The document provides an overview of the Multimedia Lab at Ghent University and iMinds research institute in Belgium. It discusses the organizational structure of Ghent University and iMinds and describes the research activities of the Multimedia Lab, including social media analysis, visual content understanding, and deep machine learning. It also outlines some specific projects on Twitter data involving hashtag recommendation, named entity recognition, and social television.
Presentation on "Practical Competences in Engineering and Technology Enhanced Learning: MOOCs and Emerging Areas at the IEEE Education Society" from the IEEE Education Society Special Technical Community on Learning Sciences at the The Chinese University of Hong Kong
Cloud-based Digital Technologies for Opening Up Education: Keep On Learning ...Demetrios G. Sampson
Demetrios G Sampson, “Cloud-based Digital Technologies for Opening Up Education: Keep Up Learning beyond the Physical Classroom at the Digital Cloud”, 1st International Summit on Education in the Cloud, Athens, Greece, 22 March 2014.
Scaling up digital learning support for smart workforce development in cluste...Ralf Klamma
4th Research Forum on Small and Medium Sized Enterprises, Chur, Switzerland, February 9-10, 2015
Ralf Klamma & Tobias Ley
RWTH Aachen University, Germany & Tallinn University, Estonia
klamma@dbis.rwth-aachen.de & tley@tlu.ee
Keynote presentation at IEEE TALE 2013 conference - A Second Step Ahead in the Future of Labs and Learning: MOOCs, Widgets, Ubiquity and Mobility - Bali, Indonesia, August 2013 http://www.tale-conference.org/tale2013/
Ralf Klamma
Advanced Community Information Systems (ACIS)RWTH Aachen University, Germany
klamma@dbis.rwth-aachen.de
Dresden, January 22, 2015
las2peer is a distributed, highly reliable and secure platform for creating community information systems and community services.
The main goal of las2peer is to provide a fast and flexible way to create services which may communicate with each other and their users through standard protocols. The used and stored information is handled in a trustworthy way and within full control of the communities.
Presentation on "Practical Competences in Engineering and Technology Enhanced Learning: MOOCs and Emerging Areas at the IEEE Education Society" from the IEEE Education Society Special Technical Community on Learning Sciences at the The Chinese University of Hong Kong
Cloud-based Digital Technologies for Opening Up Education: Keep On Learning ...Demetrios G. Sampson
Demetrios G Sampson, “Cloud-based Digital Technologies for Opening Up Education: Keep Up Learning beyond the Physical Classroom at the Digital Cloud”, 1st International Summit on Education in the Cloud, Athens, Greece, 22 March 2014.
Scaling up digital learning support for smart workforce development in cluste...Ralf Klamma
4th Research Forum on Small and Medium Sized Enterprises, Chur, Switzerland, February 9-10, 2015
Ralf Klamma & Tobias Ley
RWTH Aachen University, Germany & Tallinn University, Estonia
klamma@dbis.rwth-aachen.de & tley@tlu.ee
Keynote presentation at IEEE TALE 2013 conference - A Second Step Ahead in the Future of Labs and Learning: MOOCs, Widgets, Ubiquity and Mobility - Bali, Indonesia, August 2013 http://www.tale-conference.org/tale2013/
Ralf Klamma
Advanced Community Information Systems (ACIS)RWTH Aachen University, Germany
klamma@dbis.rwth-aachen.de
Dresden, January 22, 2015
las2peer is a distributed, highly reliable and secure platform for creating community information systems and community services.
The main goal of las2peer is to provide a fast and flexible way to create services which may communicate with each other and their users through standard protocols. The used and stored information is handled in a trustworthy way and within full control of the communities.
Digital Technologies for Supporting Educational Innovations in K-12 Demetrios G. Sampson
Demetrios G Sampson, Digital Technologies for Supporting Educational Innovations in K-12, Invited Seminar, ICT in Language Learning, Athens, Greece - 24 June 2014.
Supporting Professional Communities in the Next Web Ralf Klamma
Keynote
PWM Wissenstag Social Enterprise @ I-KNOW 2013
Wednesday, September 4, 2013 in Graz (Austria)
Ralf Klamma
Advanced Community Information Systems (ACIS)
RWTH Aachen
MOVING presentation at the Course in Open Education Design, July 2018, SloveniaMOVING Project
The aim of the course was to equip the participants with basic knowledge, practical advice and hands-on experience to prepare them for their own design of Open Educational Resources (OER).
Department of Information and Communication Technologies. Univ. Pompeu Fabra,...Aurelio Ruiz Garcia
Slides introducing the Department of Information and Communication Technologies at Universitat Pompeu Fabra (DTIC-UPF) in Barcelona.
The activity of DTIC-UPF lies in the broad range of fields created around the convergence of ICT with biomedical and cognitive sciences, computation and networks.
The DTIC-UPF has since its beginnings emphasized scientific excellence and internationalization as core aspects of its activities.
Its community is composed by over 300 people from 58 nationalities. DTIC-UPF is the leading university department in Spain in participating in EU programmes (65 projects in FP7, including 9 prestigious ERC grants).
DTIC delivers both undergraduate and master training in its areas of interest of research (including brain and cognition, cognitive systems and interactive media, intelligent interactive sytems, music and sound computing, computer vision and an upcoming program in wireless communications) and professional activity (digital arts, animation, videogame creation, ICT strategic management, data visualisation, etc),
DTIC-UPF holds agreements with both leading international companies (including Yamaha, General Electric, Siemens, Philips, Telefónica, etc) and a rich ecosystem of local and international SMEs.
