Date: 5/5/2012
Time: 11am - 12noon.
Location: AG06, College Building
Title: Visual Analytical Approaches to Evaluate Uncertainty and Bias in Crowdsourced Crisis Information
Abstract: Increasing numbers of people are using social media to exchange information during crisis and conflict events. On the one hand, the humanitarian community is reluctant to use this information in the response effort as it fears the cost of untrustworthy and inaccurate information. On the other, the volunteer and technical communities have attempted to resolve this impasse by crowdsourcing crisis information; for example, by asking volunteers to ascertain whether a crisis report is trustworthy and accurate.
Trust and accuracy are two characteristics of uncertainty: The fact that each is likely to have spatial, temporal and thematic aspects is supported by research, which suggests that geography characterises crisis information. Consequently, a research programme grounded in geographic information science, (geo)visualization and (geo)visual analytics is presented that 1) seeks to evaluate the degree to which uncertainty and bias (systematic variation) are found in crowdsourced crisis information; and 2) provide heuristics to help manage these factors. This programme consists of 1) a methodology for undertaking interactive, analysis-guided software development that is informed by action research, scenario-based design and Munzner's model of visualization validation; and 2) a prototype software application that combines interactive visual representations with spatial statistical functions to explore two datasets of crowdsourced crisis information.
Following a review of the literature and a description of the data, the methodology and its implementation are placed within an appropriate work plan. Three supporting publications are included, as well as supporting statements regarding the author's skills and engagement with the academic community.
Speaker: Stephann Makri
Room: AG07, College Building
Date: Monday 12th March 2012
Time: 1:00-2:00pm
Abstract: We wanted to gain a detailed empirical understanding of how researchers come across information serendipitously, grounded in real-world examples. To gain this understanding, we asked 28 researchers from a broad cross-section of disciplines to discuss in detail memorable examples of coming across information serendipitously from their research or everyday life. We found that although the examples provided were varied, they shared common elements (specifically, they involved a mix of unexpectedness and insight and led to a valuable, unanticipated outcome). These elements form the core of 1) a descriptive model of serendipity and 2) a framework for subjectively classifying whether or not a particular experience might be considered serendipitous and, if so, how serendipitous. In this talk, we discuss this model and framework and the implications of our findings on the design of interactive information resources.
Bio: Stephann is a Research Associate at UCL Interaction Centre and is conducting research as part of a on a £1.87m UK Research Council funded project (SerenA: Chance Encounters in the Space of Ideas) which aims to understand how people come across information 'serendipitously' and to design ubiquitous computing systems based on this understanding.
Speaker: Sandra Tury
Room: AG07, College Building
Date: Friday 27th January 2012
Time: 1:00-2:00pm
Abstract:
Understanding the Information seeking behaviour of distance learners will lead to better-designed online libraries that effectively support the information and learning needs of distance learners. It will also inform a debate on how national guidelines for distance learning library provision in the UK could be formulated. For purposes of this paper I have concentrated on the initial phase of my research which is in two parts. The first part is a review of literature in the field of information seeking behaviour (part 1A) which includes an evaluation of relevant models in order to assess their practical application to the distance learning context and an assessment of existing distance library provision in the UK(Part 1B). The review identified the questions already addressed in published research and gaps in existing research and enabled a model to be selected as a framework for further research. Part 2 is a pilot study which was undertaken in order to inform the methodology of the research and an analysis of its results and the implications for the subsequent main study.
Speaker: Andrew Robson
Room: AG11, College Building
Date: Thursday, 17/11/2011
Time: 1:00-2:00pm
Abstract:
Evidence-based clinical practice requires health professionals to use up-to-date information when making treatment decisions. It thus provides a fertile field in which to study information behaviour, including information seeking, information use and communication. There are many theories and models of information behaviour but most of them focus on information seeking or communication rather than considering how the two types of activity link together and affect each other. The aim of this research is to develop a new more comprehensive model, building on insights from existing frameworks, and to test its validity. This report first reviews the literature on the information behaviour of physicians and considers this in the context of evidence-based clinical practice in the UK. Established models of information seeking and communication are then described and a new model is put forward, which covers both information seeking and communication. Then the planned empirical research to test the model in the health care environment is described including the aims, methods and timetable.
