Navigation

Important Dates

closed	Registration
Mar 16, 2012	Training data release
May 18, 2012	Test data release
Jun 15, 2012	Run submission
Jun 22, 2012	Abstract submission
Jul 06, 2012	Result notification
Jul 06, 2012	Review notification
Aug 17, 2012	Notebook submission
Sep 17-20, 2012	Conference

Contact

Email	pan@webis.de
Webpage	http://pan.webis.de

Mailing List

Related Events

PAN @ FIRE'12

Previous Events

PAN @ FIRE'11
PAN @ CLEF'11
PAN @ CLEF'10
PAN @ SEPLN'09
PAN @ ECAI'08
PAN @ SIGIR'07

Supported by

Author Identification

Task Description
Evaluation Corpus
Performance Measures
Resources
Run Submission
Evaluation Results
Task Committee

Task Description

This year we divided the task into two different sub-task:

Traditional Authorship Attribution:
Within the traditional authorship tasks there are different flavors:
- Traditional (closed-class /open-class, with varying numbers of candidate authors) authorship attribution. Within the closed class you will be given a closed set of candidate authors and are asked to identify which one of them is the author of an anonymous text. Withing the open class you have to consider also that it might be that none of the candidates is the real author of the document.
- Authorship clustering/intrinsic plagiarism: in this problem you are given a text (which, for simplicity, is segmented into a sequence of "paragraphs") and are asked to cluster the paragraphs into exactly two clusters: one that includes paragraphs written by the "main" author of the text and another that includes all paragraphs written by anybody else. (Thus, this year the intrinsic plagiarism has been moved from the plagiarism task to the author identification track.).
Sexual Predator Identification:
The goal of this sub-task is to identify classes of authors, namely online predators. You will be given chat logs involving two (or more) people and have to determine who is the one trying to convince the other partecipants(s) to provide some sexual favour . You will also need to identify the particular conversation where the person exploits his bad behavior.
The task can therefore be divided into two parts:
1. Identify the predators (within all the users)
2. Identify the part (the lines) of the predator conversations which are the most distinctive of the predator bad behavior
Given the public nature of the dataset, we ask the participants not to use external or online resources for resolving this task (e.g. search engines) but to extract evidence from the provided datasets only.

Evaluation Corpus

For each of the two sub-tasks your will be given separate evaluation resources.

Traditional Authorship Attribution:
- Training corpus:
  pan12-authorship-attribution-training-corpus-2012-03-28.tar.gz
  (256 KB, MD5 sum: 6665b5f5122679bb62cec1c46fcf25aa)
  pan12-authorship-attribution-training-corpus-supplement-2012-04-03.tar.gz
  (2.9 MB, MD5 sum: ca4ae34a8f70faafd043edb63abb7db1)
  pan12-authorship-attribution-training-corpus-supplement-2012-5-24.tar.gz
  (4.9 MB, MD5 sum: 73e2dd17443af045dca9b477300b6f89)
- Test corpus:
  pan12-authorship-attribution-test-corpus-2012-05-24.tar.gz
  (5.7 MB, MD5 sum: f3351bb3561c65a31daf01c7ad6400d7)
- Ground truth:
  pan12-authorship-attribution-groundtruth.txt
  (2.2 KB, MD5 sum: df5da3d972a473da87beb1a0bc3bc395)
Sexual Predator Identification:
- Training corpus:
  pan12-sexual-predator-identification-training-data-2012-03-31.zip
  (26.7 MB, MD5 sum: 90f324497f860c5d564e2016f3f437f3)
  pan12-sexual-predator-identification-training-data-2012-05-01.zip
  (26.2 MB, MD5 sum: 9ef55e76273be1884adfdcbdcf46f103)
- Test corpus:
  pan12-sexual-predator-identification-test-data-2012-05-21.zip
  (57.6 MB, MD5 sum: 2ad5aabbb8ac637989521bfa1869bca3)
- Ground truth:
  pan12-sexual-predator-identification-groundtruth.zip
  (38.3 KB, MD5 sum: 04b0a918f0b562b20960171df9516b56)

Performance Measures

For each of the two sub-tasks the performance be determined based on standard performance measures:

Traditional Authorship Attribution:
The performance of your authorship attribution will be judged by average precision, recall, and F1 over all authors in the given training set. A reference implementation will be forthcoming.
Sexual Predator Identification:
The performance of your authorship attribution will be judged by average precision, recall, and F1 over all persons involved and lines of the conversations.