DTIC-UPF is also very active in its engagement with society in general, and students in particular, leading and participating regular activities in artistic centres, museums, schools and NGOs.
For more information please visit http://www.upf.edu/dtic/en/
Demetrios G Sampson, Digital Technologies for Opening Up Education, European Network of Educational Councils, Seminar on "Learning in the Digital Age", Athens, Greece, 5-6 May 2014
Digital Technologies for Supporting Educational Innovations in K-12 Demetrios G. Sampson
Demetrios G Sampson, Digital Technologies for Supporting Educational Innovations in K-12, Invited Seminar, ICT in Language Learning, Athens, Greece - 24 June 2014.
Supporting Professional Communities in the Next Web Ralf Klamma
Keynote
PWM Wissenstag Social Enterprise @ I-KNOW 2013
Wednesday, September 4, 2013 in Graz (Austria)
Ralf Klamma
Advanced Community Information Systems (ACIS)
RWTH Aachen
MOVING presentation at the Course in Open Education Design, July 2018, SloveniaMOVING Project
The aim of the course was to equip the participants with basic knowledge, practical advice and hands-on experience to prepare them for their own design of Open Educational Resources (OER).
Department of Information and Communication Technologies. Univ. Pompeu Fabra,...Aurelio Ruiz Garcia
Slides introducing the Department of Information and Communication Technologies at Universitat Pompeu Fabra (DTIC-UPF) in Barcelona.
The activity of DTIC-UPF lies in the broad range of fields created around the convergence of ICT with biomedical and cognitive sciences, computation and networks.
The DTIC-UPF has since its beginnings emphasized scientific excellence and internationalization as core aspects of its activities.
Its community is composed by over 300 people from 58 nationalities. DTIC-UPF is the leading university department in Spain in participating in EU programmes (65 projects in FP7, including 9 prestigious ERC grants).
DTIC delivers both undergraduate and master training in its areas of interest of research (including brain and cognition, cognitive systems and interactive media, intelligent interactive sytems, music and sound computing, computer vision and an upcoming program in wireless communications) and professional activity (digital arts, animation, videogame creation, ICT strategic management, data visualisation, etc),
DTIC-UPF holds agreements with both leading international companies (including Yamaha, General Electric, Siemens, Philips, Telefónica, etc) and a rich ecosystem of local and international SMEs.
DTIC-UPF is also very active in its engagement with society in general, and students in particular, leading and participating regular activities in artistic centres, museums, schools and NGOs.
For more information please visit http://www.upf.edu/dtic/en/
Demetrios G Sampson, Digital Technologies for Opening Up Education, European Network of Educational Councils, Seminar on "Learning in the Digital Age", Athens, Greece, 5-6 May 2014
Authors/Presenters: Vasileios Mezaris and Benoit Huet.
Video hyperlinking is the introduction of links that originate from pieces of video material and point to other relevant content, be it video or any other form of digital content. The tutorial presents the state of the art in video hyperlinking approaches and in relevant enabling technologies, such as video analysis and multimedia indexing and retrieval. Several alternative strategies, based on text, visual and/or audio information are introduced, evaluated and discussed, providing the audience with details on what works and what doesn’t on real broadcast material.
Learning Analytics and Sensemaking in Digital Learning Ecosystems - Examples ...tobold
Presentation given at the Seminar "Opportunities and Challenges of Learning with Technologies: Evidence-based Education" at the Permanent Representation of Estonia to the EU on 12 November 2014 in Brussels.
Open Source for Higher Conventional and Open Education in IndiaRamesh C. Sharma
In this presentation we discussed about what are open source softwares and how higher, conventional and open education system in India is making use of open source tools.
The Wellcome Trust is examining the possibility of a cloud platform for the storage and delivery of digitised artefacts. This platform is intended for the Trust's own use as well as others. A version of this presentation with embedded notes and video can be viewed on Google docs: http://bit.ly/1GRKqN4 or PowerPoint online: http://bit.ly/1CwGsrE
Visual Information Analysis for Crisis and Natural Disasters Management and R...Yiannis Kompatsiaris
Invited talk at the Ninth International Conference on Image Processing Theory, Tools and Applications IPTA 2019 (http://www.ipta-conference.com/ipta19/)
Crises and natural disasters are unwelcome, but also unavoidable features of modern society, affecting more communities than ever. Visual information analysis plays an important role in efficient pre-event (e.g. risk modeling), during the event (response) and post-event (recovery) emergency situation management. This talk will describe the potential role of visual information sources including satellite images, surveillance and traffic cameras, social multimedia and aerial video in applications such as floods, fires, and oil spills. Multimodal and fusion techniques will be presented combining satellite and social data and how deep neural networks can be applied in this domain. The talks will include demos and results from the relevant BeAware and EOPEN projects and from our participation in the 2018 Multimedia Satellite Task of the MediaEval Benchmarking Initiative.
European Open Science Cloud: Concept, status and opportunitiesEOSC-hub project
European Open Science Cloud: Concept, status and opportunities.
Presentation given by Gergely Sipos at the International Symposium on Grids and Clouds 2019 event in Taiwan.
Data Science: History repeated? – The heritage of the Free and Open Source GI...Peter Löwe
Data Science is described as the process of knowledge extraction from large data sets by means of scientific
methods. The discipline draws heavily from techniques and theories from many fields, which are jointly used to
furthermore develop information retrieval on structured or unstructured very large datasets. While the term Data
Science was already coined in 1960, the current perception of this field places is still in the first section of the hype cycle according to Gartner, being well en route from the technology trigger stage to the peak of inflated
expectations.
