Apologies for cross-posting,
Submissions are invited for MASALA (Machine-learning Approaches to Sentiment
Analysis and Learning Algorithms), an ICML14 workshop exploring the new
frontiers of big data computing for opinion mining through machine-learning
techniques and sentiment learning methods. For more information, please visit:
http://sentic.net/masala
RATIONALE
The distillation of knowledge from social media is an extremely difficult task
as the content of today's Web, while perfectly suitable for human consumption,
remains hardly accessible to machines. The opportunity to capture the opinions
of the general public about social events, political movements, company
strategies, marketing campaigns, and product preferences has raised growing
interest both within the scientific community, leading to many exciting open
challenges, as well as in the business world, due to the remarkable benefits to
be had from marketing and financial market prediction.
Statistical NLP has been the mainstream NLP research direction since late 1990s.
It relies on language models based on popular machine-learning algorithms such
as maximum-likelihood, expectation maximization, conditional random fields, and
support vector machines. By feeding a large training corpus of annotated texts
to a machine-learning algorithm, it is possible for the system to not only learn
the valence of keywords, but also to take into account the valence of other
arbitrary keywords, punctuation, and word co-occurrence frequencies. However,
standard statistical methods are generally semantically weak if they merely
focus on lexical co-occurrence elements with little predictive value
individually.
Endogenous NLP, instead, involves the use of machine-learning techniques to
perform semantic analysis of a corpus by building structures that approximate
concepts from a large set of documents. It does not involve prior semantic
understanding of documents; instead, it relies only on the endogenous knowledge
of these (rather than on external knowledge bases). The advantages of this
approach over the knowledge engineering approach are effectiveness, considerable
savings in terms of expert manpower, and straightforward portability to
different domains. Endogenous NLP includes methods based either on lexical
semantics, which focuses on the meanings of individual words (e.g., LSA, LDA,
and MapReduce), or compositional semantics, which looks at the meanings of
sentences and longer utterances (e.g., HMM, association rule learning, and
probabilistic generative models).
TOPICS
MASALA aims to provide an international forum for researchers in the field of
machine learning for opinion mining and sentiment analysis to share information
on their latest investigations in social information retrieval and their
applications both in academic research areas and industrial sectors. The broader
context of the workshop comprehends opinion mining, social media marketing,
information retrieval, and natural language processing. Topics of interest
include but are not limited to:
• Endogenous NLP for sentiment analysis
• Sentiment learning algorithms
• Big social data analysis
• Opinion retrieval, extraction, classification, tracking and summarization
• Domain specific sentiment analysis and model adaptation
• Emotion detection
• Sentiment pattern mining
• Concept-level sentiment analysis
• Biologically-inspired opinion mining
• Social-network motivated methods for natural language processing
• Topic modeling for aspect-based sentiment analysis
• Learning to rank for social media
• Content-based and social-based recommendation
• Multimodal sentiment analysis
• Content-, concept-, and context-based sentiment analysis
TIMEFRAME
• April 20th, 2014: Submission deadline
• May 11th, 2014: Notification of acceptance
• May 18th, 2014: Final manuscripts due
• June 25th, 2014: Workshop date
ORGANIZERS
• Yunqing Xia, Tsinghua University (China)
• Erik Cambria, National University of Singapore (Singapore)
• Newton Howard, MIT Media Laboratory (USA)
Hoi,
In the latest iteration of the "Reasonator", you will find the option to
see what items exist in a radius of 15 km around the current item.
Obviously, you will find only those items that have a geo-coordinate. There
are however many of them :) They are everything from cities to amusement
parks like Disney land and its rides.
Essentially the functionality shows what is the result of a query.
The current maps are based on OpenStreetMap and Wikidata. This is a first
iteration of this functionality so we love to hear from you what more and
what else you would like us to do with maps and Wikidata.
Thanks,
GerardM
http://ultimategerardm.blogspot.nl/2014/03/reasonator-cambridge-revisited-i…
There is interaction between Wikidata, the OHM, the historians working with
gazetteers, LOD researchers and Jochen Topf & Tim Alder's work. The
Wikimaps project is trying to stay abreast of the development to build on
that.
