LECTURERS & COURSES

  1. Introduction to IR: Foundations and Models by Prof. Keith van Rijsbergen (University of Glasgow - United Kingdom)
  2. Task-based Information Interaction Evaluation by Prof. Kalervo Järvelin (University of Tampere - Finland)
  3. Cross-language and Multilingual IR by Prof. Martin Braschler (Zurich University of Applied Sciences - Switzerland)
  4. Scalability and Efficiency Challenges in Large-Scale Web Search Engines, by Dr. Berkant Barla Cambazoglu (Yahoo Research - Spain)
  5. Opinion Retrieval in Social Media by Prof. Georgios Paltoglou (University of Wolverhampton - United Kingdom)
  6. Faceted Search by Prof. Sebastien Ferre (University of Rennes 1, IRISA - France)
  7. Recommender Systems by Dr. Kostas Stefanidis (FORTH-ICS - Greece)
  8. IR Evaluation++ by Dr. Mihai Lupu (TU-WIEN - Austria)
  9. Interactive Content-based Multimedia Retrieval by Dr. Stefanos Vrochidis (Informatics and Telematics Institute - Greece)
  10. Bridging the Web of Documents with the Web of Data by Prof. Yannis Tzitzikas (University of Crete, FORTH-ICS - Greece)
  11. Contextual Search and Contextual Factors Aggregation by Prof. Gabriella Pasi (Milano Bicocca - Italy)
  12. Collaborative Information Seeking and Search Interfaces by Prof. Preben Hansen (Stockholm University, SICS - Sweden) 
  13. MUMIA: Integrating IR Technologies for Professional Search by Prof. Michail Salampasis (ATEI of Thessaloniki - Greece)

Keith van Rijsbergen Keith van Rijsbergen (homepage)
University of Glasgow - United Kingdom
Introduction to IR: Foundation and Models
Slides part1, Slides part2



 
Introduction in Information Retrieval (IR), concentrating on the basics, discussing foundational issues and elaborating on some of the common models for IR. At all times the emphasis will be at the underlying logics, geometry, and probability.




Short Biography

C.J. ‘Keith’ van Rijsbergen was born in Holland in 1943. He was educated in Holland, Indonesia, Namibia and Australia. He took a degree in mathematics at the University of Western Australia. As a graduate he spent two years tutoring in mathematics while studying computer science. In 1972 he completed a Ph.D. in computer science at Cambridge University. After almost three years of lecturing in information retrieval and artificial intelligence at Monash University he returned to the Cambridge Computer Laboratory to hold a Royal Society Information Research Fellowship. In 1980 he was appointed to the chair of computer science at University College Dublin; from there he moved in 1986 to the Glasgow University where he is Professor Emeritus. From 2008 to 2012 he was Visiting Professor in Residence at the Computer Laboratory in Cambridge University. Since January 2013 he is an Honorary member of that Computer Laboratory. He is a Fellow of the Royal Society of Edinburgh, Royal Academy of Engineering, BCS, and ACM.

Since about 1969 his research has been devoted to information retrieval, covering both theoretical and experimental aspects. He has specified several theoretical models for IR and seen some of them from the specification and prototype stage through to production. His current research is concerned with the design of appropriate logics to model the flow of information. He has been involved in a number of EU projects and
working groups on IR, including Fermi, Miro, Mira, Idomeneus, and more recently K-space. In 1993 he was appointed Editor-in-Chief of The Computer Journal, an appointment he held until 2000. He has served as a programme committee member and editorial board member of the major IR conferences and journals. He is the author of a well-known book Information Retrieval, Butterworths, 1979. In 1999, together with Crestani and Lalmas,he published a book entitled "Information Retrieval: Uncertainty and Logics". His most recent book is The Geometry of Information Retrieval, CUP, 2004, which inspired the EPSRC funded research project, Renaissance. In 2004 he was awarded the Tony Kent Strix award. In 2006, he was awarded the Gerard Salton Award. More details can be found in his Wikipedia entry.