In our view the future development of Data Science could benefit from the analysis of experiences from
related evolutionary processes. One predecessor is the area of Geographic Information Systems (GIS). The
intrinsic scope of GIS is the integration and storage of spatial information from often heterogeneous sources, data
analysis, sharing of reconstructed or aggregated results in visual form or via data transfer. GIS is successfully
applied to process and analyse spatially referenced content in a wide and still expanding range of science
areas, spanning from human and social sciences like archeology, politics and architecture to environmental and
geoscientific applications, even including planetology.
This paper presents proven patterns for innovation and organisation derived from the evolution of GIS,
which can be ported to Data Science. Within the GIS landscape, three strategic interacting tiers can be denoted: i) Standardisation, ii) applications based on closed-source software, without the option of access to and analysis of the implemented algorithms, and iii) Free and Open Source Software (FOSS) based on freely accessible program code enabling analysis, education and ,improvement by everyone. This paper focuses on patterns gained from the synthesis of three decades of FOSS development. We identified best-practices which evolved from long term FOSS projects, describe the role of community-driven global umbrella organisations such as OSGeo, as well as the standardization of innovative services. The main driver is the acknowledgement of a meritocratic attitude.
These patterns follow evolutionary processes of establishing and maintaining a web-based democratic culture
spawning new kinds of communication and projects. This culture transcends the established compartmentation and
stratification of science by creating mutual benefits for the participants, irrespective of their respective research
interest and standing. Adopting these best practices will enable
Slides for presentation given at the first Digital Humanities Congress held in Sheffield from 6 – 8 September 2012 with the support of the Network of Expert Centres and Centernet.
URL http://www.shef.ac.uk/hri/dhc2012
Research and Development at Sound and Vision Victor de Boer
Slides for guest lecture about R&D at the Netherlands Institute for Sound and Vision for the lecture series "Introduction to IMM" at VU Amsterdam.
With slides by Lotte Belice Baltussen, Maarten Brinkerink, Johan Oomen, Bouke Huurnink and Victor de Boer
Into the Night - Technology for citizen scienceMuki Haklay
Current citizen science seems effortless...just download an app and start using it. However, there are many technical aspects that are necessary to make a citizen science project work. In this session, we will provide an overview of all the technical elements that are required - from the process of designing an app., to designing and managing a back-end system, to testing the system end to end before deployment. Participants will have the opportunity to engage in a short exercise to consider the design of an app for a citizen science project that addresses light pollution.
Participation of ADITESS LTD in Career and Entrepreneurship Fair in CYPRUS UNIVERSITY OF TECHNOLOGY where we will present our company, "The case of a Cyprus start up investing in information systems and empowering staff"
Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...Wesley De Neve
Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional neural networks. Paper presented at the Workshop on Computational Biology at the International Conference on Machine Learning, Long Beach, USA, 2019.
Investigating the biological relevance in trained embedding representations o...Wesley De Neve
Investigating the biological relevance in trained embedding representations of protein sequences. Paper presented at the Workshop on Computational Biology at the International Conference on Machine Learning, Long Beach, USA, 2019.
Towards reading genomic data using deep learning-driven NLP techniquesWesley De Neve
Towards reading genomic data using deep learning-driven NLP techniques. Slides presented at BIOINFO 2016 – Precision Bioinformatics & Machine Learning.
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...Wesley De Neve
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to Smart Farming. Presentation given at the Korea-Europe International Conference on the 4th Industry Revolution.
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...Wesley De Neve
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Target Interaction and DNA Analysis.
Poster presented at the BIG N2N Symposium 2016.
Towards using multimedia technology for biological data processingWesley De Neve
Towards using multimedia technology for biological data processing.
Presentation given during the Ghent University Global Campus (GUGC) Research Seminar on 19/1/2014.
PHP Frameworks: I want to break free (IPC Berlin 2024)Ralf Eggert
In this presentation, we examine the challenges and limitations of relying too heavily on PHP frameworks in web development. We discuss the history of PHP and its frameworks to understand how this dependence has evolved. The focus will be on providing concrete tips and strategies to reduce reliance on these frameworks, based on real-world examples and practical considerations. The goal is to equip developers with the skills and knowledge to create more flexible and future-proof web applications. We'll explore the importance of maintaining autonomy in a rapidly changing tech landscape and how to make informed decisions in PHP development.
This talk is aimed at encouraging a more independent approach to using PHP frameworks, moving towards a more flexible and future-proof approach to PHP development.
State of ICS and IoT Cyber Threat Landscape Report 2024 previewPrayukth K V
The IoT and OT threat landscape report has been prepared by the Threat Research Team at Sectrio using data from Sectrio, cyber threat intelligence farming facilities spread across over 85 cities around the world. In addition, Sectrio also runs AI-based advanced threat and payload engagement facilities that serve as sinks to attract and engage sophisticated threat actors, and newer malware including new variants and latent threats that are at an earlier stage of development.
The latest edition of the OT/ICS and IoT security Threat Landscape Report 2024 also covers:
State of global ICS asset and network exposure
Sectoral targets and attacks as well as the cost of ransom
Global APT activity, AI usage, actor and tactic profiles, and implications
Rise in volumes of AI-powered cyberattacks
Major cyber events in 2024
Malware and malicious payload trends
Cyberattack types and targets
Vulnerability exploit attempts on CVEs
Attacks on counties – USA
Expansion of bot farms – how, where, and why
In-depth analysis of the cyber threat landscape across North America, South America, Europe, APAC, and the Middle East
Why are attacks on smart factories rising?