Clare Thornley (University College Dublin) - Speaker
Andrea Johnson (University College Dublin)
Alan Smeaton (Dublin City University)
Hyowon Lee (Dublin City University)
Room: AG08, College Building
Date: Friday, 11/11/2011
Time: 1:00-2:00pm
Abstract:
This seminar reports on a scholarly impact and bibliometric study of the TRECVid (TREC Video Retrieval Evaluation) benchmarking conferences between 2003 and 2009. The contribution of TRECVid to research in video retrieval was assessed by analysing publication content to show the development of techniques over time and by analysing publication impact through publication numbers and citation analysis. For a selection of participants at different career stages, the relative importance of TRECVid publications in terms of citations vis a vis their other publications was also investigated. We also investigated whether papers reporting on successful video retrieval techniques (in terms of scoring against TRECVid benchmarks) also tended to receive more citations than those reporting on less successful techniques.
The theoretical and methodological implications of the study in terms of the value of bibliometric analysis as a research evaluation tool will also be discussed. These include: motivations for scholarly impact studies; the advantages and disadvantages of the different tools available (Google Scholar, Scopus, Web of Science); different approaches to data cleaning including the use of expert input and automated methods; the significance or otherwise of the results; differences between disciplines; the possible role of emerging techniques for measuring usage as a factor in ascertaining scholarly impact.
Speaker: Kashif Khan
Room: A130, College Building
Date: Tuesday, 24/05/2011
Time: 14:00-15:00pm
Abstract:
The research is intended to be an analytical study of the influence of globalization on EU information policy. It examines how well defined and implemented information policy in turn can create a more closely knit global society by examining specific directives currently in place within the EU. This will allow for a better understanding of the complex information policy environments and their interaction with society. The existing EU Information Policy framework broadly focuses on three main areas; Privacy and Data Protection, Intellectual Property Rights and Electronic Commerce. To focus the research on EU Information Policy as a complete entity, the study aims to analyse all the three main components of the EU Information Policy framework by analyzing the EU Directive on Privacy and Electronic Communications, the EU Copyright Directive, also called as the Information Society Directive and the EU Electronic Commerce Directive.
The study will examine the inter-relationships between the three policy areas. The study will use the grounded theory approach to develop model based on the data gathered through literature reviews, documentary analysis, interviews and the case study on information policy in the UK NHS. The study assumes that it is the effect of globalisation on the free movement of information across geographical borders in a digital world, which has presented a real need to harmonize EU information policy in this respect. The research aim therefore being to further explore and analyze the ways in which the three Directives were formulated, with a view to propose ways to future proof these policies in light of evolving technological and political challenges.
At a substantive level, the research consists of a study and analysis of the inter-connected issues surrounding the three Directives to examine the development process of those Directives. At a methodological level, the research aims to provide an analysis & propose a framework, to allow for the information policy development, with clear roles and responsibilities of each policy actor and expected outcomes at each stage of policy development. At a theoretical level, the aim is to use the case-study and all analytical work to conceptualize the use of the proposed information policy framework and therefore allow for further academic study. The research concludes a presentation of an analysis of the relationship between information policy and flow of information in society. The expected contribution is directed to integrating a social theory on the flow of information in society and understanding the creation of an IP framework/model based on this theory.
Speaker: Jonathan Levitt
Room: A227, College Building
Date: Wednesday, 18/05/2011
Time: 13:00-14:00pm
Abstract:
This talk examines ways in which the analysis of citation data has been
used to inform policy decisions on research. 'Citation' can be defined in
terms of 'reference'; Article A receives a citation from Article B, if,
and only if, Article B refers to Article A. Although Dr. Levitt's
research focuses on the potential application of citation analysis to
research policy, this talk does not presuppose knowledge of citation
analysis or research policy. It focuses on two perceptions in research
policy: (a) interdiscipinarity research is conducive to higher quality
research and (b) collaboration in research is conducive to higher quality
research.
