- From: Giuseppe Rizzo <giuseppe.rizzo@eurecom.fr>
- Date: Wed, 17 Dec 2014 11:34:03 +0100
- To: semantic-web@w3.org, public-lod@w3.org
====================================================================== Named Entity rEcognition and Linking (NEEL) Challenge at the 5th Making Sense of Microposts Workshop #Microposts2015 @ WWW 2015 http://www.scc.lancs.ac.uk/microposts2015/challenge/index.html 18th/19th May 2015, Florence, Italy ====================================================================== Microposts are a highly popular medium to share facts, opinions or emotions. They are an invaluable wealth of data, ready to be mined for training predictive modellings. Following the success of the last year, we are pleased to announce the NEEL challenge which will be part of the #Microposts2015 Workshop at the World Wide Web 2015 conference. The overall task of the challenge is to automatically recognise entities and their types from English microposts, and to link them to the corresponding English DBpedia 2014 resources (if the linkage exists). Participants will have to automatically extract expressions that are formed by discrete (and typically short) sequences of words (e.g., Obama, London, Rakuten) and recognise their types (e.g., Person, Location, Organisation) from a collection of microposts. As linking stage we aim to disambiguate the spotted entity to the corresponding DBpedia resource, or to a NIL reference if the spotted named entity does not match any resource in DBpedia. This year challenge will also evaluate the end-to-end performance of the system by measuring the computation time for analyzing the corpus using the submitted algorithms. We welcome and hope participants from NEEL, TREC, TAC KBP, ERD shared tasks to participate in this year challenge. DATASET ------- The dataset comprises tweets extracted from a collection of over 18 million tweets. The dataset includes event-annotated tweets provided by the Redites project (http://demeter.inf.ed.ac.uk/redites/) covering multiple noteworthy events from 2011, 2013 (including the death of Amy Winhehouse, the London Riots, the Oslo bombing and the Westgate Shopping Mall shootout) and tweets extracted from the Twitter firehose from 2014. Since the task of this challenge is to automatically recognise and link entities, we have built our dataset considering both event and non-event tweets. While event tweets are more likely to contain entities, non-event tweets enable us to evaluate the performance of the system in avoiding false positives in the entity extraction phase. The training set is built on top of the entire corpus of the NEEL 2014 Challenge. We have further extended it for typing the entities and adding the NIL references. Following the Twitter TOS we will only provide tweet IDs and annotations for the training set; and tweet IDs for the test set. We will also provide a common framework to mine these datasets from Twitter. The training set will be released as tsv following the TAC KBP format, where each line consists of the following features: 1st: tweet id 2nd,3rd: start/end offsets expressed as the number of UTF8 characters starting from 0 (the beginning of the tweet) 4th: link to DBpedia resource or NIL (it may exist different NIL in the corpus. Each NIL may be reused if there are multiple mentions in the text which represent the same entity) 5th: salience (confidence score) 6th: type Tokens are separated by TABs. Entity mentions and URIs are listed according to their appearance order in the tweet. We will timely advertise the release of the data sets on the workshop mailing list. Please subscribe to https://groups.google.com/d/forum/microposts2015. EVALUATION ---------- Participants are required to implement their systems as a publicly accessible web service following a REST based protocol (which will be advertised on the mailing list before the release of the training set) and submit their contending entries (up to 10) to a registry of the NEEL challenge services. Upon receiving the registration of the service, calls to the contending entry will be scheduled in two different time windows, namely D-Time (meant to test the APIs) and T-Time for the final evaluation and metric computations. In the final stage, each participant can submit up to 3 final contending entries. We will use the metrics proposed by TAC KBP 2014 (https://github.com/wikilinks/neleval/wiki/Evaluation) and in particular we will focus on: [tagging] strong_typed_mention_match (check entity name boundary and type) [linking] strong_mention_match [clustering] mention_ceaf (NIL detection) To ensure the correctness of the results and avoid any loss we will trigger N number of calls and we will statistically evaluate the metrics. PAPER SUBMISSION ---------------- A paper of 2 pages describing your approach, how you tuned/tested it using the training split, and your results. All submissions must be in English. All written submissions should be prepared according to the ACM SIG Proceedings Template (see http://www.acm.org/sigs/publications/proceedings-templates), and should include author names and affiliations, and 3-5 author-selected keywords. Where a submission includes additional material submission this should be made as a single, unencrypted zip file that includes a plain text file listing its contents.Submission is via EasyChair, at: https://www.easychair.org/conferences/?conf=microposts2015. Each submission will receive at least 2 peer reviews. The #Microposts2015 proceedings will be published as a single volume containing all three tracks, via CEUR. The same publication conditions however apply as for other workshop proceedings included in the WWW conference companion: "Any paper published by the ACM, IEEE, etc. which can be properly cited constitutes research which must be considered in judging the novelty of a WWW submission, whether the published paper was in a conference, journal, or workshop. Therefore, any paper previously published as part of a WWW workshop must be referenced and suitably extended with new content to qualify as a new submission to the Research Track at the WWW conference." WORKSHOP STRUCTURE ------------------ A keynote address from an invited speaker will open the day, and followed by paper presentations. We will hold a poster and demo session to trigger further, in-depth interaction between workshop participants. The last set of presentations will be brief overviews of selected submissions to the Challenge. The workshop will close with the presentation of awards. IMPORTANT DATES --------------- Intent to participate: 20 Jan 2015 (soft - further instructions will be shared on the mailing list ) Release of the REST API specs: 2 Feb 2015 Release of training set: 15 Feb 2015 Registration of contending entries: 2 Mar 2015 D-Time: 10-15 Mar 2015 (hard) T-Time: 20-25 Mar 2015 (hard) Paper submission: 28 Mar 2015 (hard) Challenge Notification: 21 Apr 2015 (hard) Challenge camera-ready deadline: 31 Apr 2015 (hard) Workshop program issued: 22 Apr 2015 Challenge proceedings to be published via CEUR Workshop - 18/19 May 2015 (Registration open to all) (All deadlines 23:59 Hawaii Time) PRIZE ----- A prize of 1500 euros, generously sponsored by SpazioDati, will be awarded to the challenge winner. SpazioDati is an Italian startup focused on text analytics and big data. One the SpazioDati's key components is DataTXT, a text-analytics engine available on SpazioDati API platform, Dandelion. DataTXT named-entity extraction system has been proven to be very effective and efficient on short and fragmented texts, like microposts. By teaming up with SpazioDati to make the challenge possible, the #Microposts workshop organisers wish to highlight new entity extraction methods and algorithms to pursue in such a challenging scenario. CONTACT ------- Mailing list : https://groups.google.com/forum/#!forum/microposts2015 E-mail: microposts2015@easychair.org Twitter hashtags: #neel #microposts2015 Twitter account: @Microposts2015 W3C Microposts Community Group: http://www.w3.org/community/microposts Challenge Organizers: --------------------- Challenge Chair: A. Elizabeth Cano, Knowledge Media Institute, The Open University, UK Giuseppe Rizzo, EURECOM, France Dataset Chair: Andrea Varga, Swiss Re, UK Challenge Committee: --------------------- Gabriele Antonelli, SpazioDati, Italy Ebrahim Bagheri, Ryerson University, Canada Pierpaolo Basile, University of Bari, Italy Leon Derczynski, The University of Sheffield, UK Milan Dojchinovski, Czech Technical University, Czech Republic Guillaume Ereteo, Vigiglobe, France Andrés García-Silva, Univesidad Politécnica de Madrid, Spain Anna Lisa Gentile, Sheffield, UK Miguel Martinez-Alvarez, Signal, UK Jose M. Morales-Del-Castillo, University of Granada, Spain Georgios Paltoglou, University of Wolverhampton, UK Bernardo Pereira Nunes, PUC-Rio, Brazil Daniel Preoţiuc-Pietro, University of Pennsylvania, USA Giles Reger, The University of Manchester, UK Irina Temnikova, Qatar Computing Research Institute, Qatar Raphaël Troncy, EURECOM, France Victoria Uren, Aston University, UK
Received on Wednesday, 17 December 2014 10:35:48 UTC