Cyber risk predictions
Axis of attacks – Europe
Systemic attacks in the Middle East
Download the full report from here:
https://sectrio.com/resources/ot-threat-landscape-reports/sectrio-releases-ot-ics-and-iot-security-threat-landscape-report-2024/
"Impact of front-end architecture on development cost", Viktor TurskyiFwdays
I have heard many times that architecture is not important for the front-end. Also, many times I have seen how developers implement features on the front-end just following the standard rules for a framework and think that this is enough to successfully launch the project, and then the project fails. How to prevent this and what approach to choose? I have launched dozens of complex projects and during the talk we will analyze which approaches have worked for me and which have not.
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualityInflectra
In this insightful webinar, Inflectra explores how artificial intelligence (AI) is transforming software development and testing. Discover how AI-powered tools are revolutionizing every stage of the software development lifecycle (SDLC), from design and prototyping to testing, deployment, and monitoring.
Learn about:
• The Future of Testing: How AI is shifting testing towards verification, analysis, and higher-level skills, while reducing repetitive tasks.
• Test Automation: How AI-powered test case generation, optimization, and self-healing tests are making testing more efficient and effective.
• Visual Testing: Explore the emerging capabilities of AI in visual testing and how it's set to revolutionize UI verification.
• Inflectra's AI Solutions: See demonstrations of Inflectra's cutting-edge AI tools like the ChatGPT plugin and Azure Open AI platform, designed to streamline your testing process.
Whether you're a developer, tester, or QA professional, this webinar will give you valuable insights into how AI is shaping the future of software delivery.
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...James Anderson
Effective Application Security in Software Delivery lifecycle using Deployment Firewall and DBOM
The modern software delivery process (or the CI/CD process) includes many tools, distributed teams, open-source code, and cloud platforms. Constant focus on speed to release software to market, along with the traditional slow and manual security checks has caused gaps in continuous security as an important piece in the software supply chain. Today organizations feel more susceptible to external and internal cyber threats due to the vast attack surface in their applications supply chain and the lack of end-to-end governance and risk management.
The software team must secure its software delivery process to avoid vulnerability and security breaches. This needs to be achieved with existing tool chains and without extensive rework of the delivery processes. This talk will present strategies and techniques for providing visibility into the true risk of the existing vulnerabilities, preventing the introduction of security issues in the software, resolving vulnerabilities in production environments quickly, and capturing the deployment bill of materials (DBOM).
Speakers:
Bob Boule
Robert Boule is a technology enthusiast with PASSION for technology and making things work along with a knack for helping others understand how things work. He comes with around 20 years of solution engineering experience in application security, software continuous delivery, and SaaS platforms. He is known for his dynamic presentations in CI/CD and application security integrated in software delivery lifecycle.
Gopinath Rebala
Gopinath Rebala is the CTO of OpsMx, where he has overall responsibility for the machine learning and data processing architectures for Secure Software Delivery. Gopi also has a strong connection with our customers, leading design and architecture for strategic implementations. Gopi is a frequent speaker and well-known leader in continuous delivery and integrating security into software delivery.
GraphRAG is All You need? LLM & Knowledge GraphGuy Korland
Guy Korland, CEO and Co-founder of FalkorDB, will review two articles on the integration of language models with knowledge graphs.
1. Unifying Large Language Models and Knowledge Graphs: A Roadmap.
https://arxiv.org/abs/2306.08302
2. Microsoft Research's GraphRAG paper and a review paper on various uses of knowledge graphs:
https://www.microsoft.com/en-us/research/blog/graphrag-unlocking-llm-discovery-on-narrative-private-data/
JMeter webinar - integration with InfluxDB and GrafanaRTTS
Watch this recorded webinar about real-time monitoring of application performance. See how to integrate Apache JMeter, the open-source leader in performance testing, with InfluxDB, the open-source time-series database, and Grafana, the open-source analytics and visualization application.
In this webinar, we will review the benefits of leveraging InfluxDB and Grafana when executing load tests and demonstrate how these tools are used to visualize performance metrics.
Length: 30 minutes
Session Overview
-------------------------------------------
During this webinar, we will cover the following topics while demonstrating the integrations of JMeter, InfluxDB and Grafana:
- What out-of-the-box solutions are available for real-time monitoring JMeter tests?
- What are the benefits of integrating InfluxDB and Grafana into the load testing stack?
- Which features are provided by Grafana?
- Demonstration of InfluxDB and Grafana using a practice web application
To view the webinar recording, go to:
https://www.rttsweb.com/jmeter-integration-webinar
Kubernetes & AI - Beauty and the Beast !?! @KCD Istanbul 2024Tobias Schneck
As AI technology is pushing into IT I was wondering myself, as an “infrastructure container kubernetes guy”, how get this fancy AI technology get managed from an infrastructure operational view? Is it possible to apply our lovely cloud native principals as well? What benefit’s both technologies could bring to each other?
Let me take this questions and provide you a short journey through existing deployment models and use cases for AI software. On practical examples, we discuss what cloud/on-premise strategy we may need for applying it to our own infrastructure to get it to work from an enterprise perspective. I want to give an overview about infrastructure requirements and technologies, what could be beneficial or limiting your AI use cases in an enterprise environment. An interactive Demo will give you some insides, what approaches I got already working for real.
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...Ramesh Iyer
In today's fast-changing business world, Companies that adapt and embrace new ideas often need help to keep up with the competition. However, fostering a culture of innovation takes much work. It takes vision, leadership and willingness to take risks in the right proportion. Sachin Dev Duggal, co-founder of Builder.ai, has perfected the art of this balance, creating a company culture where creativity and growth are nurtured at each stage.