Resources

For an overview of approaches to automated authorship attribution, we refer you to recent survey papers in the area:

Patrick Juola. Authorship Attribution. In Foundations and Trends in Information Retrieval, Volume 1, Issue 3, December 2006.
Moshe Koppel, Jonathan Schler, and Shlomo Argamon. Computational Methods in Authorship Attribution. Journal of the American Society for Information Science and Technology, Volume 60, Issue 1, pages 9-26, January 2009.
Efstathios Stamatatos. A Survey of Modern Authorship Attribution Methods. Journal of the American Society for Information Science and Technology, Volume 60, Issue 3, pages 538-556, March 2009.

Run Submission

Traditional Authorship Attribution:
As per repeated requests, here is a sample submission format to use for the Traditional Authorship Attribution Competition for PAN/CLEF. Please note that following this format is not mandatory and we will continue to accept anything we can interpret.
For traditional authorship problems (e.g. problem A), use the following (all words in ALL CAPS should be filled out appropriately):
```
team TEAM NAME : run RUN NUMBER
task TASK IDENTIFIER
file TEST FILE = AUTHOR IDENTIFIER
file TEST FILE = AUTHOR IDENTIFIER
...
```
For problems E and F, there are no designated sample authors, so we recommend listing paragraph numbers. Author identifier is optional and arbitrary -- if it makes you feel better to talk about authors A and B or authors 1 and 2 you can insert it into the appropriate field. Any paragraphs not listed will be assumed to be part of an unnamed default author.
```
team TEAM NAME : run RUN NUMBER
task TASK IDENTIFIER
file TEST FILE = AUTHOR IDENTIFIER (PARAGRAPH LIST)
file TEST FILE = AUTHOR IDENTIFIER
...
```
For example:
```
team Jacob : run 1
task B
file 12Btest01.txt = A
file 12Btest02.txt = A
file 12Btest03.txt = A
file 12Btest04.txt = None of the Above
file 12Btest05.txt = A
file 12Btest06.txt = A
file 12Btest07.txt = A
file 12Btest08.txt = A
file 12Btest09.txt = A
file 12Btest10.txt = A

task C
file 12Ctest01.txt = A
file 12Ctest02.txt = A
file 12Ctest03.txt = A
file 12Ctest04.txt = A
file 12Ctest05.txt = A
file 12Ctest06.txt = A
file 12Ctest07.txt = A
file 12Ctest08.txt = A
file 12Ctest09.txt = A

task F
file 12Ftest01.txt = (1,2,3,6,7)
file 12Ftest01.txt = (4,5)
```
In this sample file, we consider anything not listed in task F (paragraphs 8 and beyond) to be a third, unnamed author.

Sexual Predator Identification:

For each of the two parts we require a different format.

