Warning: Your browser doesn't support all of the features in this Web site. Please view our accessibility page for more details.
Thursday December 13th in IT207 from 14:00-15:00.
You are all invited to attend and biscuits, tea/coffee will be available in the staff room following the talks. Please circulate to anyone who maybe interested.
Title: Games AI in an Online Car Combat/Racing Game
Abstract: This presentation will discuss key aspects of the Artificial Intelligence (A.I.) techniques underlying the author's online game "Darkwind: War on Wheels". Topics will include: the use of Genetic Algorithms for racing-line optimisation; the development of a novel, crowd-sourced approach to pathfinding; and, a first look at 2.6 million weaponsfire datapoints which have been gathered over the past 18 months, and some ideas about how these might be use to improve tactical decision-making.
Padraig O'Flaithearta
Title: Optimising QoS of VoIP over Wireless LANs via Synchronized Time
Abstract: As well as supporting conventional data applications such as email, file transfer and web access, mobile handheld devices that are WiFi-enabled are being used increasingly for Voice over IP (VoIP). Wireless LANs (WLANs) are thus increasingly required to support Quality of Service (QoS)-centric applications, which are delay sensitive and require a certain level of throughput. Whilst 802.11e goes some way towards meeting this need, severe congestion leading to unacceptable delays and packet loss can still occur. In the existing 802.11e standard, all VoIP sessions contend within the same prioritization Access Category (AC), despite having potentially varying M2E (Mouth to ear) delays. In this paper we show how synchronized time can help optimize 802.11e EDCA parameters in order to prioritize VoIP sessions with relatively large M2E delays and thus distinguish between VoIP sessions. Using the NS-3 Network Simulator, we quantify the benefits achievable through synchronization of an 802.11e network handling multiple VoIP calls in the presence of other TCP traffic. We also present our EDCA tuning algorithm which uses the E-Model R-Factor QoS rating mechanism as the basis of control.
Abstract: Using Twitter as an effective marketing tool has become a gold mine for companies interested in their online reputation. A quite significant research challenge related to the above issue is to disambiguate tweets with respect to company names. In fact, finding if a particular tweet is relevant or irrelevant to a company is an important task not satisfactorily solved yet; to address this issue in this paper we propose a Wikipedia-based two-pass algorithm. The experimental evaluations demonstrate the effectiveness of the proposed approach.
Abstract: The aim of our work is to easily monitor the reputation of a company in the Twittersphere. We propose a strategy that organizes a stream of tweets into different clusters based on the tweets topics. Furthermore, the obtained clusters are prioritized into different priority levels. A cluster with high priority represents a topic which may affect the reputation of a company, and that consequently deserves immediate attention. The evaluation results show that our method is competitive even though the method does not make use of any external knowledge resource.
1pm, IT 206 (IT building)
1pm
1pm
Abstract. Modelling the score distribution of documents returned from any information retrieval (IR) system is of both theoretical and practical importance. The goal of which is to be able to infer relevant and non-relevant documents based on their score to some degree of confidence.
In this paper, we show how the performance of mixtures of score distributions can be compared using inference of query performance as a measure of utility. We (1) outline methods which can directly calculate average precision from the parameters of a mixture distribution. We (2) empirically evaluate a number of mixtures for the task of inferring query performance, and show that the log-normal mixture can model more relevance information compared to other possible mixtures. Finally, (3) we perform an empirical analysis of the mixtures using the recall-fallout convexity hypothesis.
--------------------------------------------------------------
--------------------------------------------------------------
--------------------------------------------------------------
Date: Wednesday, 12 October, 2011
Time: 15:00
Location: IT202 (IT Building)
Speaker: Dr. Donn Morrison (Digital Enterprise Research Institute)
Title: TagCaptcha: Annotating images with CAPTCHAs
Abstract: In this talk I will present a review of existing CAPTCHA technologies
(Completely Automated Public Turing test to tell Computers and Humans Apart) as well as a novel image-based CAPTCHA that is used to annotate images for use in a retrieval setting. The annotation system, called TagCaptcha, presents the user with a number of images that must be correctly labelled in order to pass the test. Given a partially annotated database used as a control set, the unannotated images will be incrementally labelled over time. I will examine robustness against automated attacks, report on usability results from a small user study and present sample annotations from the online demonstration system.
You are cordially invited to attend the next IT seminar, presented by Dr. Ronan Cummins, postdoctoral researcher with the Information Retrieval Group, Glasgow University.