Järvelin, Kalervo Kalervo Järvelin (homepage)
University of Tampere - Finland




Task-based Information Interaction Evaluation
Slides



 
It is commonly believed that information retrieval (IR) positively affects task performance. Traditional IR evaluation cannot justify such beliefs because the evaluation designs focus on the evaluation of the quality of the ranked output, excluding factors related to tasks and often also search interactions. The talk focuses on understanding and evaluation of information systems and information behaviors for their contribution to task performance. In Part I, extensions to the traditional Cranfield evaluation paradigm and user-centered evaluation are discussed. These cover evaluation of multiple query sessions though simulations and experiments with human participants. In Part II, an evaluation framework for task-based information interaction is proposed. It is based on generic activities in task-processes that can be supported by information systems. Tasks are understood as the larger tasks triggering information interaction. Particular focus is given to learning tasks to avoid the excessive complexity of embracing all possible larger tasks. Five generic activity types within information interaction in learning tasks are proposed: (1) task planning and reflective assessment, (2) searching sources, (3) selecting sources, (4) working with sources, and (5) synthesizing and reporting. Program theory is an approach to evaluation and used here to structure evaluation. A program theory aims to explicate how the system being evaluated is expected to achieve its goals regarding outputs and task outcomes in each activity. The talk presents, for each activity type, at the general level, the input, process, output and outcome factors and how they are connected across activity types. This type of modeling and analysis helps to evaluate information interaction, and its facilitating systems, properly for their contributions to task outcomes.




Short Biography

Professor Kal Järvelin is Professor of Information Retrieval at the School of Information Sciences, University of Tam­pere, Finland. He holds a PhD in Information Studies (1987) from the same university. He was Academy Professor, Academy of Finland, in 2004 –2009.  Kal Järvelin’s research covers information seeking and retrieval, linguistic and conceptual methods in IR, IR evaluation, and database management. He has coauthored over 280 scholarly publications. One of them is the co-authored seminal paper on the discounted cumulated gain evaluation metric (ACM SIGIR 2000 Best Paper Award). He has su­pervised 18 doctoral dissertations. Several of his former doctoral students have a recognized standing within Information Science. He has been a principal investigator of several research projects funded by EU, industry, and the Academy of Finland.

Kal Järvelin has frequently served the ACM SIGIR Conferences as a program committee member (1992-2009), Conference Chair (2002) and Program Co-Chair (2004, 2006, 2014); and the ECIR, the ACM CIKM and many other conferences as program committee member. He was an Associate Editor of Information Processing and Management (USA, 2008-2012).

Kal Järvelin received the Finnish Computer Science Dissertation Award 1986; the Tony Kent Strix Award 2008 in recognition of contributions to the field of Information Retrieval; and ASIS&T Research Award 2012 in recognition of contributions to the field of Information Science.





Martin Braschler (homepage)
Zurich University of Applied Sciences - Switzerland



Cross-language and Multilingual IR
Slides



A lot of early work in Information Retrieval was exclusively focused on retrieval of English text documents. This bias still exists to some extent today: many of the widely used mechanisms, such as popular weighting schemes, have been developed with the English language in mind and have been tested mainly on collections comprised of English text. They are often blindly used in non-English scenarios.

First serious steps to address multilingualism in IR better have started in the early 80s, and by the mid-90s, evaluation campaigns such as TREC had begun to include tracks concerned with non-English documents. A major boost was the founding of the NTCIR (for eastern Asian languages) and CLEF (for European languages) evaluation campaigns. The lecture will cover how to systematically extend basic monolingual indexing and matching, and adapt them for working with other languages. Issues such as language identification, tokenization/segmentation (including Asian languages), word normalization, stemming/decompounding are discussed. Throughout, numbers for the effect of these measures on retrieval effectiveness will be given.