Identify the predators (within all the users)

Participants should update a text file containing an user-id per line, of those identified as predator only:
```
…
a7c5056a2c30e2dc637907f448934ca3
58f15bbb100bbeb6963b4b967ce04bdf
e040eb115e3f7ad3824e93141665fc2a
3d57ed3fac066fa4f8a52432db51c019
…
```

Identify the part (the lines) of the predator conversations which are the most distinctive of the predator bad behavior

Participants should update an xml file similar to the corpus ones, containing conversation-ids and message line numbers considered suspicious (line numbers together with all the others message information: author, time, text):

<conversations>
	…
	<conversation id="0042762e26ed295a8576806f5548cad9">
		<message line="3">
			<author>f069dbec9ab3e090972d432db279e3eb</author>
			<time>03:20</time>
			<text>whats up?</text>
		</message>
		<message line="4">
			<author>f069dbec9ab3e090972d432db279e3eb</author>
			<time>03:21</time>
			<text>how u doing?</text>
		</message>
		…
		<message line="10">
			<author>f069dbec9ab3e090972d432db279e3eb</author>
			<time>04:00</time>
			<text>sse you llater?</text>
		</message>
	</conversation>
	…
	<conversation id="0209b0a30c8eced86863631ada73a530">
		<message line="3">
			<author>0042762e26ed295a8576806f5548cad9</author>
			<time>01:17</time>
			<text>and that i dont touch u</text>
		</message>
	</conversation>
	…
<conversations>

Once prepared, please submit your runs via mail to pan@webis.de. You may submit more than one run, however, please limit yourself to a reasonable number of submissions. With regard to an overall ranking of participants, only the latest run submitted is used. If you submit more than one run at a time, please rank your runs accordingly so that it is clear to us which run is your preferred choice.

Evaluation Results

Traditional Authorship Attribution:

The results of the traditional authorship attribution task can be found here.

Sexual Predator Identification:

Due to an error in computing the F measure (with β=0.5 and β=3; β was not squared) and to the addition of 4 more predators (which produced 76 more lines) in the ground truth, results have sligtly changed compared to the first release. Please find the updated results below and grund truth in the corpus section above.

For the predator identification subtask we received 16 submissions for the first part of the problem (identifying the predators) and 14 for the second part (identifying the distinctive chat lines of the predator behavior).

We provide in the tables below the results for the first problem, for the main run of each team and for all the submitted runs. We evaluate the results with Precision (P: retrieved authors that are relevant), Recall (R: relevant authors that are retrieved) and F1 measure (weighted harmonic mean between Precision and Recall, with β factor equal to 1). If we interpret the results in a realistic scenario, we might observe that retrieving lot of relevant authors is important (Recall), since a police agent would like to receive the major number of suspect. However, what is more important is the fact that the retrieved authors are relevant (Precision), to optimize the time of a police agent towards the "right" suspect rather than "all" the possible suspects. For this reason we introduced another measure of F, with the β factor equal to 0.5, for emphasizing the precision. As it can be observed in the table, this influences the first 3 positions of the ranks only.

Preliminary results of the Sexual Predator Identification: 1) Identify the predators
Participant main run	Retrieved	Relevant	P	R	F(β=1)	F(β=0.5)	Rank F(β=0.5)
							.
villatorotello-run-2012-06-15-2157g	204	196	0.9608	0.7840	0.8634	0.8936	1
snider12-run-2012-06-16-0032	186	181	0.9731	0.7240	0.8303	0.8730	2
eriksson12-run-2012-06-15-1949	265	223	0.8415	0.8920	0.8660	0.8577	3
parapar12-run-2012-06-15-0959j	181	168	0.9282	0.6720	0.7796	0.8235	4
morris12-run-2012-06-16-0752-main	159	152	0.9560	0.6080	0.7433	0.8028	5
peersman12-run-2012-06-15-1559	170	148	0.8706	0.5920	0.7048	0.7525	6
grozea12-run-2012-06-14-1706b	215	160	0.7442	0.6400	0.6882	0.7059	7
sitarz12-run-2012-0615-1515	218	156	0.7156	0.6240	0.6667	0.6822	8
vartapetiance12-run-2012-06-15-1411	160	97	0.6063	0.3880	0.4732	0.5105	9
kontostathis-run-2012-06-16-0317e	475	167	0.3516	0.6680	0.4607	0.4175	10
kang12-run-2012-06-15-0904b	930	199	0.2140	0.7960	0.3373	0.2829	11
kern12-run-2012-06-18-1827b	1172	173	0.1476	0.6920	0.2433	0.2001	12
bogdanova12-run-2012-06-14-1117	2109	54	0.0256	0.2160	0.0458	0.0363	13
prasath12-run-2012-06-15-2122	10289	204	0.0198	0.8160	0.0387	0.0294	14
vilarino12-run-2012-06-14-2121b	5225	97	0.0186	0.3880	0.0354	0.0272	15
gomezhidalgo12-2012-06-15-1900	150	1	0.0067	0.0040	0.0050	0.0055	16

