality of overlong, redundant and obscure financial related text, there are specialized challenges when processing them. Our methodology is to devise knowledge-injected neural models to combine financial experts' experiences with AI solutions.</div>
<hr class="line" style="width: 25%;">
</div>
</div>
</section>
<section class="mbr-section content4 cid-scdXPNK4rc" id="content4-2t">
<div class="container">
<div class="media-container-row">
<div class="title col-12 col-md-8">
<h2 class="align-center pb-3 mbr-fonts-style display-5">Joint Chinese Word Segmentation and Part-of-speech Tagging
via Two-way Attentions of Auto-analyzed Knowledge (ACL 2020)</h2>
</div>
</div>
</div>
</section>
<section class="mbr-section content6 cid-scdXQrHcSm" id="content6-2u">
<div class="container">
<div class="media-container-row">
<div class="col-12 col-md-8">
<div class="media-container-row">
<div class="mbr-figure" style="width: 75%;">
<img src="assets/images/acl2020-596x296.png" alt="Mobirise" title="">
</div>
<div class="media-content">
<div class="mbr-section-text">
<p class="mbr-text mb-0 mbr-fonts-style display-7">In this paper, we propose a neural model named TWASP for joint CWS and POS tagging following the character-based sequence labeling paradigm, where a two-way attention mechanism is used to incorporate both context feature and their corresponding syntactic knowledge for each input character. Particularly, we use existing language processing toolkits to obtain the auto-analyzed syntactic knowledge for the context, and the proposed attention module can learn and benefit from them although their quality may not be perfect. Our experiments illustrate the effectiveness of the two-way attentions for joint CWS and POS tagging, where state-of-the-art performance is achieved on five benchmark dataset. <a href="https://www.aclweb.org/anthology/2020.acl-main.735.pdf" target="_blank">[paper]</a><br></p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="mbr-section content4 cid-scdZbvKzfM" id="content4-2v">
<div class="container">
<div class="media-container-row">
<div class="title col-12 col-md-8">
<h2 class="align-center pb-3 mbr-fonts-style display-5">A Challenge Dataset and Effective Models for Aspect-Based Sentiment Analysis (EMNLP 2019)</h2>
</div>
</div>
</div>
</section>
<section class="mbr-section content6 cid-scdZcdJmCy" id="content6-2w">
<div class="container">
<div class="media-container-row">
<div class="col-12 col-md-8">
<div class="media-container-row">
<div class="mbr-figure" style="width: 70%;">
<img src="assets/images/capnet-355x359.png" alt="Mobirise" title="">
</div>
<div class="media-content">
<div class="mbr-section-text">
<p class="mbr-text mb-0 mbr-fonts-style display-7">In
this paper, we present a new large-scale Multi Aspect Multi-Sentiment (MAMS) dataset, in which each sentence contains at least two different aspects with different sentiment polarities. The release of this dataset would push forward the research in this field. In addition, we propose simple yet effective CapsNet and CapsNet-BERT models which combine the strengths of recent NLP advances. Experiments on our new dataset show that the proposed model significantly outperforms the
state-of-the-art baseline methods. <a href="https://www.aclweb.org/anthology/D19-1654.pdf" target="_blank">[paper]</a><br></p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="mbr-section content4 cid-rSkjKSjd21" id="content4-1p">
<div class="container">
<div class="media-container-row">
<div class="title col-12 col-md-8">
<h2 class="align-center pb-3 mbr-fonts-style display-5">Reading Like HER: Human Reading Inspired Extractive Summarization (EMNLP2019)</h2>
</div>
</div>
</div>
</section>
<section class="mbr-section content6 cid-rSkjLm2E0b" id="content6-1q">
<div class="container">
<div class="media-container-row">
<div class="col-12 col-md-8">
<div class="media-container-row">
<div class="mbr-figure" style="width: 110%;">
<img src="assets/images/444-490x161.png" alt="Mobirise" title="">
</div>
<div class="media-content">
<div class="mbr-section-text">
