Nikolaos Kaklanis - PhD presentation

905 views

Published on

Nikolaos Kaklanis - PhD presentation

Published in: Technology, Education
1 Comment
1 Like
Statistics
Notes
No Downloads
Views
Total views
905
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
20
Comments
1
Likes
1
Embeds 0
No embeds

No notes for slide
  • Γενικά μπορούμε να πούμε ότι, η συναισθηματική υπολογιστική απαρτίζεται από δύο τομείς, Ο ένας αφορά διεπαφές ΑΑΣ, ενώ ο άλλος αφορά αναπροσαρμοζόμενες γραφικές διεπαφές χρήστη. Η διατριβή ασχολήθηκε και με τους δύο αυτούς τομείς, εστιάζοντας την έρευνά της στις διεπαφές αυτόματης αναγνώρισης συναισθημάτων. Η βασική συνεισφορά της έγκειται πρώτα απ’όλα στο ότι… Για την ανάλυση όλων το προτεινόμενων χαρακτηριστικών εκτελέστηκαν τρία διαφορετικά πειράματα. Στα πλαίσια της διατριβής διεξήχθησαν τρία διαφορετικά πειράματα που θα περιγραφούν στη συνέχεια. Η συναισθηματική υπολογιστική μπορεί γενικά να περιγραφεί από αυτό το σχήμα. Αποτελείται
  • Γενικά μπορούμε να πούμε ότι, η συναισθηματική υπολογιστική απαρτίζεται από δύο τομείς, Ο ένας αφορά διεπαφές ΑΑΣ, ενώ ο άλλος αφορά αναπροσαρμοζόμενες γραφικές διεπαφές χρήστη. Η διατριβή ασχολήθηκε και με τους δύο αυτούς τομείς, εστιάζοντας την έρευνά της στις διεπαφές αυτόματης αναγνώρισης συναισθημάτων. Η βασική συνεισφορά της έγκειται πρώτα απ’όλα στο ότι… Για την ανάλυση όλων το προτεινόμενων χαρακτηριστικών εκτελέστηκαν τρία διαφορετικά πειράματα. Στα πλαίσια της διατριβής διεξήχθησαν τρία διαφορετικά πειράματα που θα περιγραφούν στη συνέχεια. Η συναισθηματική υπολογιστική μπορεί γενικά να περιγραφεί από αυτό το σχήμα. Αποτελείται
  • Nikolaos Kaklanis - PhD presentation