Updated results of the Sexual Predator Identification: 1) Identify the predators
Participant main run	Retrieved	Relevant	P	R	F(β=1)	F(β=0.5)	Rank F(β=0.5)
							.
villatorotello-run-2012-06-15-2157g	204	200	0.9804	0.7874	0.8734	0.9346	1
snider12-run-2012-06-16-0032	186	183	0.9839	0.7205	0.8318	0.9168	2
parapar12-run-2012-06-15-0959j	181	170	0.9392	0.6693	0.7816	0.8691	3
morris12-run-2012-06-16-0752-main	159	154	0.9686	0.6063	0.7458	0.8652	4
eriksson12-run-2012-06-15-1949	265	227	0.8566	0.8937	0.8748	0.8638	5
peersman12-run-2012-06-15-1559	170	152	0.8941	0.5984	0.7170	0.8137	6
grozea12-run-2012-06-14-1706b	215	163	0.7581	0.6417	0.6951	0.7316	7
sitarz12-run-2012-0615-1515	218	159	0.7294	0.6260	0.6737	0.7060	8
vartapetiance12-run-2012-06-15-1411	160	99	0.6188	0.3898	0.4783	0.5537	9
kontostathis-run-2012-06-16-0317e	475	170	0.3579	0.6693	0.4664	0.3946	10
kang12-run-2012-06-15-0904b	930	203	0.2183	0.7992	0.3429	0.2554	11
kern12-run-2012-06-18-1827b	1172	177	0.1510	0.6969	0.2482	0.1791	12
bogdanova12-run-2012-06-14-1117	2109	55	0.0261	0.2165	0.0466	0.0316	13
prasath12-run-2012-06-15-2122	10289	207	0.0201	0.8150	0.0393	0.0250	14
vilarino12-run-2012-06-14-2121b	5225	98	0.0188	0.3858	0.0358	0.0232	15
gomezhidalgo12-2012-06-15-1900	150	1	0.0067	0.0039	0.0050	0.0059	16