Understanding how to adapt IR systems successfully for many languages is a necessary pre-requisite to tackle the problem of Cross-Language Information retrieval (CLIR), i.e. the retrieval of documents written in a language different to the language of the user's request. The lecture will cover the different translation strategies to address the CLIR problem. By extending the system into CLIR territory, issues of cultural differences, translation ambiguity, privot languages and extension to a large number of languages arise.



Short Biography

Martin Braschler holds the title of Professor ZFH and is a lecturer at the Zurich University of Applied Sciences. He heads the Information Engineering group, which is part of the Institute of Applied Information Technology. He studied computer science at the Swiss Federal Institute of Technology ETH in Zurich, leading to a degree of MSc ETH, and received the degree of Dr. sc. from University of Neuchatel. His doctoral thesis is entitled “Robust Multilingual Information Retrieval.” He visited the National Institute of Standards and Technology (NIST) in Gaithersburg, MD, USA, as a Guest Researcher in 1997. His main research interests include multilingual information retrieval, IR application evaluation and enterprise search. He has served as the technical coordinator of the cross-language track at the American TREC (Text Retrieval Conference) series of evaluation campaigns from 1997-1999, and was one of the original initiators of the European CLEF (Cross-Language Evaluation Forum) campaigns, of which he was technical coordinator from their start in 2000 until 2004. Apart from his academic record, which includes numerous research papers and third-party funded projects, he has a stong industrial background that includes extensive experience in technology transfer and implementation of IR solutions in the commercial marketplace.




Järvelin, Kalervo Berkant Barla Cambazoglu (homepage)
Yahoo Research - Spain



Scalability and Efficiency Challenges in Large-Scale Web Search Engines
Slides




Large-scale web search engines rely on massive compute infrastructures to be able to cope with the continuous growth of the Web and their user bases. In such search engines, achieving scalability and efficiency requires making careful architectural design choices while devising algorithmic performance optimizations. Unfortunately, most details about the internal functioning of commercial web search engines remain undisclosed due to their financial value and the high level of competition in the search market. The main objective of this tutorial is to provide an overview of the fundamental scalability and efficiency challenges in commercial web search engines, bridging the existing gap between the industry and academia.




Short Biography

Berkant Barla Cambazoglu received his BS, MS, and PhD degrees, all in computer engineering, from the Computer Engineering Department of Bilkent University in 1997, 2000, and 2006, respectively. He has then worked as a postdoctoral researcher in the Biomedical Informatics Department of the Ohio State University. He is currently employed as a senior researcher in Yahoo Labs, where he is heading the web retrieval group. He has many papers published in prestigious journals including IEEE TPDS, JPDC, Inf. Syst., ACM TWEB, and IP&M, as well as top-tier conferences, such as SIGIR, CIKM, WSDM, WWW, and KDD.




Ferre, Sebastien Sébastien Ferré (homepage)
University of Rennes 1, IRISA - France



Faceted Search
Slides



Context: Faceted Search (FS) is an interactive information access paradigm that offers more flexibility, and cover more search tasks, compared to search engines, database queries, and static navigational structures. It retains the usability, guidance and feedback of navigational structures while offering more expressivity. FS is user-centered, and supports exploratory search and understanding-at-a-glance.

Motivation: The Semantic Web (SW) provides more and more semantic data, aka. Linked Open Data (LOD), that is at the same level of abstraction as FS data but is essentially relational in nature. Semantic search remains a scientific challenge because full-text search is not so relevant, Web navigation can only answer the simplest questions, and SPARQL queries are as difficult to write as SQL queries. Therefore, we think that Faceted Search has a key role to play in semantic search to reconcile usability, interactivity and expressivity.

Learning objectives: This course will teach how the principles of faceted search can be successfully applied to semantic relational data. The course will be based on the technique of Query-based Faceted Search (QFS) that refines classical FS, and that enables to reach a very high expressivity, close to SPARQL, while retaining the simple interaction model and efficiency of FS. The technique is not tied to the Semantic Web, and can be adapted to other kinds of data and queries. At the end of the course, the participants should be able to adapt QFS principles to their own needs in interactive information access.