Bio:
Jonathan Levitt first studied mathematics, but because of substantially
reduced sight could not complete his doctoral research. In 1997, he
obtained an M.Sc. in computer science and, after working as a software
engineer at NTL Internet, he became an associate lecturer in computer
science at the Open University in 2000. He subsequently obtained an M.A.
in social research methodologies, a Ph.D. specialising in scholarly
communication, and from 2009 to 2010 was an ESRC Postdoctoral Fellow at
Loughborough University. In November 2010 he secured, as Project Head and
Principal Investigator, ESRC funding to investigate the effective
allocation of research funding. Dr. Levitt was a research student at City
University from October 2003 to September 2004.
Speaker: Cassidy Sugimoto, School of Library and Information Science, Indiana University Bloomington
Room: AG08, College Building
Date: Friday, 06/05/2011
Time: 13:00-14:00pm
Abstract:
This talk will explore the use of academic genealogy networks to explore
the formation, maturation, and intersection of disciplines. Using LIS
as a case study, this presentation will explore the potential
applications of these networks for providing empirical evidence to
describe the development of disciplines. The talk focus on issues of
maturation and interdisciplinarity and will review potential sources and
tools for collecting and analyzing academic genealogy networks. Future
research and broad applications for this topic will be discussed.
Speaker: Ying Ding School of Library and Information Science Indiana University Bloomington
Room: AG08, College Building
Date: Tuesday, 03/05/2011
Time: 13:00-14:00pm
Abstract:
Scientific collaboration and endorsement have been well-established
research topics utilizing three kinds of methods: survey/questionnaires,
bibliometrics, and complex network analysis. Still there are many
interesting questions remaining unanswered. This paper applies the
combination of topic modeling and path-finding algorithms to address
some of them, such as, whether productive authors like to collaborate
with or cite researchers with same or different interests, and whether
highly cited authors like to collaborate with or cite each other. Taken
Information Retrieval as a test field, we found that productive authors
tend to directly coauthor with and closely cite colleagues sharing same
research interests; they do not directly collaborate with but closely
cite colleagues having different research topics; and highly cited
authors indirectly coauthor with but closely cite each other.
Room: AG08, College Building
Date: Monday, 14/03/2011
Time: 13:00-14:00pm
Title: Is Crowdsourcing the Answer to the Evaluation of Book Search Engines?
Abstract:
Crowdsourcing has become a widely popular mechanism for solving a range of
human intelligence tasks. Such tasks include the labelling of images or
search results, a job where humans still outperform machines. As a result,
crowdsourcing is increasingly relied upon as a feasible alternative to
traditional methods of gathering relevance labels for the evaluation of
search engines. Examples include the INEX Book Track where the scale of
the assessment task rendered traditional methods of relevance data
gathering inhibitive.
However, crowdsourcing raises a range of questions regarding the quality of the resulting labels. What indeed can be said about the quality of the data that is contributed by anonymous workers who are only paid cents for their efforts?
In this talk, I will provide an introduction into the world of crowdsourcing for search engine evaluation with specific focus on considerations for quality control within the design of crowdsourcing experiments. I will then discuss the findings of a recent large scale crowdsourcing experiment to gather relevance labels for the evaluation of the INEX Book Track. The experiments offer insights that can aid in the design of HITs for improved output quality.
Bio:
Gabriella Kazai is a research consultant, working for Microsoft Research
in Cambridge, UK. Her research interests include crowdsourcing, social
information retrieval, IR evaluation measures, test collection building,
book search and active reading, and personal digital libraries. She is
founder and organiser of the INEX Book Track since 2007, in the context of
which she developed a crowdsourcing system for collecting relevance
judgements for digitized books as part of a social game. She is also
currently working on a book on Crowdsourcing for Search Engine Evaluation
with Omar Alonso and Stefano Mizzaro. Gabriella holds a PhD in Information
Retrieval from Queen Mary University of London. She published over 40
papers and organised several IR conferences and workshops.