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf91mobiles
91mobiles recently conducted a Smart TV Buyer Insights Survey in which we asked over 3,000 respondents about the TV they own, aspects they look at on a new TV, and their TV buying preferences.
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Jeffrey Haguewood
Sidekick Solutions uses Bonterra Impact Management (fka Social Solutions Apricot) and automation solutions to integrate data for business workflows.
We believe integration and automation are essential to user experience and the promise of efficient work through technology. Automation is the critical ingredient to realizing that full vision. We develop integration products and services for Bonterra Case Management software to support the deployment of automations for a variety of use cases.
This video focuses on the notifications, alerts, and approval requests using Slack for Bonterra Impact Management. The solutions covered in this webinar can also be deployed for Microsoft Teams.
Interested in deploying notification automations for Bonterra Impact Management? Contact us at sales@sidekicksolutionsllc.com to discuss next steps.
AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outline Research Activities
1. ELIS – Multimedia Lab
Multimedia Lab @ Ghent University - iMinds:
Organizational Overview & Outline Research Activities
Research Seminar
KAIST, 1 August 2014
Wesley De Neve
@wmdeneve
Ghent University – iMinds & KAIST
2. 2
ELIS – Multimedia Lab
Outline
• Organizational overview (15 minutes)
- Ghent University
- iMinds
- Multimedia Lab
• Outline research activities (45 minutes)
- social media analysis
- visual content understanding
- deep machine learning
3. 3
ELIS – Multimedia Lab
Outline
• Organizational overview (15 minutes)
- Ghent University
- iMinds
- Multimedia Lab
• Outline research activities (45 minutes)
- social media analysis
- visual content understanding
- deep machine learning
4. 4
ELIS – Multimedia Lab
Ghent University (1/3)
• A Dutch-speaking public university
- located in Ghent, Belgium
- established in 1817
Ghent
Brussels
5. 5
ELIS – Multimedia Lab
Ghent University (2/3)
• Consists of 38,000 students and 8,000 staff members
- about 4,000 foreign students and 800 foreign staff members
• Consists of eleven faculties, composed of more than 130 departments
- campus buildings distributed all over the city
Congress Center
‘Het Pand’
Faculty of Engineering
and Architecture
Aula Academia
6. 6
ELIS – Multimedia Lab
Ghent University (3/3)
• Ghent University Global Campus in Songdo
- offers academic programs in molecular biotechnology, environmental
technology, and food technology
- operates together with the State University of New York (SUNY), George
Mason University, and University of Utah
Songdo Global University Campus Visit to Samsung Biologics
7. 7
ELIS – Multimedia Lab
• Organizational overview
- Ghent University
- iMinds
- Multimedia Lab
Outline
• Outline research activities
- social media analysis
- visual content understanding
- deep machine learning
8. 8
ELIS – Multimedia Lab
iMinds
Research institute founded in 2004 by the Flemish
government, with the aim of creating lasting
economic and social value through ICT innovation
9. 9
ELIS – Multimedia Lab
iMinds: A Virtual Research Institute
Leverages the strengths of 5 universities,
20 research groups, and more than 850 researchers
10. 10
ELIS – Multimedia Lab
iMinds’ Research Departments
ICT Media Health Energy
Smart
Cities
Manu-facturing
Internet Technologies
Multimedia Technologies
Security
Medical Information Technologies
Digital Society
11. 11
ELIS – Multimedia Lab
From Idea to Business: The iMinds Innovation Toolbox
5+ years Time-to-market …1 year
Strategic research
Incubation &
entrepreneurship
Applied
research
Pre-competitive
testing
Knowledge-driven
Explorative
Basics for applied
research
Training &
coaching
Financing
Facilities
Networking
Internationali-zation
Business-driven
Interdisciplinary
Cooperative
Demand-driven
Proof of Concept
ICON projects
Large-scale user
trials & living
labs
Evaluate
technical
feasibility
Simulations
12. 12
ELIS – Multimedia Lab
iMinds ICON: Example Projects
• iRead+ – The intelligent reading companion
- January 2012 to December 2013
- finished project that built a text analysis
pipeline for enriching digital news articles
in Dutch and French with links to Wikipedia,
dictionary definitions, and images
• GiPA – Generic platform for augmented reality
- January 2014 to December 2015
- aims at building an interoperable platform
for augmented reality applications, ranging
from games to simulations, addressing diverse
requirements, from capturing to rendering
13. 13
ELIS – Multimedia Lab
• Organizational overview
- Ghent University
- iMinds
- Multimedia Lab
Outline
• Outline research activities
- social media analysis
- visual content understanding
- deep machine learning
14. 14
ELIS – Multimedia Lab
People (Speech Lab excluded)
• Staff
- Rik Van de Walle – senior full professor, head of MMLab
- Peter Lambert – associate professor
- Piet Verhoeve – guest lecturer (ICON program manager at iMinds)