Course overview: There are no strict prerequisites for this course, but a basic knowledge of FS would be a plus. The course will (approximately) follow the sequence below.

1. FS compared to other information access paradigms;
2. The FS data and interaction model;
3. Existing extensions of FS for the Semantic Web, and their limits;
4. Query-based Faceted Search (QFS): merging FS interaction and query  languages;
5. Natural language syntax for queries;
6. QFS on top of SPARQL endpoints, with demo on DBpedia;
7. Summary, open issues, and discussion.

Learners with a computer and Internet connection will have the opportunity to experiment the presented interaction model (QFS) on public semantic datasets (via SPARQL endpoints).




Short Biography

Sébastien Ferré is an assistant professor in computer science at the University of Rennes 1, France, since 2004. His research takes place in the LIS team (Logical Information Systems), in the Data and Knowledge Management (DKM) departement of the IRISA laboratory. He holds a PhD in Computer Science from the University of Rennes 1 (2002), and has also been an assistant researcher at the University of Wales, Aberystwyth.

He has published papers about Formal Concept Analysis (FCA), logics for knowledge representation and reasoning, information retrieval and exploration, faceted search, Semantic Web, and controlled natural languages. His application domains have been personal information management, geographical information systems, bioinformatics, software engineering, and group decision support. He has also developed and maintains several softwares for the guided exploration of datasets: Camelis for collections of items with rich descriptors, Sewelis for RDF graphs, and more recently Sparklis on top of SPARQL endpoints. In the scope of MUMIA topics, Sébastien Ferré is a co-author of several chapters of the book “Dynamic taxonomies and faceted search”, edited by G.M. Sacco and Y. Tzitzikas.

Sébastien Ferré has taught programming languages, algorithmics (graphs, sequences), software engineering, compilers, symbolic datamining, and Semantic Web. He has taught at all levels, from under-graduate to master, in front of diverse audiences (computer science, business informatics, life sciences).




Hansen, Preben Preben Hansen (homepage)
Stockholm University, SICS - Sweden



Collaborative Information Seeking and Search Interfaces
Slides



This course attempts to introduce the topic of Collaborative Information Seeking (CIS) and its background and frameworks within the research areas of IR, IS and CSCW. It will also highlight concrete examples of current CIS systems, discuss CIS in the context of work-tasks, and 2 projects and illustrate some research in the area. Introducing CIS we address two fundamental questions. First, why collaborate for information seeking? Such situations require people coming together with intention, looking for and sharing information, and making sense out of it to reach their common goals. CIS can give teams (in companies or organizations) the capability to search more deeply and more broadly. Groups of people or work teams employ different strategies based on division of labor, member roles, and awareness of collaborators’ actions. Second, we consider how to collaborate for information seeking. This is still an open area of research. As scholars have found (e.g., Morris, 2008), there are many instances when people desire to collaborate on information seeking tasks, but either they find a lack of tools to support their need, or simply do not bother to depart from tried-and-tested methods including email, chat, and teleconferencing.




Short Biography

Preben Hansen is an Associate Professor at Stockholm University (Dept. of Computer and Systems Sciences), Sweden. His research interests include interactive information seeking and retrieval, human-computer interaction and work-task based research in collaborative contexts. He is working with the EU-projects Promise, MUMIA and KNOWeSCAPE. He received a PhD in information studies from University of Tampere, Finland, supervised by Professor Kalervo Järvelin.




Stefanidis, Kostas Kostas Stefanidis (homepage)
FORTH-ICS - Greece



Recommender Systems
Slides



With the growing complexity of WWW, users often find themselves overwhelmed by the mass of choices available. Shopping for DVDs, books or clothes online becomes more and more difficult, as the variety of offers increases rapidly and gets unmanageable. To facilitate users in their selection process, recommender systems provide suggestions on items, which could be interesting for the end user. In particular, recommender systems aim at giving recommendations to users or groups of users by estimating their item preferences and recommending those items featuring the maximal predicted preference. The prerequisite for determining such recommendations is historical information on the users' interests, e.g., the users' purchase history.

