User modelling for adapted accessible interaction Julio Abascal #, Olatz Arbelaitz *, Myriam Arrue #, Javier Muguerza * # EGOKITUZ: Laboratory of HCI for.

Slides:



Advertisements
Similar presentations
Andrea Maurino Web Service Design Methodology Batini, De Paoli, Maurino, Grega, Comerio WP2-WP3 Roma 24/11/2005.
Advertisements

Data Mining and the Web Susan Dumais Microsoft Research KDD97 Panel - Aug 17, 1997.
Integrating 3D Geodata in Service-Based Visualization Systems Jan Klimke, Dieter Hildebrandt, Benjamin Hagedorn, and Jürgen Döllner Computer Graphics Systems.
1 "Quality requirements for web-sites – recommendations for best practice for fulfilling user needs Best practice in communicating statistics: Making statistics.
Submissions November 2007 Stephen McCann, NSNSlide 1 IEEE 802 Emergency Services (ES) Call for Interest (CFI) Date: Stephen McCann
ASYCUDA Overview … a summary of the objectives of ASYCUDA implementation projects and features of the software for the Customs computer system.
Cultural Heritage in REGional NETworks REGNET Project Meeting Content Group Part 1: Usability Testing.
Improving Learning Object Description Mechanisms to Support an Integrated Framework for Ubiquitous Learning Scenarios María Felisa Verdejo Carlos Celorrio.
Designing Services for Grid-based Knowledge Discovery A. Congiusta, A. Pugliese, Domenico Talia, P. Trunfio DEIS University of Calabria ITALY
GMD German National Research Center for Information Technology Darmstadt University of Technology Perspectives and Priorities for Digital Libraries Research.
Introduction to HTML, XHTML, and CSS
Peer-to-peer and agent-based computing Peer-to-Peer Computing: Introduction.
Profiles Construction Eclipse ECESIS Project Construction of Complex UML Profiles UPM ETSI Telecomunicación Ciudad Universitaria s/n Madrid 28040,
Introduction Lesson 1 Microsoft Office 2010 and the Internet
Configuration management
Software change management
IONA Technologies Position Paper Constraints and Capabilities for Web Services
Distributed search for complex heterogeneous media Werner Bailer, José-Manuel López-Cobo, Guillermo Álvaro, Georg Thallinger Search Computing Workshop.
KEOD 2013 – 20 th September 2013 A Comprehensive Framework for Semantic Annotation of Web Content Manuel Fiorelli 1, Maria Teresa Pazienza 2, Armando Stellato.
Help communities share knowledge more effectively across the language barrier Automated Community Content Editing PorTal.
Chapter 22: Cloud Computing and Related Security Issues Guide to Computer Network Security.
Page16/2/2015 Sirlan Usage and usability considerations for SIRLAN solution success.
Project 1 Introduction to HTML.
WebMiningResearch ASurvey Web Mining Research: A Survey Raymond Kosala and Hendrik Blockeel ACM SIGKDD, July 2000 Presented by Shan Huang, 4/24/2007.
WebMiningResearchASurvey Web Mining Research: A Survey Raymond Kosala and Hendrik Blockeel ACM SIGKDD, July 2000 Presented by Shan Huang, 4/24/2007 Revised.
The W3C Web Accessibility Initiative (WAI) Inclusive learning through technology Damien French.
Building Knowledge-Driven DSS and Mining Data
1 An overview Always Best Connected Networks Dênio Mariz Igor Chaves Thiago Souto Aug, 2004.
Semantic Web and Web Mining: Networking with Industry and Academia İsmail Hakkı Toroslu IST EVENT 2006.
1st Project Introduction to HTML.
Community Manager A Dynamic Collaboration Solution on Heterogeneous Environment Hyeonsook Kim  2006 CUS. All rights reserved.
HTML 1 Introduction to HTML. 2 Objectives Describe the Internet and its associated key terms Describe the World Wide Web and its associated key terms.
Chapter ONE Introduction to HTML.
HTML Comprehensive Concepts and Techniques Intro Project Introduction to HTML.
FALL 2012 DSCI5240 Graduate Presentation By Xxxxxxx.
Tanenbaum & Van Steen, Distributed Systems: Principles and Paradigms, 2e, (c) 2007 Prentice-Hall, Inc. All rights reserved DISTRIBUTED SYSTEMS.
1. Human – the end-user of a program – the others in the organization Computer – the machine the program runs on – often split between clients & servers.
Chapter 1 Introduction to HTML, XHTML, and CSS
Competitive Differentiation: Using Technology to Deliver on Staples EASY Brand Promise Scott Floeck, Sr. Vice President, Staples, Inc.
Research paper: Web Mining Research: A survey SIGKDD Explorations, June Volume 2, Issue 1 Author: R. Kosala and H. Blockeel.
First things, First Do you belong in here? – 10 – 12 – Comp. Discovery or Keyboard/Comp Apps – Do you have any experience with Web Page Design?????
Active Monitoring in GRID environments using Mobile Agent technology Orazio Tomarchio Andrea Calvagna Dipartimento di Ingegneria Informatica e delle Telecomunicazioni.
1 Research Groups : KEEL: A Software Tool to Assess Evolutionary Algorithms for Data Mining Problems SCI 2 SMetrology and Models Intelligent.
Computer Measurement Group, India Optimal Design Principles for better Performance of Next generation Systems Balachandar Gurusamy,
HTML, XHTML, and CSS Sixth Edition Chapter 1 Introduction to HTML, XHTML, and CSS.
DC 2004 Metadata Generation and Accessibility Auditing Liddy Nevile La Trobe University, Australia Mail
PERVASIVE COMPUTING MIDDLEWARE BY SCHIELE, HANDTE, AND BECKER A Presentation by Nancy Shah.
Cloud platforms Lead to Open and Universal access for people with Disabilities and for All Cloud4all User Forum and Focus group 30/08/2012.
Page 1 WWRF Briefing WG2-br2 · Kellerer/Arbanowski · · 03/2005 · WWRF13, Korea Stefan Arbanowski, Olaf Droegehorn, Wolfgang.
Linking Tasks, Data, and Architecture Doug Nebert AR-09-01A May 2010.
A Context Model based on Ontological Languages: a Proposal for Information Visualization School of Informatics Castilla-La Mancha University Ramón Hervás.
FDT Foil no 1 On Methodology from Domain to System Descriptions by Rolv Bræk NTNU Workshop on Philosophy and Applicablitiy of Formal Languages Geneve 15.
NGCWE Expert Group EU-ESA Experts Group's vision Prof. Juan Quemada NGCWE Expert Group IST Call 5 Preparatory Workshop on CWEs 13th.
Cloud platforms Lead to Open and Universal access for people with Disabilities and for All Pofile initialisation and management Teleconference.
Module 10: Windows Firewall and Caching Fundamentals.
HTML Concepts and Techniques Fifth Edition Chapter 1 Introduction to HTML.
Chapter 1 Introduction to HTML, XHTML, and CSS HTML5 & CSS 7 th Edition.
11 ROUTING IP Chapter 3. Chapter 3: ROUTING IP2 CHAPTER INTRODUCTION  Understand the function of a router.  Understand the structure of a routing table.
6/28/ A global mesh of interconnected networks (internetworks) meets these human communication needs. Some of these interconnected networks are.
NCP training day ICT 23- Interfaces for accessibility Juan Pelegrin "Youth, Skills and Inclusion" DG CONNECT European Commission Luxembourg.
Chapter 1 Introduction to HTML
Automatic cLasification d
Securing the Network Perimeter with ISA 2004
Project 1 Introduction to HTML.
Chapter 18 MobileApp Design
Chapter 21: Cloud Computing and Related Security Issues
Chapter 22: Cloud Computing Technology and Security
CIS 375 Teaching Effectively-- snaptutorial.com
Web Mining Department of Computer Science and Engg.
Web Mining Research: A Survey
Presentation transcript:

User modelling for adapted accessible interaction Julio Abascal #, Olatz Arbelaitz *, Myriam Arrue #, Javier Muguerza * # EGOKITUZ: Laboratory of HCI for Special Needs * Algorithms, Data mining and Parallelism Research Team University of the Basque Country/Euskal Herriko Unibertsitatea

Rationale This paper briefly describes the diverse approaches to user adapted accessible interaction that we have developed in the last years Purpose: To discuss the possibility of advancing towards a comprehensive approach to shared-user modelling

Index 1.Introduction of EGOKITUZ 2.Objectives 3.Personal accessibility to the web 4.Accessibility to Ubiquitous Computing environments 5.Web mining for user modelling 6.Conclusions

EGOKITUZ: Laboratory of HCI for Special Needs Created in Main goal: the application of computer technology to provide support to people with disabilities and elderly people. Staff: Variable (currently 10 fulltime researchers). 4

EGOKITUZ: Laboratory of HCI for Special Needs International activities IFIP TC13 Human-Computer Interaction (1991-) IFIP WG 13.3 HCI and Disability (1993-) IFIP WG 13.1 Education in HCI and HCI Curriculum (1999-) EU Adviser, reviewer, evaluator, expert roles EU projects, COST European actions Research HCI & Assistive Technology Ambient Intelligence & Ubiquitous Computing Web Accessibility Teaching Advanced interaction systems Networks, OS & HW design Accessibility & Usability 5

Index 1.Introduction of EGOKITUZ 2.Objectives 3.Personal accessibility to the web 4.Accessibility to Ubiquitous Computing environments 5.Web mining for user modelling 6.Conclusions

Objectives Enhancing the accessibility for people with temporary or permanent restrictions. Adapting the interaction system to – the features, needs and likes of each specific user, and – the characteristic of each place and task. By compiling information about the users and their environment to create suitable user models And dynamically creating personalized interfaces. Future: Sharing or exporting models among remote applications