Date: Monday 28th February 2011
Time: 1.00-2.00pm
Place: AG07, College Building
Title: Bargain of the year - website, academic and digital library visibility, plus effective information retrieval, plus a case study, in 1 hour flat....
Aim: The main aim of this seminar is to provide the attendee with academic insight into website visibility. Secondly, it is to show how this should be applied in various areas of academic existence.
Abstract:
This seminar combines theoretical presentation with live, practical
examples. The principles of website visibility as determined by academic
research rather than popular belief are uncovered. These are then applied
to; a university website, personal academic visibility, and the visibility
of pdf documents stored in university digital libraries. Three common
searching methodologies are tested on a sample of university digital
libraries, proposing the most efficient one as a search methodology of
choice.
Biography:
Melius Weideman is currently a Head of Department in the Faculty of
Informatics and Design, Cape Peninsula University of Technology in Cape
Town. After working in the electronics and computer industry, he joined
the academe in 1984. His research interests were initially focussed on
computer viruses, but after 1994 the Internet, and specifically search
engines started fascinating him.
He graduated with a Doctorate in Information Science from the University of Cape Town in 2001. Numerous of his publications have seen the light since then on topics including website visibility and usability, search engines and information retrieval. He published an academic book in 2009, titled: "Website Visibility: The theory and practice of improving rankings". Implementation of website visibility by publishing academics has become one of his trademarks.
Date: Monday 31st January 2011
Time: 1pm to 2pm.
Venue: AG08, College Building
Title:
Interacting with Knowledge, Interacting with People: Web Searching in
Early Childhood
Abstract:
The presentation explores new and emerging dimensions in our understanding
of information behaviour and Web searching abilities in early childhood.
Spink (2010) proposed that information behaviour is shaped by both
instinctive and environmental dimensions, and emerges in early childhood.
The presentation explores what we currently know about the development of
Web searching abilities in young children. We report results from a recent
study of Web searching by four and five year old children who were video-
and audio-taped Google Web searching in a preschool classroom (Spink, et
al., 2010). The study found that young children engage in complex Web
searches and information behaviours, including keyword searching and
browsing, query formulation and reformulation, relevance judgments,
successive related Web searches overtime, information multitasking
(multiple topics in single Web search sessions) and collaborative
behaviours. The next research phase is a large Australian Research Council
funded study investigating the extent of young children's Web searching in
classrooms and home environments, what they access and in what social
contexts. Findings will inform educators, Web designers and families about
young childrens' Web use for socially interactive learning and
knowledge-building.
Local search is of interest in IR, for problems which are computationally complex and are difficult to solve in a reasonable timeframe. In this talk we give background on previous resesarch in the area, using a query based framework and give some areas on research issues which are to be tackled. We give a brief overview of the work led by City on the term selection problem, using Hill-climbers and GA's.
It is widely believed that information systems and technologies can function as engines for political and social change in authoritarian societies. This politicised version of technological determinism has almost gained ontological status of common sense although little evidence has been provided in support of such claims by academic disciplines. As one of the last communist countries in the world, Cuba offers a very interesting scenario to study how different societies actively adopt alternative appropriation models of information systems and technologies in such a way that, rather than undermining social structure, help re-produce the political system.
The seminar concentrates in outlining the main characteristics of the Cuban model of information production, transfer, and consumption that shapes and informs national information policies and strategies. The emphasis is placed in the conceptual frameworks, discourses, metaphors, and representations that underpin and legitimise information policies. The aim is to illustrate how ruling groups in Cuba proactively work through discourse the appropriation and use of information systems to promote their own interests and priorities. The role of the symbolic universe surrounding technologies in shaping how Cubans make sense of information systems, and what they can and should be used for is also highlighted.