The goal of this tutorial is to give an overview and general understanding of the existing approaches for computing recommendations. Since, typically, recommendations are established by considering users sharing similar preferences as the query user or group, and scanning the whole database to find such like-minded users is a costly process, in this course, we will give special emphasis on recent approaches that are based on building user models. One way to do this is by applying clustering for organizing users into clusters and using these clusters, instead of linear scanning the database, for predictions. Moreover, we will focus on several approaches that handle different temporal aspects of recommendations. We will consider two different types of time effects based upon the freshness and the temporal context of the user preferences, or ratings.




Short Biography

Kostas Stefanidis is a collaborating researcher at the Institute of Computer Science of the Foundation for Research and Technology in Greece. Previously, he worked at the research group of Prof. Kjetil Norvag at the Norwegian University of Science and Technology, and at the group of Prof. Yufei Tao at the Chinese University of Hong Kong. He got his PhD in personalized data management from the Computer Science Department of the University of Ioannina, Greece, in 2009, under the supervision of Prof. Evaggelia Pitoura. He obtained his MSc and BSc degrees from the same university in 2005 and 2003, respectively. His research interests lie in the intersection of databases, web and information retrieval, and include personalized and context-aware data management systems, recommender systems, keyword-based search, user modeling, information extraction, resolution and integration, and social media. Kostas has served as a PC co-chair of the ExploreDB 2014 workshop, in conjunction with EDBT/ICDT 2014, as a reviewer in the board of several journals, including IEEE TKDE, Elsevier IS, Elsevier JWS, and as a PC member in various conferences and workshops.




Lupu, Mihai Mihai Lupu (homepage)
TU-WIEN - Austria



IR Evaluation ++
Slides



Information Retrieval is a heavily experimental science, which has consistently relied on experimental results to drive its progress. For any researcher, young or not, it is important to have an overview of the best practice in evaluation, whether they are planning research on IR Evaluation itself or in any other Information Access-­‐related area. Participants will gain an overview of the available collections, tools and best practices. At the end of the lecture, you will be able to identify the best way to test their own experiments, to consider critically experiments presented in other papers, and to create your own test collections, if none is available. The lecture will cover the fundamental issues of IR evaluation, as observed in evaluation campaigns, considering the perspectives of both their supporters and critics. The lecture is entitled IR Evaluation++ because, while focusing on the large amount of know-­‐how present in the IR community, it also touches upon other aspects of Information Access Evaluation and puts the work in the context of professional environments. 




Short Biography

Mihai Lupu has obtained his PhD from the Singapore-­‐MIT alliance at the National University of Singapore in November 2008. Since then, his work has concentrated on evaluation of domain-­‐specific IR systems. He has been the co-­‐organizer of the TREC Chemical Track, the CLEF Intellectual Property Track and of the Workshop on Patent Information Retrieval. More recently, Mihai is also involved in the Retrieving Diverse Images Track at MediaEval 2014, bringing his expertise on text-­‐retrieval evaluation in close cooperation with colleagues working on image retrieval evaluation. Mihai is currently a post-­‐doctoral researcher at the Vienna University of Technology, where he also lectures on Information Retrieval to MSc students in Informatics and Business Informatics. Mihai is also working as an independent consultant on patent search, currently devoting part of his time to the European Patent Office.




Ferre, Sebastien George Paltoglou (homepage)
University of Wolverhampton - United Kingdom



Opinion Retrieval in Social Media
Slides



The course will introduce its audience to the basic principles of Opinion search in social media. Opinion search deals with discovery and retrieval of content, primarily from social media, that is relevant to the user’s information needs and contains opinions that pertain to them. It combines methodologies from two distinct areas of research: information retrieval and sentiment analysis. The former deals with the issues of representing, storing and providing access to information, while the latter focuses on the detection, extraction and analysis of affective content. In the session, we will provide a brief but concise introduction to the area, focusing on the most relevant and influential work that has taken place in both distinct areas of research, as well as discuss how those approaches can be combined effectively and efficiently to fulfil the field’s stated goal.