- Erik Mannens, Jan De Cock & Wesley De Neve – research management
- Ellen Lammens & Laura Smekens – administrative management
• 35 researchers
- 50% PhD students
• Miscellaneous
- about 15 master’s thesis students per year
- a few Summer internships each year
15. 15
ELIS – Multimedia Lab
Research Activities (1/2)
• Cluster 1: Video Coding (Jan De Cock)
- compression and transport of video
- transcoding and scalable coding
- high-dynamic range video
• Cluster 2: Game Tech & Graphics (Peter Lambert)
- augmented and virtual reality
- texture and mesh compression
- path planning
16. 16
ELIS – Multimedia Lab
Research Activities (2/2)
• Cluster 3: Semantic Web (SWTF; Erik Mannens)
- multimedia and interactivity on the Web
- knowledge representation and reasoning
- (big) data analytics and visualization
- digital publishing
• Cluster 4: Social & Visual Intelligence (SaVI; Wesley De Neve)
- social media analysis
- visual content analysis
- machine learning
17. 17
ELIS – Multimedia Lab
Teaching Activities
• Bachelor/Master Computer Science and Bachelor/Master Electronics
(Faculty of Engineering and Architecture)
- Multimedia Techniques
- Design of Multimedia Applications
- Advanced Multimedia Applications
• Bachelor Informatics
(Faculty of Sciences)
- Multimedia
- Internet Technology
• Bachelor Biotechnology
(Songdo Global Campus)
- Structured Programming
+ New graduate course on
Big Data Analytics
(pending approval)
18. 18
ELIS – Multimedia Lab
Standardization Activities
• W3C (World Wide Web Consortium)
- new Web techniques
- e.g., HTML5 and Media Annotations
• MPEG (Moving Picture Experts Group)
- new compression techniques
• e.g., H.264/AVC and 3-D Video Coding
- new storage and transport techniques
• e.g., MP4 file format and MPEG DASH
• VQEG (Video Quality Experts Group)
- measurement of video quality
- e.g., subjective quality evaluations
19. 19
ELIS – Multimedia Lab
• Organizational overview
- Ghent University
- iMinds
- Multimedia Lab
Outline
• Outline research activities
- social media analysis
- visual content understanding
- deep machine learning
20. 20
ELIS – Multimedia Lab
Twitter
• An online social network service that enables users to send and read
short 140-character text messages, called "tweets" or "microposts"
Hashtag
(starts with #)
Tweet or
Mention
(starts with @)
Favorite
(like or
bookmark)
Retweet micropost
(sharing)
21. 21
ELIS – Multimedia Lab
Famous Tweets
Note the presence of both textual and (embedded) visual information!
22. 22
ELIS – Multimedia Lab
• Usage in general
Twitter Statistics
- 271 million monthly active users
- 500 million Tweets are sent per day
- 78% of active users are on mobile
- expected revenue for 2014 is $1.33 billion
• mobile advertising + data licensing
• Usage during the World Cup 2014
- fans sent 672 million related tweets in total
- during the semi-final between Brazil and Germany, fans sent more
than 35.6 million tweets
- during the final, the number of tweets sent by fans peaked at
618,725 Tweets Per Minute (TPM)
23. 23
ELIS – Multimedia Lab
Twitter Research Goal and Challenges
• Research goal
- to make sense of the vast amounts of textual and visual information
communicated on Twitter by means of machine learning
• Challenges
- microposts are noisy in nature
- microposts are short-form in nature
- microposts are multi-lingual in nature
- microposts come in highly varying quantities
- microposts are real-time in nature
- microposts are multi-modal in nature (textual & visual, a/o)
24. 24
ELIS – Multimedia Lab
• What?
Deep Learning (1/4)
- simply speaking: use of multi-layered neural networks that are able
to learn complicated mappings between inputs and outputs
x y = hθ(x)
learned intermediate features
deep learning = (hierarchical) representation learning
25. 25
ELIS – Multimedia Lab
Deep Learning (2/4)
• Example learned features
Supervised handwritten
digit recognition
Unsupervised visual object recognition
(Google Brain)
26. 26
ELIS – Multimedia Lab
Deep Learning (3/4)
• Why the resurgence of neural networks?
- availability of large data sets (cf. social media & Internet of Things)
- availability of cheap computing power (cf. GPU & cloud)
- availability of algorithmic improvements (cf. DropOut & max pooling)
• Current achievements
- top performance in handwritten digit recognition
- top performance in automatic speech recognition
- top performance in large-scale visual concept detection
• Attracts substantial private R&D investments
- Google (Geoffrey Hinton & Ray Kurzweil), Facebook (Yann LeCun),
Baidu (Andrew Ng & Kai Yu), Microsoft, Twitter, Netflix, and so on
27. 27
ELIS – Multimedia Lab
Deep Learning (4/4)
• Plenty of open research challenges
- how to tailor deep neural networks to novel applications?
- how to scale up deep neural networks?
- how to scale down neural networks at no cost in effectiveness?
- how to take advantage of massively parallel hardware?
- how to develop effective hybrid architectures?
- how to take into account long-term temporal dependencies?
- how to implement multi-modal approaches?
- how to establish solid theoretical foundations?
- how to bridge the gap between deep learning and strong A.I.?