Updated results (all the runs) for the Sexual Predator Identification: 1) Identify the predators
Participant run	Retrieved	Relevant	P	R	F(β=1)	F(β=0.5)
						.
bogdanova12-run-2012-06-14-1117	2109	55	0.0261	0.2165	0.0466	0.0316
eriksson12-run-2012-06-15-1949	265	227	0.8566	0.8937	0.8748	0.8638
gomezhidalgo12-2012-06-15-1900	150	1	0.0067	0.0039	0.0050	0.0059
grozea12-run-2012-06-14-1706a	322	142	0.4410	0.5591	0.4931	0.4604
grozea12-run-2012-06-14-1706b	215	163	0.7581	0.6417	0.6951	0.7316
kang12-run-2012-06-15-0904a	1049	202	0.1926	0.7953	0.3101	0.2270
kang12-run-2012-06-15-0904b	930	203	0.2183	0.7992	0.3429	0.2554
kern12-run-2012-06-18-1827a	1172	177	0.1510	0.6969	0.2482	0.1791
kern12-run-2012-06-18-1827b	1172	177	0.1510	0.6969	0.2482	0.1791
kontostathis-run-2012-06-16-0317a	5225	206	0.0394	0.8110	0.0752	0.0487
kontostathis-run-2012-06-16-0317b	5625	221	0.0393	0.8701	0.0752	0.0486
kontostathis-run-2012-06-16-0317c	3696	206	0.0557	0.8110	0.1043	0.0685
kontostathis-run-2012-06-16-0317d	688	172	0.2500	0.6772	0.3652	0.2861
kontostathis-run-2012-06-16-0317e	475	170	0.3579	0.6693	0.4664	0.3946
morris12-run-2012-06-16-0752-main	159	154	0.9686	0.6063	0.7458	0.8652
morris12-run-2012-06-17-0126	152	147	0.9671	0.5787	0.7241	0.8527
parapar12-run-2012-06-15-0959a	200	128	0.6400	0.5039	0.5639	0.6072
parapar12-run-2012-06-15-0959b	205	160	0.7805	0.6299	0.6972	0.7449
parapar12-run-2012-06-15-0959c	169	145	0.8580	0.5709	0.6856	0.7796
parapar12-run-2012-06-15-0959d	175	151	0.8629	0.5945	0.7040	0.7914
parapar12-run-2012-06-15-0959e	182	164	0.9011	0.6457	0.7523	0.8350
parapar12-run-2012-06-15-0959f	202	154	0.7624	0.6063	0.6754	0.7250
parapar12-run-2012-06-15-0959g	171	162	0.9474	0.6378	0.7624	0.8635
parapar12-run-2012-06-15-0959h	223	161	0.7220	0.6339	0.6751	0.7024
parapar12-run-2012-06-15-0959i	173	161	0.9306	0.6339	0.7541	0.8510
parapar12-run-2012-06-15-0959j	181	170	0.9392	0.6693	0.7816	0.8691
peersman12-run-2012-06-15-1559	170	152	0.8941	0.5984	0.7170	0.8137
prasath12-run-2012-06-15-2122	10289	207	0.0201	0.8150	0.0393	0.0250
sitarz12-run-2012-0615-1515	218	159	0.7294	0.6260	0.6737	0.7060
snider12-run-2012-06-16-0032	186	183	0.9839	0.7205	0.8318	0.9168
vartapetiance12-run-2012-06-15-1411	160	99	0.6188	0.3898	0.4783	0.5537
vilarino12-run-2012-06-14-2121a	9071	236	0.0260	0.9291	0.0506	0.0323
vilarino12-run-2012-06-14-2121b	5225	98	0.0188	0.3858	0.0358	0.0232
villatorotello-run-2012-06-15-2157a	108	103	0.9537	0.4055	0.5691	0.7507
villatorotello-run-2012-06-15-2157b	204	12	0.0588	0.0472	0.0524	0.0561
villatorotello-run-2012-06-15-2157c	211	200	0.9479	0.7874	0.8602	0.9107
villatorotello-run-2012-06-15-2157d	240	36	0.1500	0.1417	0.1457	0.1483
villatorotello-run-2012-06-15-2157e	305	6	0.0197	0.0236	0.0215	0.0204
villatorotello-run-2012-06-15-2157f	269	143	0.5316	0.5630	0.5468	0.5376
villatorotello-run-2012-06-15-2157g	204	200	0.9804	0.7874	0.8734	0.9346

We report in this other table below all the results for the second problem. An expert manually evaluated all the lines that where returned at least by 1 participant (these accounts for more than 90% of all the predator lines). As in the first problem, we computed Precision (P: retrieved lines that are relevant), Recall (R: relevant lines that are retrieved) and F1 measure (weighted harmonic mean between Precision and Recall, with β factor equal to 1). Still referring to the first problem, if we think at a realistic scenario we might noticed that in this second problem retrieving lot of relevant lines (Recall) is more important that finding only the relevant ones (Precision). Having lot of relevant lines augment the possibility of finding good evidences towards a suspect. For this reason we introduced another measure of F, with the β factor equal to 3, for emphasizing the recall, that slightly modifies the upper part of the ranking compared to the standard F1.

