PageSourceSearch

https://michaeljpaul.com/resources.html

html michaeljpaul.com collected 2026-10-02 01:35:46 UTC 5,519 bytes, 165 lines download raw bytes

1
2
3<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
4<html>
5  <head>
6  <title>Michael J. Paul - Homepage</title>
7  
8  <meta http-equiv="content-type" content="text/html">
9  
10	<link rel="stylesheet" href="styles.css" type="text/css" media="screen" />
11<style type="text/css">
12#photo {
13width:175px;
14height:211px;
15float: right;
16position: relative;
17right: 10px;
18bottom: 0px;
19margin-left: 20px;
20margin-bottom: -20px;
21border:2px solid black;
22-webkit-border-radius: 5px;
23-moz-border-radius: 5px;
24border-radius: 5px;
25}
26#news {
27height:15ex;
28overflow: scroll; 
29overflow-y: scroll;}
30#travel {
31height:130px;
32overflow:auto;
33}
34.newsli {
35margin-bottom: 4px;
36}
37.leftul {
38margin-left: -1.5em;
39margin-bottom: 0em;
40}
41</style>
42
43<script src="http://platform.twitter.com/widgets.js" type="text/javascript"></script>
43
44
45</head>
46<body>
47  
48<table border="0" width="100%" cellspacing="0" cellpadding="0">
49
50<tr><td id="corner">&nbsp;</td>
51 <td id="topbar">Michael J. Paul</td>
52</tr>
53
54<tr>
55  
56<td width="85" id="leftbar">&nbsp;</td>
57<td width="750" id="content">
58
59     
60  <p class="intro" style="margin-bottom:20px">Here is an archive of my academic work (see also <a href="https://scholar.google.com/citations?user=_pRB0HIAAAAJ&hl=en">Google Scholar</a>):</p>
61
62 
63<div class="tabbar"><div class="tab" style="margin-left: 15px;"><a href="index.html" class="tablink" style="">Research</a></div>
64<div class="tab" style=""><a href="teaching.html" class="tablink" style="">Teaching</a></div>
65<div class="tab" style=""><a href="group.html" class="tablink" style="">Group</a></div>
66<div class="tab" style="border: 3px solid #353D57;background-color:#dddddd;"><a href="resources.html" class="tablink" style="font-weight:bold;color:black;">Resources</a></div>
67<div class="tab" style=""><a href="book.html" class="tablink" style="">Book</a></div></div>
68<p class="sectionhead">Code</p>
69
70   <ul style="margin-left:-1em;margin-top:-1em;">
71
72    <li class="list1"><a href="https://bitbucket.org/adrianbenton/sprite/"><span style="font-size: 11pt">S</span><span style="font-size: 9pt;font-weight:bold;">PRITE</span></a></li>
73    <li class="list1"><a href="downloads/flda.php">Factorial LDA</a></li>
74    <li class="list1"><a href="https://github.com/michaeljpaul/sequence-models">Topic modeling for document sequences</a></li>
75      <ul>
76       <li class="list2">Java implementations of the Block HMM and Mixed Membership Markov Model (M4)</li>
77      </ul>
78    <li class="list1"><a href="downloads/mftm.php">Multi-Faceted Topic Modeling Package</a></li>
79      <ul>
80       <li class="list2">Java implementations of Cross-Collection LDA and the Topic-Aspect Model</li>
81      </ul>
82    <li class="list1"><a href="https://github.com/mdredze/carmen">Carmen: Geolocation for Twitter</a></li>
83
84   </ul>
85
86
87<p class="sectionhead">Data</p>
88
89   <ul style="margin-left:-1em;margin-top:-1em;">
90
91    <li class="list1"><a href="https://doi.org/10.6084/m9.figshare.6025697.v1">Zika Tweets, 2015-2016</a></li>
92      <ul>
93       <li class="list2">Data used in Pruss et al., PLOS ONE 2019.</li>
94      </ul>
95    <li class="list1"><a href="https://github.com/abenton/collsuptmdata">Tweets for Survey Prediction</a></li>
96      <ul>
97       <li class="list2">Data used in Benton et al., AAAI 2016.</li>
98      </ul>
99    <li class="list1"><a href="http://www.jmir.org/article/downloadSuppFile/3875/18525">Weibo Air Pollution Dataset</a></li>
100      <ul>
101       <li class="list2">Data used in Wang et al., JMIR 2015.</li>
102      </ul>
103    <li class="list1"><a href="https://figshare.com/articles/Discovering_health_topics_in_social_media_using_topic_models/1007712">Health Tweets</a></li>
104      <ul>
105       <li class="list2">Data used in Paul and Dredze, PLOS 2014. Includes health-related tweet IDs and ATAM output.</li>
106      </ul>
107    <li class="list1"><a href="http://www.byronwallace.com/static/ratemd-dr-ratings.csv.zip">Doctor Review Dataset</a></li>
108      <ul>
109       <li class="list2">Data used in Wallace et al., JAMIA 2014.</li>
110      </ul>
111    <li class="list1"><a href="downloads/flu_data.php">Influenza Twitter Annotations</a></li>
112      <ul>
113       <li class="list2">Data used in Lamb et al., NAACL 2013. Tweet IDs annotated with flu relevance.</li>
114      </ul>
115    <li class="list1"><a href="http://www.cs.jhu.edu/~mdredze/datasets/health_tweetIDs.txt">Health Twitter Annotations</a></li>
116      <ul>
117       <li class="list2">Data used in Paul and Dredze, ICWSM 2011. Tweet IDs annotated with health relevance.</li>
118      </ul>
119    <li class="list1"><a href="downloads/ccdata.php">Cross-Cultural Blog and Forum Dataset</a></li>
120      <ul>
121       <li class="list2">Data used in Paul and Girju, EMNLP 2009.</li>
122      </ul>
123
124   </ul>
125
126
127<p class="sectionhead">Other Resources</p>
128
129   <ul style="margin-left:-1em;margin-top:-1em;">
130
131    <li class="list1">The code from my group's recent projects can be found on the GitHub pages for <a href="https://github.com/xiaoleihuang">Xiaolei Huang</a> and <a href="https://github.com/akkikiki">Yoshinari Fujinuma</a>.</li>
132    <li class="list1">Some annotated tweets are available from the <a href="https://healthlanguageprocessing.org/smm4h/">SMM4H shared tasks</a>.</li>
133    <li class="list1">See <a href="http://dredze.com/code.php">Mark Dredze's website</a> for other code/data that I have worked with.</li>
134     
135   </ul>
136
137
138
139
140
141
142
143
144
145
146
147   
148<br / ><br /><br /><br />
149<br / ><br /><br /><br />
150<br / ><br /><br /><br />
151
152 
153  
154  <br />
155  <br />
156    <br />
157  <br /><br /><br />
158</td>
159
160<td>&nbsp;</td>
161</tr></table>
162
163
164</body>
165</html>

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.