During the session, we will discuss the particular challenges in the Opinion retrieval field and describe recent research that attempts to address them. Concretely, we will examine the issue of IR within social media, which significantly differs from standard web-based search due to the unique nature of the domain (e.g., ephemerality of content, importance of timeliness). Similarly, we will define and provide solutions to the challenges of conducting sentiment analysis in social media content, such as the extensive use of informal, abbreviated language and unique prose elements, like hashtags. We will also formally define and analyze the constituent parts of “opinions” from a linguistic point of view and discuss their role in the effective analysis of relevant and opinionated content. Importantly we will address the issue of combining the research findings of those two areas of research in successfully ranking opinionated documents in response to user’s information needs.





Short Biography

Dr Georgios Paltoglou is a Senior Lecturer at the Faculty of Science and Engineering of the University of Wolverhampton, UK and a member of the Statistical Cybermetrics Research Group (SCRG) of the same university. His research interests include sentiment analysis, information retrieval and machine learning. He has 17 journal publications with an average impact factor of 2.27 (including prestigious journals such as IP&M, IEEE TAC, JASIST and others), 3 book chapters (in press), and 18 publications in peer-reviewed conferences and workshops. His work has been cited over 800 times and has an h-index of 13.
 
He has also given numerous public talks and seminars on information retrieval and sentiment analysis, including talks at the University of Sheffield (6/12/2013), Aston Business School (25/5/2013) and the Oxford Internet Institute (1/3/2013). He has been invited and given lectures in 2 international research summer schools, including the 9th European Summer School in Information Retrieval (ESSIR 2013) and the 1st Mu.M.I.A. Training School – Building Next Generation Search Systems.




Pasi, Gabriella Gabriella Pasi (homepage)
Milano Bicocca - Italy



Contextual Search and Contextual Factors Aggregation
Slides



The central concept in IR is the concept of relevance; the main objective of Search Engines is to estimate the relevance of Web pages to the information needs expressed in the user’s query. However, only the user can determine the true relevance of a document, i.e., the usefulness, pertinence, appropriateness, or utility of that document with respect to the user search intents formally sketched in a query. Relevance is time, situation, and user specific; in assessing the utility of a document, users are influenced by many factors that go beyond topical relevance, i.e. whether or not a given document is about the topics covered in the users’ queries. Relevance is in fact a multi-dimensional notion that depends on many complex factors related to the search task, to the user, to her/his context, and to the document and search context as well. An effective estimate of the relevance of documents to a query should take into account a context model. An important objective in contextualizing search is not to burden the user-system interaction with the need of explicitly specifying contextual information; on the contrary, a good context aware system should be able to identify the useful context information by making the user unaware of this. Moreover a good system should be able to detect the often complex and dynamic characteristics of users search tasks. This lecture addresses the problem of contextualizing search, by trying to outline both the research challenges and the user role and involvement in the various identified tasks. Moreover,  the important issue of aggregation in contextual search will be addressed.





Short Biography

Gabriella Pasi received a PhD in Computer Science at the Université de Rennes, France. She has been working at the National Council of Research in Italy till 2005. She is actually Associate Professor at the Università Degli Studi di Milano Bicocca, Milano, Italy, where she leads the Information Retrieval Research Laboratory. Her research activities mainly focus on personalised/contextual access to information, and on the problem of aggregation in search. She has served as the Program Chair of several international conferences and workshops, and she has been the chair or co-chair of several International events among which the IEEE / WIC / ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology, Università degli Studi di Milano Bicocca, 15-18 September 2009, the PhD School on Web Information Retrieval (WebBar 2007), the Seventh International Conference on Flexible Query Answering Systems (FQAS 2006), the European Summer school in Information Retrieval (ESSIR 2000), and the annual track “Information Access and Retrieval” within the ACM Symposium on Applied Computing. She has published more than 180 papers on International Journals and Books, and on the Proceeding of International Conferences, and she is member of the Editorial Board of the several International Journals.