Preliminary results of the Sexual Predator Identification: 2) Identify predators line
Participant main run	Retrieved	Relevant	P	R	F(β=1)	F(β=3)	Rank F(β=3)
							.
kontostathis-run-2012-06-16-0317e	19535	3215	0.1646	0.5022	0.2479	0.3319	1
grozea12-run-2012-06-14-1706b	63290	5715	0.0903	0.8927	0.1640	0.2771	2
peersman12-run-2012-06-15-1559	4717	1650	0.3498	0.2577	0.2968	0.2759	3
sitarz12-run-2012-0615-1515	4558	1469	0.3223	0.2295	0.2681	0.2473	4
morris12-run-2012-06-16-0752-main	2685	1195	0.4451	0.1867	0.2630	0.2184	5
kern12-run-2012-06-18-1827b	15533	1328	0.0855	0.2074	0.1211	0.1529	6
eriksson12-run-2012-06-15-1949	10416	1116	0.1071	0.1743	0.1327	0.1507	7
prasath12-run-2012-06-15-2122	77255	1041	0.0135	0.1626	0.0249	0.0432	8
vartapetiance12-run-2012-06-15-1411	607	91	0.1499	0.0142	0.0260	0.0184	9
parapar12-run-2012-06-15-0959j	2037	96	0.0471	0.0150	0.0228	0.0181	10
vilarino12-run-2012-06-14-2121b	6787	47	0.0069	0.0073	0.0071	0.0072	11
bogdanova12-run-2012-06-14-1117	49	4	0.0816	0.0006	0.0012	0.0008	12
villatorotello-run-2012-06-15-2157g	50	1	0.0200	0.0002	0.0003	0.0002	13
gomezhidalgo12-2012-06-15-1900	400	0	0.0000	0.0000	0.0000	0.0000	14

Updated results of the Sexual Predator Identification: 2) Identify predators line
Participant main run	Retrieved	Relevant	P	R	F(β=1)	F(β=3)	Rank F(β=3)
							.
grozea12-run-2012-06-14-1706b	63290	5790	0.0915	0.8938	0.1660	0.4762	1
kontostathis-run-2012-06-16-0317e	19535	3249	0.1663	0.5015	0.2498	0.4174	2
peersman12-run-2012-06-15-1559	4717	1688	0.3579	0.2606	0.3016	0.2679	3
sitarz12-run-2012-0615-1515	4558	1486	0.3260	0.2294	0.2693	0.2364	4
morris12-run-2012-06-16-0752-main	2685	1211	0.4510	0.1869	0.2643	0.1986	5
kern12-run-2012-06-18-1827b	15533	1357	0.0874	0.2095	0.1233	0.1838	6
eriksson12-run-2012-06-15-1949	10416	1122	0.1077	0.1732	0.1328	0.1633	7
prasath12-run-2012-06-15-2122	77255	1044	0.0135	0.1612	0.0249	0.0770	8
parapar12-run-2012-06-15-0959j	2037	105	0.0515	0.0162	0.0247	0.0174	9
vartapetiance12-run-2012-06-15-1411	607	91	0.1499	0.0140	0.0257	0.0154	10
vilarino12-run-2012-06-14-2121b	6787	48	0.0071	0.0074	0.0072	0.0074	11
bogdanova12-run-2012-06-14-1117	49	4	0.0816	0.0006	0.0012	0.0007	12
villatorotello-run-2012-06-15-2157g	50	1	0.0200	0.0002	0.0003	0.0002	13
gomezhidalgo12-2012-06-15-1900	400	0	0.0000	0.0000	0.0000	0.0000	14

Reference for the evaluation measures (and beta values): http://nlp.stanford.edu/IR-book/html/htmledition/evaluation-of-unranked-retrieval-sets-1.html (last check: August 9, 2012)

