KMi News

KMi Researchers Win the Best Poster Award at JCDL 2016

KMi Researchers Win the Best Poster Award at JCDL 2016

Drahomira Herrmannova and Petr Knoth have won the Best Poster Award at JCDL 2016 in Newark, USA with their contribution “Semantometrics: Towards fulltext-based research evaluation.” This was a very good timing as the full experimental report on semantometrics commissioned by Jisc was published in this announcement a week prior the conference. 

The CORE team at KMi have also organised a successful 5th International Workshop on Mining Scientific Publications (WOSP 2016). The workshop was attended by key people in the area of text and data mining research papers from both Europe and the USA. The workshop featured this year two keynotes. Yuxiao Dong of Notre Dame University gave a talk titled “AMiner: Towards Understanding Big Scholarly Data” and Michael J. Kurtz of Harvard-Smithsonian Centre for Astrophysics presented the "Astrophysics Data System: The Joy of Text”.

At the workshop, Drahomira Herrmannova also presented a joint long paper with Petr Knoth titled: "An Analysis of the Microsoft Academic Graph."

The WOSP workshop was this year sponsored by the OpenMinTeD project in which KMi participates and we invited two speakers on this.  Stelios Piperidis of Athena Research Centre gave a talk on “Making sense of scientific textual content” and Peter Mutschke of GESIS presented a discussed in his talk the "Challenges and potential of text mining in scholarly information retrieval.” 

Webscience Summer School 2016

Webscience Summer School 2016

The WSTNet Web Science Summer School 2016 was held from Thursday, 30 June to Wednesday, 6 July 2016. The Summer School is the first edition of the oncoming WSTNet Summer School series on Web Science. It is organised by the Institute for Web Science and Technologies (WeST) at the University of Koblenz-Landau, Germany. Keynotes, tutorials and project work in the following areas of Web Science: Introduction to Web Science, Computational Social Science, Social Machines, Politics, Entrepreneurship, and Law. The summer school featured keynote lectures such as Leveraging Web/Internet/Network Sciences (WINS) to address Grand Societal Challenges by Prof. Noshir Contractor and Social Machines 2.0:  The Coming Collision of Artificial Intelligence, Social Networking, and Humanity by Prof. James Alexander Hendler. 
 
Apart from the lectures and tutorials each participant presented their research in a poster session where I shared my PhD research topic that I am currently working on in KMi under the supervision of Dr. Miriam Fernandez and Prof. Harith Alani. My research is in Natural Language Processing (NLP); I am focusing on extracting sentiment from dialectal social data "specifically for Arabic”. We have done some studies on the language usage on Twitter within Arab regions and the identification of Romanized Arabic from multilingual data streams. Next, we will investigate dialect identification and sentiment analysis. 
 
The organisers of the summer school also distributed us into teams of 4-5 students and assigned projects for us to work on and present by the end of the summer school. Our project title was “Detecting Hate Speech from Reddit Data” supervised by Dr. Claudia Wagner and Dr. Fabian Flöck. We implemented an unsupervised approach that relies on a hate database to extract hate scores from reddit comments.
 
I am a first-year PhD student and this was my first summer school, in short, I am very delighted by the rich experience. Besides the academic part, the Webscience Summer School has a great value by connecting students and professors from different disciplines and different parts of the world. I have to thank my supervisors for giving me this opportunity and Prof. Steffan Staab and Dr. Ulrich Wechselberger for a great Webscience Summer School. 

Related Links:


European Data Forum a Great Success

European Data Forum a Great Success

The 2016 European Data Forum (EDF) which took place last week was a great success. This event which was setup by the European Commission supported by STI International and a number of prominent European researchers in 2012 has grown from under 200 participants to over 1000 registrations at this year’s event. EDF brings together key policy makers, industrial players and researchers to discuss issues around the area of Big Data.
                   
The event was opened by the CEO of Philips, Frans van Houten and included a video message from Gunther Oettinger the European Commissioner for Digital Economy and Society. The CEO and co-founder of TomTom, Harold Goddijn, explained his vision for self-driving cars. By 2020 the first commercial semi-autonomous cars will be available. Within 50 years self-driving taxis will transport city dwellers arriving within two minutes of a button press. He speculated that this mode of transport will include specific personal services such as built-in hair salons. Anders Arpteg, the Head of Analytics at Spotify, articulated how his data-first company uses data analytics to provide music recommendations, target advertising and help patients with Alzheimer’s.
 
Other presentations included Stef Oud from Deloitte Consulting who outlined how they are using Big Data to support arriving refugees in finding jobs and Peter Scheer from the Amsterdam Arena (home to the Ajax football team) who showed how soil sensors and data analytics are used to support grass maintenance by monitoring ambient moisture and tracking the individual paths taken by players. Wesley Goatley, a sound artist currently working on an MK:Smart art installation for the forthcoming Milton Keynes International Festival, expounded the relationship between art and Big Data and explained the concepts underlying some of his creations.
 
KMi was heavily involved in the organisation of EDF. John Domingue was a Conference Chair, Enrico Motta articulated the MK:Smart vision for smart urban environments, Zdenek Zdrahal talked about his team’s work on OU Analyse, the European Data Science Academy a KMi-coordinated project involving John Domingue, Alex Mikroyonnidis, Aba-Sah Dadzie and Aneta Tumilowicz was presented, and Damian Dadswell, Harriett Cornish, Paul Hogan and Jamie Daniels designed and developed the conference website and accompanying smart phone App.
 
EDF 2016 was an extremely stimulating event showcasing the very best in European Big Data research and innovation demonstrating the tremendous value that European collaboration can bring. Long may it continue!!   
 

Related Links:


KMi News Image 0
John Domingue introducing Zdenek Zdrahal for his OU Analyse Talk
KMi News Image 1
Enrico Motta presenting MKSmart
KMi News Image 2
John introducing Elena Simperl for her talk on the European Data Science Academy

CORE wins Best Poster Award at the Open Repositories Conference #OR2016

CORE wins Best Poster Award at the Open Repositories Conference #OR2016