Tzitzikas, Yannis Yannis Tzitzikas (homepage)
(University of Crete, FORTH-ICS - Greece)



Opening of 3rd MUMIA Training School
Opening



Bridging the Web of Documents with the Web of Data
Slides



Closing of 3rd MUMIA Training School
Closing



This tutorial aims at showing how one could bridge the Web of Documents with the Web of Data at Search Time. At first the tutorial will describe the required background and context. Then it will describe the ways semantic data can affect and search process, and finally it will focus more on methods for semantic post-processing search results. For this topic, the tutorial, will provide the motivation, will describe the vision and the related challenges, will summarize what existing approaches and systems can do and cannot do, will demonstrate such systems (publicly available and not, including those designed by the tutor), and sketch directions for future research.

Learning objectives:
  • Provide an overview of the current semantic technologies
  • Understand how existing search systems currently exploit and integrate semantic information
  • Understand the limitations of the current approaches
  • Understand the value of semantic post-processing of search results
  • Deliver the practical experience of designing and implementing such systems
  • Understand the difficulties in offering an effective interaction model
  • Understand the difficulties in evaluating such systems
  • Investigate research opportunities



Short Biography

Yannis Tzitzikas is currently Assistant Professor of Information Systems in the Computer Science Dep. at University of Crete (Greece) and Associate Researcher of the Information Systems Laboratory at FORTH-ICS (Greece). Before joining UofCrete and FORTH-ICS he was postdoctoral fellow at the University of Namur (Belgium) and ERCIM postdoctoral fellow at ISTI-CNR (Pisa, Italy) and at VTT Technical Research Centre of Finland. He conducted his undergraduate and graduate studies (MSc, PhD) in the Computer Science Department at the University of Crete. In parallel, he was a member of the Information Systems Lab of FORTH-ICS for about 8 years, where he conducted basic and applied research around semantic network-based information systems within several EU-founded research projects. His research interests fall in the intersection of the following areas: Information Systems, Information Indexing and Retrieval, Conceptual Modeling, Knowledge Representation and Reasoning. The results of his research have been published in more than 80 papers in refereed international conferences and journals, and he has received two best paper awards (at CIA’2003 and ISWC’07). He is one of the editors of the book Dynamic Taxonomies and Faceted Search (Springer 2009), he is currently involved in the ongoing EU projects iMarine (FP7 Research Infrastructures), SCIDIP-ES (FP7 Research Infrastructures), Aparsen (FP7 Network of Excellence), and national contact of the MUMIA Cost Action (vice chair of WG4). His current research revolves around Faceted Interactive Search, and lately on methods that can bridge the web of documents with the web of data.




Vrochidis, Stefanos Stefanos Vrochidis (homepage)
Informatics and Telematics Institute - Greece



Interactive Content-based Multimedia Retrieval
Slides



The rapid advances of digital technologies have resulted in a great increase of multimedia content worldwide. This calls for the need for the research and development of intelligent content-based methods for multimedia indexing and retrieval. In parallel, the exploitation of aggregated implicit user feedback is an important research challenge given the large amounts of interaction data that are available nowadays (i.e. log files, social media interactions).

This course presents an overview of the techniques that deal with the aforementioned research challenges and includes three main parts: a) an overview of content-based multimedia analysis and retrieval techniques; b) the exploitation of implicit user feedback during interactive image and video search; c) demonstrations of image and video search systems.

The first part will present techniques for video segmentation and low-level feature generation from audiovisual content.  In addition it will discuss semantic indexing of video and images using concepts and events based on supervised machine learning. Finally, we will present the main retrieval functionalities supported by multimedia search systems. Since the focus is on interactive systems, the second part of the course will deal with the exploitation of implicit user feedback. Specifically we will discuss the modeling of user navigation patterns, such as mouse clicks, as well as the exploitation of involuntarily actions such as gaze movements using graph representation and machine learning techniques. In the third part, the lecture will include a live demonstration of interactive multimedia systems (including patent search) with discussion on how the aforementioned techniques are integrated and applied.