Sexual Predator Identification: Participants
Partecipant main run	Partecipant and affiliation
					.
villatorotello-run-2012-06-15-2157g	Esaú Villatoro-Tello Instituto Nacional de Astrfísica, Óptica y Electrónica (INAOE) and Universidad Autónoma Metropolitana Mexico	Antonio Juárez-González Instituto Nacional de Astrfísica, Óptica y Electrónica (INAOE) Mexico	Hugo J. Escalante Instituto Nacional de Astrfísica, Óptica y Electrónica (INAOE) Mexico	Manuel Montes-y-Gómez Instituto Nacional de Astrfísica, Óptica y Electrónica (INAOE) Mexico	Luis Villaseñor-Pineda Instituto Nacional de Astrfísica, Óptica y Electrónica (INAOE) Mexico
snider12-run-2012-06-16-0032	Tim Snider Porfiau Inc. Canada
eriksson12-run-2012-06-15-1949	Gunnar Eriksson Gavagai AB Sweden	Jussi Karlgren Gavagai AB Sweden
parapar12-run-2012-06-15-0959j	Javier Parapar University of A Coruña Spain	David E. Losada Universidade de Santiago de Compostela Spain		Alvaro Barreiro University of A Coruña Spain
morris12-run-2012-06-16-0752-main	Colin Morris University of Toronto Canada	Graeme Hirst University of Toronto Canada
peersman12-run-2012-06-15-1559	Claudia Peersman University of Antwerp Netherlands	Frederik Vaassen University of Antwerp Netherlands	Vincent Van Asch University of Antwerp Netherlands	Walter Daelemans University of Antwerp Netherlands
grozea12-run-2012-06-14-1706b	Cristian Grozea Fraunhofer Institute FIRST Germany	Marius Popescu University of Bucharest Romania
sitarz12-run-2012-0615-1515	Rachel Sitarz Purdue University United States
vartapetiance12-run-2012-06-15-1411	Anna Vartapetiance University of Surrey UK	Lee Gillam University of Surrey UK
kontostathis-run-2012-06-16-0317e	April Kontostathis Ursinus College USA	Andy Garron The University of Maryland USA	Kelly Reynolds Lehigh University USA	Will West Lehigh University USA	Lynne Edwards Ursinus College USA
kang12-run-2012-06-15-0904b	In-Su Kang Kyungsung University South Korea	Chul-Kyu Kim Kyungsung University South Korea	Shin-Jae Kang Daegu University South Korea	Seung-Hoon Na Electronics and Telecommunications Research Institute South Korea
kern12-run-2012-06-18-1827b	Roman Kern Graz University of Technology and Know-Center GmbH Austria	Stefan Klampfl Know-Center GmbH Austria	Mario Zechner Know-Center GmbH Austria
bogdanova12-run-2012-06-14-1117	Dasha Bogdanova University of Saint Petersburg Russia	Paolo Rosso Universitat Politècnica de València Spain
prasath12-run-2012-06-15-2122	Sriram Prasath Elango KTH/Gavagai Sweden
vilarino12-run-2012-06-14-2121b	Darnes Vilariño Benemérita Universidad Autónoma de Puebla Mexico	Esteban Castillo Benemérita Universidad Autónoma de Puebla Mexico	David Pinto Benemérita Universidad Autónoma de Puebla Mexico	Iván Olmos Benemérita Universidad Autónoma de Puebla Mexico	Saul León Benemérita. Universidad Autonóma de Puebla Mexico
gomezhidalgo12-2012-06-15-1900	José María Gómez Hidalgo Optenet Spain	Andrés Alfonso Caurcel Díaz Universidad Politécnica de Madrid Spain

Task Committee

Patrick Juola
Duquesne University

Shlomo Argamon
Illinois Institute of Technology

Efstathios Stamatatos
University of the Aegean

Moshe Koppel
Bar-Ilan University

Giacomo Inches and Fabio Crestani
IRGroup @ University of Lugano