Index 1.Introduction of EGOKITUZ 2.Objectives 3.Personal accessibility to the web 4.Accessibility to Ubiquitous Computing environments 5.Web mining for user modelling 6.Conclusions

Universal Accessibility to the web The problem of Web accessibility is mainly treated from the Design for All or Universal Accessibility point of view. This approach is extremely convenient for ensuring the civil rights to electronic inclusion of people with any type of disability. Many methodologies and tools have been created to apply these guidelines. This approach failed to help specific users to find suitable web sites

EvalAccess: Automatic Web Accessibility Evaluator A result of the IRIS European Project: Built as a web-service to be used from mainstream applications. Not built-in Guidelines: able to evaluate diverse sets of guidelines. Tool to allow the creation of machine-readable new guidelines: specific purpose guidelines. It provides statistical data to create quantitative accessibility metrics.

Personal vs. Universal Accessibility Evalaccess allowed us to tackle Personal Accessibility: Starting from a combination of – Quantitative metrics and – The use of specific guidelines or WAI subsets In order to select the most adequate guidelines users where modelled. – Abilities and restrictions to access the Web

Index 1.Introduction of EGOKITUZ 2.Objectives 3.Personal accessibility to the web 4.Accessibility to Ubiquitous Computing environments 5.Web mining for user modelling 6.Conclusions

EGOKI INREDIS Project INterfaces for RElations between Environment and people with DISabilities – Consortium: 14 companies, 18 research institutions. – Period: 2007 to – Investment 23.6 million. – Purpose: to develop basic accessible and interoperable technologies that enable the communication and interaction between people with disabilities and their environment. – Some work-packages: Interoperability protocols. Assistive technology and ubiquitous software. Adaptive user interfaces and device configuration. Interoperability in mobile devices. – INREDIS project inspired us to create EGOKI 14

Scenario: Interacting with Ubiquitous Computing Environments 15 Middleware 1. The user device and the target machine somehow transparently communicate (through a wireless network) 2. The ATM service is offered to the user. He/She accepts it (using his/her mobile personal device) 3. The system creates (and downloads to the user device) an instance of the UI adapted to the user/device/context

Automatic generation of accessible User Interfaces EGOKI: Automatic generation of adapted UIs for ubiquitous computing For users with restrictions: people with disabilities, elderly people. people performing other activities (driving) or using special devices (mobiles). Goal: to provide ubiquitous access to intelligent machines (ATMs, information kiosks, intelligent home appliances, etc.). Context: users are provided with their own device adapted to their features and needs. 16

Automatic generation of accessible User Interfaces Service designers provide abstract specifications of the UI for each service by means of a User Interface Modelling Language (UIML) The system maintains user/task/context models (in an ontology) EGOKI selects the most suitable interaction resources (from those supplied by the service provider) taking into account the users capability for each communication modality. It creates an accessible adapted UI, which is suited to the service. 17

Automatic generation of accessible User Interfaces Case Study: Underground Ticketing 18

Validation In order to to prove the correct functionality and the accessibility of interfaces that the EGOKI generated automatically, it was carried out: – Barrier Walkthrough exercise – User Based Testing: Blind users and Users with cognitive disabilities

Index 1.Introduction of EGOKITUZ 2.Objectives 3.Personal accessibility to the web 4.Accessibility to Ubiquitous Computing environments 5.Web mining for user modelling 6.Conclusions

User Modelling based on Web Usage Mining 1.Data acquisition and pre-processing – Complex (the most time consuming and computationally expensive step) – Users privacy issue. – Diverse possible sources (client machines, proxies, servers...) – It includes: user and session identification, and data fusion and cleaning. 2.Pattern discovery and pattern analysis. – Machine learning techniques are applied in order to find sets of web users with common web-related characteristics and the corresponding patterns. – Paradigms : unsupervised learning (or clustering), association rules, and paradigms used to find frequent patterns such as frequent episodes. – Subsequently: selection of the most meaningful patterns. manually by experts in the area or based on the parameters of the machine learning algorithms used

ModelAcces Project Currently we are working on profiling functional abilities of the users, using data extracted from their web interaction. Logs from a large website DISCAPNET run by ONCE. Some variables automatically extracted from the server log data, can have direct connections with t user abilities: – number of different URLs visited – average time spent on each URL (taking into account if the page is of authority type or hub type) – maximum and/or average depth of each session – diversity in semantic content of the visited URLs – etc. We use these types of parameters to make assumptions about the possible limitations of the users (specific disabilities, how lost they are, etc.). The results will be used to enrich the recommendations generated using other strategies.

Conclusion Proliferation of adaptive applications (each one handling and maintaining its own model). But Public Ubiquitous Computing environments do not have a model of each user. Is it possible to share models among applications? Development of methods to… – …(partly or completely) share models. – …provide remote access to private models. – …define formats for interoperable model description. – …protect user privacy. – …adopt Virtual User Modelling [VUMS White Paper].

EGOKITUZ: Laboratory of HCI for Special Needs Location University of the Basque Country/ Euskal Herriko Unibertsitatea Donostia. Spain 24