28. 28
ELIS – Multimedia Lab
Ongoing Research Topics with a Twitter Focus
• Hashtag recommendation
• Named entity recognition and disambiguation
• Sports analytics
• Social television
• Vine video classification
29. 29
ELIS – Multimedia Lab
Social and Visual Intelligence (SaVI)
Abhineshwar Tomar
abhineshwar.tomar@ugent.be
Fréderic Godin
frederic.godin@ugent.be
Baptist Vandersmissen
baptist.vandersmissen@ugent.be
Wesley De Neve
wesley.deneve@ugent.be
Azarakhsh Jalalvand
azarakhsh.jalalvand@ugent.be
+ 3 master’s thesis students
30. 30
ELIS – Multimedia Lab
Research Topics with a Twitter Focus
• Hashtag recommendation
• Named entity recognition and disambiguation
• Social television
• Sports analytics
• Vine video classification
31. 31
ELIS – Multimedia Lab
Hashtags on Twitter
Hashtag usage:
- topic-based indexing & search
• #socialnetwork
• #Reddit
- conversational/event clustering
• #www2014
Observation: only about 10% of tweets contain a hashtag
Research challenge: develop techniques for Twitter hashtag recommendation
32. 32
ELIS – Multimedia Lab
Twitter Hashtag Recommendation
Using Deep Learning (1/2)
• Training: learning the relation between tweets and hashtags
Tweet Hashtag
word2vec
300-D
tweet
vector
word2vec
300-D
hashtag
vector
Deep feed-forward
neural
network
300-D input layer
1000-D hidden layer
500-D hidden layer
400-D hidden layer
300-D output layer
Elizabeth Warren Taking on
Hillary as New Democratic
Powerhouse
#politics
33. 33
ELIS – Multimedia Lab
Twitter Hashtag Recommendation
Using Deep Learning (2/2)
• Testing: recommending hashtags to tweets
word2vec
300-D
tweet
vector
300-D
hashtag
vector
Deep feed-forward
neural
network
300-D input layer
1000-D hidden layer
500-D hidden layer
400-D hidden layer
300-D output layer
Tweet
House Democrats suggest
Obama impeachment is
imminent to raise cash
vec2word
Hashtag
Hashtag
Hashtag
Hashtags
#politics
#crisis
34. 34
ELIS – Multimedia Lab
word2vec
• Developed by Google Research
• Computes vector representations for words
- through the use of neural network technology
• trained on part of the Google News dataset (+/- 100 billion words)
• the model contains vectors for 3 million words and phrases
- capture the semantic meaning of a word
• Example word vector properties
- vector('Paris') - vector('France') + vector('Italy') ≈ vector('Rome')
- vector('king') - vector('man') + vector('woman') ≈ vector('queen')
35. 35
ELIS – Multimedia Lab
Experimental Results
Tweet Recommended hashtags
1 Someone dm/text me bc I’m so bored madd, Oh noes, rainnwilson,
sooooooo, fricken
2 The good life is one inspired by love and guided by
knowledge.
Ahh yes, FIVE THINGS About,
YANKEES TALK, Kinder gentler,
Ya gotta love
3 Method of Losing Weight http://t.co/rs64CEuo5W Shape Shifting, Treat Acne, Detect
Cancer, Warps, Calorie Burn
4 I hate today cause its room cleaning day for me!!! FAN ’S ATTIC, Puh leez, Mopping
robot, % #F######## 3v.jsn, Interest
EURO JAP
5 SPELLS AND SPELL-CASTING:ENCYCLOPEDIA OF
5000 SPELLS ( JUDIKA ILLES ):BLACKSMITH’S
WATER HEALING SPELL: A...
http://t.co/k0TfrqJFQW
DEBUTS NEW, NOW AVAILABLE FOR,
TO PUBLISH, DESIGNED TO,
IS READY TO
36. 36
ELIS – Multimedia Lab
Research Topics with a Twitter Focus
• Hashtag recommendation
• Named entity recognition and disambiguation
• Sports analytics
• Social television
• Vine video classification
37. 37
ELIS – Multimedia Lab
Named Entity Recognition and Disambiguation
• Named entity
- person
- location
- organization
- miscellaneous
• film/movie, entertainment award event, political event,
programming language, sporting event and TV show
• Recognition
- identification of a named entity in a given text
• Disambiguation
- e.g., fruit ‘apple’ versus company ‘Apple’
38. 38
ELIS – Multimedia Lab
Research Challenge
• Tools for named entity recognition and disambiguation have thus far
been developed for long-form news articles using formal language
• Need for development of tools for named entity recognition and
disambiguation for short-form microposts using informal language
39. 39
ELIS – Multimedia Lab
Natural Language Processing (NLP) for Twitter from Scratch
Tweet Tokenization
Part-of-Speech
Tagging (PoS)
Chunking
Named Entity
Recognition and
Disambiguation
Information
Retrieval
Text-to-Speech
Artificial Intelligence
(cf. Siri, Cortana, Google Now)
General Text
Parsing
pronoun verb noun
Tom likes Sprite.
40. 40
ELIS – Multimedia Lab
Our Approach: Twitter PoS using Deep Learning
Word 1
Word 2
Word 3
L
o
o
k
u
p
word
vector
word
vector
word
vector
• Use of a feed-forward neural network for learning the mapping between
a collection of word vector representations and a PoS tag
- feature learning and not feature engineering
• Use of word vector representations derived from Twitter
- not from Google News
Neural
network
PoS tag of
word 2
44. 44
ELIS – Multimedia Lab
Research Topics with a Twitter Focus
• Hashtag recommendation
• Named entity recognition
• Sports analytics
• Social television
• Video classification
45. 45
ELIS – Multimedia Lab
• What?
Rationale
- prediction of the outcome of football matches
in the English Premier League (EPL), using both
traditional statistics and Twitter microposts
• Why?
- betting on football is a billion dollar industry
- Twitter is highly popular for real-time coverage of sports events
• How?