I think also that Wikidata will lead the way and will offer a crowdsourced
platform for place names across times. The open questions would be related
to the choice of labels when displaying, while Wikidata itself would be
able to store many different names, languages and alternatives.
Discussion is needed for the modelling, eg. if a place is one entity with
changing properties or if a place is a linked continuum of separate places.
What properties to store, how to link? How can the data be linked to say
OSM DB entities? Do the notability guidelines of Wikimedia allow storing
only important places?
So, in short, the most natural site for discussion is the wikidata-l list
(now cc:d)
Best,
Susanna Ånäs
wikimaps.wikimedia.fi
2014-03-19 22:59 GMT+02:00 Laurence Penney <lorp(a)lorp.org>:
> It's great to have such things mapped, but it does need care.
>
> In this field Jochen Topf coded "Multilingual Map Test" together back in
> 2012. You might ask him to add Finnish to the languages offered.
>
> https://lists.openstreetmap.org/pipermail/talk/2012-November/065312.html
>
> Here's part of Poland, shown with German labels:
>
>
> http://mlm.jochentopf.com/?zoom=7&lat=52.57802&lon=19.11621&layers=B0T&lang…
>
> While the larger cities have well-known and current German names that are
> uncontroversial -- Warschau, Posen, Breslau, etc. -- many small towns and
> villages would only have been given German names during the Third Reich.
>
> It is therefore contentious to use the "name:de" tag for these places,
> unless one is making a map of occupied Poland during WW2. The naming was a
> political act, and most of the names were not used by Germans, even those
> living in the vicinity, before 1939 or after 1945. Taking politics out of
> it, perhaps one could use the date to indicate when the name was in use,
> thus a key of "name:de(1939-1945)".
>
> It would be good to speak to historians who specialize in this area.
>
> - L
>
> On 19 Mar 2014, at 20:37, Chris Helenius <chris.helenius(a)gmail.com> wrote:
>
> How are historical place names from annexed countries regarded? Or put in
> another way; when does a name no longer exist?
>
> In the case on Finland, which lost Karelia to Russia in the 1950s,
> hundreds of place names were translated and are now officially Russian,
> with the Finnish population gone.
> Former place names could nevertheless be of historical value (e.g. to see
> the geographical extent of the language), as physical historical features
> are.
>
> The question is, does a name disappear when it is no longer used? Larger
> cities are still called by their Finnish names in a Finnish context, so
> would towns and villages be any different? Or when they are deserted?
>
> There is also the unignorable issue of geopolitics, as there are still
> tensions between the countries.
> There is no shortage of geographical naming disputes (
> https://en.wikipedia.org/wiki/Category:Geographical_naming_disputes),
> and wikipedians themselves had a row over geographical names. (
> http://www.foreignpolicy.com/articles/2013/02/05/China_Japan_Wikipedia_War_…
> )
> I can imagine how the naming could be seen having a political agenda.
>
> For what it's worth, my agenda is only historical, although I can't shrug
> off my national bias.
> Before I go and add name:fi= place-names, I'd like to hear what the
> community thinks of this.
>
> Chris Helenius
> _______________________________________________
> Historic mailing list
> Historic(a)openstreetmap.org
> https://lists.openstreetmap.org/listinfo/historic
>
>
>
> _______________________________________________
> Historic mailing list
> Historic(a)openstreetmap.org
> https://lists.openstreetmap.org/listinfo/historic
>
>
> Am 20/mar/2014 um 07:58 schrieb Susanna Ånäs <susanna.anas(a)gmail.com>:
>
> Do the notability guidelines of Wikimedia allow storing only important places?
because the consequence of using wikidata will be to have wikidata objects not only for places but also for minor streets and squares as soon as they change name (most of these will not have Wikipedia articles)
cheers,
Martin
Organize MediaWiki's catalog of 1000s of extensions using Wikidata. Is this
a sensible idea? Reality checks and other opinions are welcome here or at
https://bugzilla.wikimedia.org/show_bug.cgi?id=46704#c33
Pasting the relevant part for convenience:
Has anybody discussed the possibility of creating Wikidata items for
extensions, after defining a set of properties to describe them? Linking
those Wikidata items to mediawiki.org extension pages, and then playing
with templates and what not to keep the semantic data up to date (version
number, last release, dependencies, compatible with MediaWiki releases...)?