    1. 1. University of SurreyDepartment of ComputingA COUPLED USER AND TASK MODELLINGMETHODOLOGY FOR ACCESSIBLEPRODUCT DESIGNPhD ThesisNikolaos KaklanisInformation & Communication Systems Engineer, MSc
    2. 2. University of SurreyDepartment of Computing• Disability is part of the human condition– Almost everyone will be temporarily or permanently impairedat some point in life.• Aging is strongly connected with difficulties infunctioning• The lowest estimate, based on the currently defineddisablement categories, estimates their total number ataround 74 Million persons, in the European Union.• Many products & services are inaccessible for peoplewith functional limitationsThe need: Accessible & ergonomic products andservices for the elderly and disabledTHE NEEDTHE NEED
    3. 3. University of SurreyDepartment of ComputingTYPICAL DEVELOPMENT CHAINTYPICAL DEVELOPMENT CHAIN
    4. 4. University of SurreyDepartment of ComputingDEVELOPMENT CHAIN USINGDEVELOPMENT CHAIN USING VIRTUAL USER MODELSVIRTUAL USER MODELSTheContext of the introduced research
    5. 5. University of SurreyDepartment of ComputingA new coupled user-task modeling methodologyis needed, which will enable– the detailed description of users with disabilities– the development of Virtual User Models (VUMs)representing specific population groupsPROBLEM IDENTIFICATIONPROBLEM IDENTIFICATION
    6. 6. University of SurreyDepartment of Computing• A novel Virtual User Model (VUM) enabling the detailed descriptionof the physical, cognitive & behavioural user characteristics with specialfocus on the elderly and people with functional limitations.• An innovative methodology (based on statistical analysis) for thedevelopment of Virtual User Models that represent specific disabledpopulation groups.• A framework for the adoption of virtual user models withinaccessibility simulation and ergonomy assessment processes of virtualdesigns.• Design and development of tools for supporting and putting theaforementioned methodologies and frameworks into practice.RESEARCH CONTRIBUTIONRESEARCH CONTRIBUTION
    7. 7. University of SurreyDepartment of ComputingSTATE OF THE ART IN VIRTUAL USER MODELINGSTATE OF THE ART IN VIRTUAL USER MODELING• Ontology-based models– OntobUM (Razmerita et al., 2003)– GUMO (Heckmann, 2006)– VICON User Model (Pierre et al., 2011)• XML-based models– UserML (Heckmann, 2003)– GUIDE User Model (Biswas & Langdon, 2012)– User Models using relational databases– MyUI User Model (Peissner et al. 2012)• Personas (Cooper, 1999; Goodwin, 2002; Nielsen, 2002 & 2003; Pruitt & Grudin, 2003)• Task modeling approaches• ConcurTaskTree (Paternò, 1999)• Hierarchical Task Analysis (HTA) (Annett & Duncan, 1967)• UsiXML (Vanderdonckt, 2005)
    8. 8. University of SurreyDepartment of ComputingLIMITATIONS OF EXISTING APPROACHESLIMITATIONS OF EXISTING APPROACHESExisting approach/methodology LimitationsPersonas Not expressed in a machine-readable formatFocused on specific body partsCannot represent in detail physical, cognitive &behavioural characteristicsVUMs in machine-readableformatDisabilities not supportedFocus on the progression of the disabilityUser modelling methodologiesCannot create VUMs representing specificpopulation groupsUser is defined independently from tasksUser & task modellingmethodologiesIt’s not clear which tasks are affected by thedisabilities
    9. 9. University of SurreyDepartment of Computing• Introduction•An innovative usermodelingmethodology focusedon the elderly anddisabled• A methodology forcreating virtual usermodels representingspecific populationgroups• The proposed usermodeling methodologyin practice• ConclusionsAn innovative user modelingmethodology focused on theelderly and disabled
    10. 10. University of SurreyDepartment of ComputingThe proposed framework is based on the following sevenmajor building blocks:• Abstract User Models– high level description of potential user models• Generic Virtual User Models– a Generic Virtual User Model (GVUM) describes a set of users having aspecific set of disabilities• Instance of a Generic Virtual User Model– an instance of a virtual user (e.g., Persona).AN INNOVATIVE USER MODELING METHODOLOGY FOCUSED ON THE ELDERLY & DISABLEDAN INNOVATIVE USER MODELING METHODOLOGY FOCUSED ON THE ELDERLY & DISABLED
    11. 11. University of SurreyDepartment of Computing• Primitive Tasks– primitive human actions• Task Models– actions that are being systematically performed in the context of the virtualprototype to be tested• Multimodal Interaction Models– alternative ways of a primitive task’s execution• alternative modalities• assistive devices• Simulation Models– simulation scenarioAN INNOVATIVE USER MODELING METHODOLOGY FOCUSED ON THE ELDERLY & DISABLEDAN INNOVATIVE USER MODELING METHODOLOGY FOCUSED ON THE ELDERLY & DISABLED
    12. 12. University of SurreyDepartment of ComputingARCHITECTURE OF THE PROPOSED FRAMEWORKARCHITECTURE OF THE PROPOSED FRAMEWORK
    13. 13. University of SurreyDepartment of Computing• Primitive tasks– define primitive human actions– are related to the disabilitycategory• limited but sufficient numberof primitive tasksPRIMITIVE TASKSPRIMITIVE TASKSPrimitive task’scategoryPrimitivetaskMotor PushMotor GraspCognitive SelectCognitive Read
    14. 14. University of SurreyDepartment of Computing• User tasks are divided in two categories:– a) primitive (e.g., grasp, pull, walk, etc.)– b) complex (e.g., driving, telephone use, computer use, etc.)• For each complex task, a Task Model is developed– to specify how the complex task can be analyzed intoprimitive tasks.TASK MODELSTASK MODELSCOMPLEX TASK PRIMITIVE TASK BODY PART OBJECTCLOSE CAR DOOR WHILESEATEDREACH ARM DOORGRASP HAND INTERIOR DOORHANDLEPULL HAND INTERIOR DOORHANDLEPUSH HAND LOCK BUTTON
    15. 15. University of SurreyDepartment of Computing• UsiXML’s taskModelTASK MODELS - IMPLEMENTATIONTASK MODELS - IMPLEMENTATION<?xml version="1.0" encoding="UTF-8"?><taskmodel><task id="st0task0" name="Close_car_door_while_seated" type="abstraction"><task id="st0task1" name="Reach_door_with_arm" type="interaction"/><task id="st0task2" name="Grasp_interior_door_handle_with_hand" type="interaction"/><task id="st0task3" name="Pull_interior_door_handle_with_hand" type="interaction"/><task id="st0task4" name="Push_lock_button_with_hand" type="interaction"/></task>……</taskmodel>UsiXML source code