Date: Monday 21st Feb
Time: 1-2 pm
Location: IT202 (IT Building)
Dr. Cummins will present a talk on Information retrieval, and modern web search, which have been vibrant research areas for several decades now. In this talk, he examines the limits of the information retrieval process from three different perspectives, and shows that there is much room for improvement in the area of query construction. He will outline preliminary research that develops a number of query performance predictors that may aid in automatically navigating the query spacePresented by Lourdes Beloqui Yuste and Dr. Hugh Melvin of the Performance Engineering Laboratory group, NUIG
'The transmission of multimedia traffic over IP Networks is increasing
year on year. This is particularly the case for IPTV systems in Europe. IPTV
differs from other TV delivery platforms by facilitating greater user
interactivity and customisation. IPTV provides an excellent delivery system for
different multimedia streams from multiple sources which can be integrated at
the receiver end, making it possible to display content in a synchronised mode.
Dr. Weiss has worked at the National Institute of Standards and Technology (NIST, formerly the National Bureau of Standards, NBS) in Boulder Colorado since 1978. He wrote the firmware for the NBS/GPS Time Transfer System for which he received the Applied Research Award of the NBS in 1983, along with the other principals. Dr. Weiss has been active in studying and developing time transfer systems especially using the Global Positioning System, for applications such as the generation of International Atomic Time. He also has led the NIST contract with the GPS program office for support of their clocks and timing systems.
In addition Dr. Weiss has specialized in new time scale algorithms and in synchronization in telecommunications systems. He has worked on problems with Relativity as they relate to GPS and to primary frequency standards. He has spear‑headed an annual Workshop on Synchronization in Telecommunications Systems (WSTS), which is now co‑sponsored each year by NIST, ATIS-OPTXS, the telecommunications synchronization standards committee, and by Telcordia. A sister conference, ITSF, was spawned from this, and held most recently in Dublin, November 2-4, 2010.
Marc Weiss received his B.S. degree from Valparaiso University, Valparaiso, Indiana in 1973. He received his M.S. degree in Mathematics in 1975, and his Ph.D. in Mathematical‑Physics in 1981, both from the University of Colorado, Boulder, Colorado.
Abstract:
This talk will discuss the problem of modeling and prediction of social network data over time, such as time-stamped emails or other communication events among a set of individuals. The talk will begin by motivating the problem of modeling such data, discussing for example the difference between discrete-time aggregated network representations and continuous-time event-based representations. We will review some of the basic strategies in building probabilistic models for such data, starting with models for static (non-temporal) data and moving to temporal models. In particular we will focus on hidden variable models which are emerging as a broadly applicable and flexible framework for network modeling. Relevant prior work in this area will be discussed as well as new ongoing work. We will also emphasize the importance of predictive evaluation in network modeling and discuss a number of issues that arise in this context. Experimental results will be presented comparing different modeling approaches using a variety of real-world event-based network data sets such as email networks.
Joint work with Arthur Asuncion, Chris DuBois, and Jimmy Foulds.
Biography:
Padhraic Smyth is a Professor in the Department of Computer Science and also serves as Director of the Center for Machine Learning and Intelligent Systems, both at the University of California, Irvine. He also has joint appointments in the Statistics and Biomedical Engineering Departments at UC Irvine. His research interests include machine learning, data mining, pattern recognition, and applied statistics. He was a recipient of best paper awards at the 2002 and 1997 ACM SIGKDD Conferences, received the NSF CAREER award in 1999, the ACM SIGKDD Innovation Award in 2009, and is a AAAI Fellow. He is co-author of Modeling the Internet and the Web: Probabilistic Methods and Algorithms (with Pierre Baldi and Paolo Frasconi in 2003), and was also co-author of Principles of Data Mining, MIT Press, August 2001, with David Hand and Heikki Mannila.
He received a first class honors degree in Electronic Engineering from University College Galway (National University of Ireland) in 1984, and the MSEE and PhD degrees from the Electrical Engineering Department at the California Institute of Technology in 1985 and 1988 respectively. From 1988 to 1996 he was a Technical Group Leader at the Jet Propulsion Laboratory, Pasadena, and has been on the faculty at UC Irvine since 1996. In addition to his academic research he is also active in industry consulting, working with companies such as Netflix (on the Netflix Prize), eBay, Oracle, Yahoo!, Nokia, and AT&T.