<p class="mbr-text mb-0 mbr-fonts-style display-7">In this work, we re-examine extractive text summarization by simulating the process of extracting summarization of human. We adopt a convolutional neural network to encode gist of paragraphs for rough reading, and a decision making policy with an adapted termination mechanism for careful reading. <a href="https://www.aclweb.org/anthology/D19-1300.pdf" target="_blank">[paper]</a><a href="https://github.com/LLluoling/HER" target="_blank">[code]</a><br></p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="mbr-section content4 cid-rSkj9Nb3Nm" id="content4-1n">
<div class="container">
<div class="media-container-row">
<div class="title col-12 col-md-8">
<h2 class="align-center pb-3 mbr-fonts-style display-5">Unsupervised Neural Aspect Extraction with Sememes<br>(IJCAI 2019)</h2>
</div>
</div>
</div>
</section>
<section class="mbr-section content6 cid-rSkjarNdK9" id="content6-1o">
<div class="container">
<div class="media-container-row">
<div class="col-12 col-md-8">
<div class="media-container-row">
<div class="mbr-figure" style="width: 110%;">
<img src="assets/images/0z61i0wmuahie9n74j-490x189.png" alt="Mobirise" title="">
</div>
<div class="media-content">
<div class="mbr-section-text">
<p class="mbr-text mb-0 mbr-fonts-style display-7">In this work, we present an unsupervised neural framework that leverages sememes to enhance lexical semantics. We propose a sememe attention structure to represent word meanings and add an RNN sentence encoder for guiding the sememe exploration. The experimental results show that our model is superior to the existing models especially on identifying infrequent aspects. <a href="https://www.ijcai.org/Proceedings/2019/0712.pdf" target="_blank">[paper]</a><a href="https://github.com/LLluoling/Unsupervised-Neural-Aspect-Extraction-with-Sememes" target="_blank">[code]</a><br></p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="mbr-section content4 cid-rSkfMkSgHh" id="content4-1k">
<div class="container">
<div class="media-container-row">
<div class="title col-12 col-md-8">
<h2 class="align-center pb-3 mbr-fonts-style display-5">Beyond Polarity: Interpretable Financial Sentiment Analysis with Hierarchical Query-driven Attention(IJCAI 2018)</h2>
</div>
</div>
</div>
</section>
<section class="mbr-section content6 cid-rSkfL8zkRX" id="content6-1j">
<div class="container">
<div class="media-container-row">
<div class="col-12 col-md-8">
<div class="media-container-row">
<div class="mbr-figure" style="width: 70%;">
<img src="assets/images/222-436x412.png" alt="Mobirise" title="">
</div>
<div class="media-content">
<div class="mbr-section-text">
<p class="mbr-text mb-0 mbr-fonts-style display-7">We propose an interpretable framework coined <strong>FISHQA</strong> (FInancial Sentiment analysis network with Hierarchical Query-driven Attention) for financial sentiment analysis. Multiple user specified queries are contributed to distill document representation with query based attention mechanism. The experiments demonstrate that our framework can learn better representation of the document, unearth meaningful clues on replying different users’ preferences and outperforms the state-of-the-art methods. <a href="https://www.ijcai.org/Proceedings/2018/0590.pdf" target="_blank">[paper]</a><a href="https://github.com/LLluoling/FISHQA" target="_blank">[code]</a></p>
</div>
</div>
</div>
</div>
</div>
</div>
</section><section style="background-color: #fff; font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'Roboto', 'Helvetica Neue', Arial, sans-serif; color:#aaa; font-size:12px; padding: 0; align-items: center; display: flex;"><a href="https://mobirise.site/l" style="flex: 1 1; height: 3rem; padding-left: 1rem;"></a><p style="flex: 0 0 auto; margin:0; padding-right:1rem;"><a href="https://mobirise.site/c" style="color:#aaa;">The</a> page was started with Mobirise</p></section><script src="assets/web/assets/jquery/jquery.min.js"></script> <script src="assets/popper/popper.min.js"></script> <script src="assets/bootstrap/js/bootstrap.min.js"></script> <script src="assets/tether/tether.min.js"></script> <script src="assets/smoothscroll/smooth-scroll.js"></script> <script src="assets/dropdown/js/nav-dropdown.js"></script> <script src="assets/dropdown/js/navbar-dropdown.js"></script> <script src="assets/touchswipe/jquery.touch-swipe.min.js"></script> <script src="assets/theme/js/script.js"></script>
</body>
</html>