The learning objective of the course is to provide an overview of video and image indexing and retrieval techniques and also provide an introduction to techniques for the exploitation of user interaction during multimedia retrieval tasks. The course will also serve as a background for developing interactive multimedia search engines.




Short Biography

Dr. Stefanos Vrochidis received the Diploma degree in Electrical Engineering from Aristotle University of Thessaloniki, Greece, the MSc degree in Radio Frequency Communication Systems from University of Southampton and the PhD degree in Electronic Engineering from Queen Mary University of London. Currently, he is a researcher with the Multimedia Knowledge and Social Media Analytics Lab of the Information Technologies Institute at the Centre for Research and Technology Hellas. His research interests include semantic multimedia analysis, indexing and retrieval, semantic search, multimedia search engines and human interaction, as well as environmental applications and patent search. Currently, Stefanos Vrochidis is the deputy Project Coordinator and Scientific Manger of the fp7 STREP project MULTISENSOR, which deals with multidimensional content integration and multimedia retrieval for intelligent media monitoring and the Interaction Coordinator of the Cost Action: European Network on Integrating Vision and Language (iV&L Net): Combining Computer Vision and Language Processing For Advanced Search, Retrieval, Annotation and Description of Visual Data. Dr. Vrochidis has successfully participated in many European and National projects and he has been involved as a co-author in more than fifty related scientific journal, conference and book chapter publications. He serves as regular Program Committee Member and Reviewer in well reputed conferences and journals such as the IEEE International Conference on Image Processing, ACM International Conference on Multimedia Retrieval, European Conference on Information Retrieval, IEEE transactions in Multimedia, Multimedia Tools and Applications Journal and Elsevier Digital Signal Processing.





Salampasis, Michail Michail Salampasis (homepage)
Alexander Technology Educational Institute (ATEI) of Thessaloniki - Greece




MUMIA: Integrating IR Technologies for Professional Search
Slides



This talk will present MUMIA which is a research networking activity that brings together various facets of state-of-the-art search technology research which can contribute to the development of search tools for next generation professional search systems. Additionally, it presents a general framework which provides a useful topology for better understanding the design space of professional search systems and how different IR/NLP technologies can be integrated to enable rich information seeking environments where different tools can support specific objectives within a typically lengthy search process. The main motivation of the talk is to present they key challenges, a framework and the best practices which will influence the design and development of next generation professional integrated search systems.





Short Biography

Dr Michail Salampasis is an associate professor at the department of Informatics of the Alexander TEI of Thessaloniki. He holds a B.Sc. in Informatics (1993) and a Ph.D. in Computing (1997). He was candidate for the best PhD thesis award in UK. His main research interests are in applied and interdisciplinary studies in information science, including models and experiments related to information seeking behaviour, information seeking in large professional search systems, Web information seeking and evaluation, distributed information retrieval including source selection and results merging algorithms, search systems usability testing, information seeking using multiple strategies/interfaces. He currently is the coordinator of the Cost Action a research networking activity on “Multilingual and Multifaceted Interactive Information Access (MUMIA)” and a Marie Curie Fellow at the Institute of Software Technology and Interactive Systems, Vienna University of Technology leading a research program for Personalised Federated Patent Search Systems (PerFedPat). Michail Salampasis has published about 70 papers in refereed journals, conferences and book chapters in various fields of computing.





 
Facebook Image

Upcoming Events

3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014


3rd Open Interdisciplinary MUMIA Conference

The 3rd Open Interdisciplinary MUMIA Conference will take place in Copenhagen on 11-12 November 2014

Read more


8th MC/final evaluation meeting

The 8th MC MUMIA /final evaluation meeting will take place in Copenhagen on 12 November 2014