- fusion of the output of four simple methods, using different features
and machine learning techniques
46. 46
ELIS – Multimedia Lab
Approach
• Method 1: Statistical features
- ranking in the league, the number of points gathered in the league,
the number of points gathered during the last five games, the
number of goals made, and the number of goals against
• Method 2: Twitter volume changes
• Method 3: Twitter sentiment analysis
• Method 4: Twitter user predictions
• Machine learning
- Naive Bayes, Logistic Regression, and SVM
social features derived from
+50 million tweets
47. 47
ELIS – Multimedia Lab
Experimental Results (1/2)
Method Accuracy
Baseline methods
Naive predictions 51%
Expert predictions 60%
Bookmaker predictions 67%
Individual methods
Statistical features 64%
Twitter volume changes 50%
Twitter sentiment analysis 52%
Twitter user predictions 63%
Combination of statistical features and
Twitter user predictions
Majority voting 64%
Early fusion 68%
Late fusion 66%
49. 49
ELIS – Multimedia Lab
Research Topics with a Twitter Focus
• Hashtag recommendation
• Named entity recognition
• Sports analytics
• Social television
• Video classification
50. 50
ELIS – Multimedia Lab
Rationale (1/2)
• Social television (second screen)
- interaction between televised content and online social networks
• Breaking Bad finale: peak of 22,373 TPM
• Super Bowl 2014: peak of 382,000 TPM
• World Cup 2014 final: peak of 618,725 TPM
51. 51
ELIS – Multimedia Lab
• Challenges
Rationale (2/2)
- how to measure engagement and reach on online social networks?
• cf. the Nielsen television ratings
- how to profile your audience?
• e.g., age, gender and location
• Addressing these challenges is important for the allocation of
advertisement budgets and targeted advertisement strategies
versus
52. 52
ELIS – Multimedia Lab
Measurement of Engagement and Reach in Flanders
• Three major difficulties
- privacy concerns
- low usage of Twitter (at that time)
- identification of Flemish users of Twitter
53. 53
ELIS – Multimedia Lab
Twitter User Profiling: Gender Detection (1/3)
• What?
- classification of Flemish Twitter users into male and female classes
• Why?
- current user profiles do not contain gender information
- gender information is important for targeted advertising
• How?
- through (mostly n-gram) features extracted from the profile of the
user, the tweets of the user, and the social network of the user
- through machine learning based on Naive Bayes and SVM
54. 54
ELIS – Multimedia Lab
Twitter User Profiling: Gender Detection (2/3)
Male
Female
E
n
s
e
m
b
l
e
averaging of
probabilities
Username
Classifier
Name Classifier
Description
Classifier
Tweet Content
Classifier
Tweet Style
Classifier
Friend Description
Classifier
@wmdeneve
Wesley De Neve
Senior Researcher at Ghent University - iMinds &
KAIST. Interested in social media analysis, visual
content understanding and machine learning.
Attending "The Future of Metadata" at CONTEC.
#TISP
URL usage, emoticon usage, and punctuation
Sports fan, basketball player, outdoor lover and
a Ph.D. researcher #SocialTV and Natural
Language Processing (#NLP) @iMinds - @UGent
55. 55
ELIS – Multimedia Lab
Twitter User Profiling: Gender Detection (3/3)
Classifier Accuracy
Username 78.86%
Name 87.54%
Description 65.74%
Tweet content 75.36%
Tweet style 66.34%
Friend description 75.34%
Test set TweetGenie Ensemble
Test set 2 82.15% 91.89%
Test set 3 86.44% 93.32%
56. 56
ELIS – Multimedia Lab
Research Topics with a Twitter Focus
• Hashtag recommendation
• Named entity recognition
• Social television
• Sports analytics
• Vine video classification
57. 57
ELIS – Multimedia Lab
What is Vine? (1/4)
• Platform for social & mobile video
- established in June 2012
• Allows creating & distributing videos of up to 6 seconds
- maximum video length resembles Twitter’s character limitation
• Acquired by Twitter in October 2012
- currently has more than 40 million users
• Has the potential to become a new social news platform
- cf. Ninja News in Belgium
61. 61
ELIS – Multimedia Lab
Automatic Understanding of Social Video Content (1/2)
Recognition of general concepts in video fragments
Categorize short and noisy video fragments
Localize and recognize named entities in video fragments
Localize and recognize products in video fragments
+
Neural
network
Output
62. 62
ELIS – Multimedia Lab
Automatic Understanding of Social Video Content (2/2)
Representation learning for social video
Learn general noise-robust features
Exploitation of temporal information in video to improve classification
Investigate recurrent neural networks and reservoir computing networks
63. Visualization
63
ELIS – Multimedia Lab
Future Research Vision SaVI & SWTF
Cognitive computing? Strong A.I.? Technological singularity ;-)?
Human &
machine action
Machine-understandable
information
Data
(online social networks &
Internet of Things)
Deep
learning
Semantic
Web
understanding
Natural
language
Visual
content
understanding
Application domains Technology stacks
65. 65
ELIS – Multimedia Lab
References
[1] F. Godin, B. Vandersmissen, A. Jalalvand, W. De Neve, and R. Van de Walle, “Alleviating manual
feature engineering for Part-of-Speech tagging of Twitter microposts using distributed word
representations,” Proceedings of the NIPS Workshop on Modern Machine Learning Methods and
Natural Language Processing, Dec. 2014.
[2] A. Tomar, F. Godin, B. Vandersmissen, W. De Neve, and R. Van de Walle, “Towards Twitter
hashtag recommendation using distributed word representations and a deep feed forward
neural network,” Proceedings of the IEEE International Workshop on Cyber-Physical Systems and
Social Computing (CSSC-2014) , Sep. 2014.
[3] F. Godin, J. Zuallaert, B. Vandersmissen, W. De Neve, and R. Van de Walle, "Beating the
bookmakers: leveraging statistics and Twitter microposts for predicting soccer results,“
Proceedings of the 2014 KDD Workshop on Large-Scale Sports Analytics, Aug. 2014.
[4] B. Vandersmissen, F. Godin, A. Tomar, W. De Neve, and R. Van de Walle, "The rise of mobile
and social short-form video: an in-depth measurement study of Vine," Proceedings of SoMuS
2014 : Workshop on Social Multimedia and Storytelling (co-located with ICMR 2014), Apr. 2014.