Then play with templates, queries and visualizations to create all kinds of
useful output, from structured extension pages to a proper and robust map
of extensions.
--
Quim Gil
Engineering Community Manager @ Wikimedia Foundation
http://www.mediawiki.org/wiki/User:Qgil
(please forward)
Dear all,
we are very happy to announce that we have succeeded in the formation of
a new organization to support DBpedia and its community.
The DBpedia Association is now officially in action. In the coming
months, we hope to raise funding to reach some of the goals outlined in
our charter: http://dbpedia.org/association
At the European Data Forum in Athens today and tomorrow, you are able to
meet many of the people from the DBpedia Community who have helped to
create the DBpedia Association for example Martin Kaltenböck, Michael
Martin and Dimitris Kontokostas will be at the the LOD2 Booth at EDF
and Asunción Gómez Pérez will be at the LD4LT side event on Friday
(Just to name a few)
From September 1st-5th in Leipzig, we hope to gather everyone to
celebrate this great achievement. Especially on the 3rd of September,
where we will have the 2nd DBpedia Community meeting, which is
co-located with the SEMANTiCS 2014 (formerly i-SEMANTICS) on September 4-5.
The people who have all worked together to create this wealth of value
under the DBpedia name are so numerous that we are hardly able to know
their exact number or all their names. For proper acknowledgement, as a
first action the DBpedia Association will start to give out Linked Data
URIs during the next months for all its contributors and supporters.
Personally, I am very proud to live in such a great age of collaboration
where we are able to work together across borders and institutions.
Hope to see you in person in September or earlier as linked data under
the http://dbpedia.org/community/$contributor namespace.
Sebastian Hellmann
--
Sebastian Hellmann
AKSW/NLP2RDF research group
Insitute for Applied Informatics (InfAI) affiliated with DBpedia
Events:
* *21st March, 2014*: LD4LT Kick-Off
<https://www.w3.org/community/ld4lt/wiki/LD4LT_Group_Kick-Off_and_Roadmap_Me…>
@European Data Forum
* *Sept. 1-5, 2014* Conference Week in Leipzig, including
** *Sept 2nd*, MLODE 2014
** *Sept 3rd*, 2nd DBpedia Community Meeting
** *Sept 4th-5th*, SEMANTiCS (formerly i-SEMANTICS) <http://semantics.cc/>
Venha para a Alemanha como PhD: http://bis.informatik.uni-leipzig.de/csf
Projects: http://dbpedia.org, http://nlp2rdf.org,
http://linguistics.okfn.org, https://www.w3.org/community/ld4lt
<http://www.w3.org/community/ld4lt>
Homepage: http://aksw.org/SebastianHellmann
Research Group: http://aksw.org
Thesis:
http://tinyurl.com/sh-thesis-summaryhttp://tinyurl.com/sh-thesis
Hoi,
Next month I will be presenting a half a day meeting in the Netherlands for
a big GLAM partner. This presentation will likely be repeated for other
GLAMs in the Netherlands at a later date.
I have a tentative program for half a day. I will present about Wikidata, I
will show information that already exists for both artists, works of art
and institutions. I will demonstrate Wikidata, Reasonator, WDQ.
One of the things that we will cover is how to add information from their
databases to Wikidata. They are very much into movies, actors, television
etc.
My question to you. What would you like to see covered in such a workshop /
presentation. What emphasis would work for you best.
Thanks,
GerardM
Hey folks :)
There is a cool hackathon around open data and culture coming up. I'm
sure there are a few possible awesome Wikidata-related projects some
of you could take on. Details (unfortunately only in German) are here:
http://daten.berlin.de/interaktion/artikel/coding-da-vinci-der-kultur-hacka…
Cheers
Lydia
--
Lydia Pintscher - http://about.me/lydia.pintscher
Product Manager for Wikidata
Wikimedia Deutschland e.V.
Tempelhofer Ufer 23-24
10963 Berlin
www.wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e. V.
Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg
unter der Nummer 23855 Nz. Als gemeinnützig anerkannt durch das
Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.