    16. 16. University of SurreyDepartment of Computing• Alternative ways of a primitive task’s execution with respect to• the different target user groups• the replacement modalities• the possible use of assistive devices.MULTIMODAL INTERACTION MODELSMULTIMODAL INTERACTION MODELSTask BodypartModality TaskobjectDisability Alternativetask(s)Body part AlternativemodalityAlternativetask object/assistivedevicePull Hand Motor InteriordoorhandleUpper limbimpairedSpeak Mouth Voice control VoiceactivateddoorsPush Hand/ElbowMotor Button thatcloses door
    17. 17. University of SurreyDepartment of Computing• UsiXML’s taskModelMULTIMODAL INTERACTION MODELS - IMPLEMENTATIONMULTIMODAL INTERACTION MODELS - IMPLEMENTATION<?xml version="1.0" encoding="UTF-8"?><taskmodel><task id="st0task0" name="Pull_interior_door_handle" type="abstraction"><task id="st0task1"name="Pull(modality:motor)(means:hand)(object:interior_door_handle)" type="interaction"/><task id="st0task2"name="Speak(modality:voice)(means:mouth)(object:voice_activated_doors)" type="interaction"/><task id="st0task3"name="Push(modality:motor)(means:hand,elbow)(object:button_that_closes_door)" type="interaction"/></task><deterministicChoice><source sourceId="st0task1"/><target targetId="st0task2"/></deterministicChoice><deterministicChoice><source sourceId="st0task2"/><target targetId="st0task3"/></deterministicChoice></taskmodel>UsiXML source code
    18. 18. University of SurreyDepartment of Computing• A high level descriptionof potential user modelsABSTRACT USER MODELSABSTRACT USER MODELSDisabilitycategoryDisabilityShortdescriptionQuantitative disability metricsFunctionallimitations(ICFClassification)Age-relatedMotor Spinalcordinjuries(Thoracicinjuries)Spinal cordinjuriescausemyelopathyor damage tonerve rootsormyelinatedfiber tractsthat carrysignals toand from thebrain.Temporal gait variables:-Gait Cycle (sec):2.17 (1.05)-Cadence (steps/min): 65.0 (23.1)-Double support (%): 42.8 (10.2)-Stride (m): 0.48 (0.13)-Velocity ((m/sec)/height): 0.27 (0.13)S120 Spinalcord andrelatedstructures,S1200Structure ofspinal cord,S12000Cervical spinalcord,s12001Thoracic spinalcord,s12002Lumbosacralspinal cord,s12008Structure ofspinal cord,other specified,s12009Structure ofspinal cordunspecified,CouldbeKinematic variables:-Hip excursion (˚): 39.3 (9.0)-Knee excursion (˚): 38.1 (13.2)-Ankle excursion (˚): 25.0 (4.9)-Hip velocity (˚/sec): 38.2 (17.5)-Knee velocity (flexion) (˚/sec): 64.1 (41.8)-Knee velocity (extension) (˚/sec): 83.8 (54.2)-Ankle velocity (˚/sec): 48.1 (30.8)
    19. 19. University of SurreyDepartment of Computing• Refers to a class of virtual users exhibiting one or morespecific disabilities• Describes the tasks affected by the disabilities and theirassociated disability-related parameters.GENERIC VIRTUAL USER MODELSGENERIC VIRTUAL USER MODELSDisabilitycategoryDisability Affected primitivetasksAffected primitive tasks’ parametersGrasp The user is able to grasp objects, withsize <= 3cm x 3cm x 3cmPull The user can pull an object with max_Force: 5NGait velocity ranges from 0.18 to 1.03 m/secMotor HemiplegiaWalkAbnormal step rhythm
    20. 20. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – SUPPORTED CHARACTERISTICSGENERIC VIRTUAL USER MODELS – SUPPORTED CHARACTERISTICS
    21. 21. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS - IMPLEMENTATIONGENERIC VIRTUAL USER MODELS - IMPLEMENTATION
    22. 22. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (BASIC CONTAINERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (BASIC CONTAINERS)
    23. 23. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (HEAD ELEMENT)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (HEAD ELEMENT)
    24. 24. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (DISABILITY MODEL)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (DISABILITY MODEL)
    25. 25. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (GENERAL PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (GENERAL PARAMETERS)
    26. 26. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (GENERAL PREFERENCES)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (GENERAL PREFERENCES)
    27. 27. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (ANTHROPOMETRIC PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (ANTHROPOMETRIC PARAMETERS)
    28. 28. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (UPPER LIMB PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (UPPER LIMB PARAMETERS)
    29. 29. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (LOWER LIMB PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (LOWER LIMB PARAMETERS)
    30. 30. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (NECK PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (NECK PARAMETERS)
    31. 31. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (SPINAL COLUMN PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (SPINAL COLUMN PARAMETERS)
    32. 32. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (GAIT PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (GAIT PARAMETERS)
    33. 33. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (VISUAL PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (VISUAL PARAMETERS)
    34. 34. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (HEARING PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (HEARING PARAMETERS)
    35. 35. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (SPEECH PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (SPEECH PARAMETERS)
    36. 36. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (COGNITIVE PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (COGNITIVE PARAMETERS)
    37. 37. University of SurreyDepartment of ComputingGENERIC VIRTUAL USER MODELS – IMPLEMENTATION (BEHAVIOURAL PARAMETERS)GENERIC VIRTUAL USER MODELS – IMPLEMENTATION (BEHAVIOURAL PARAMETERS)
    38. 38. University of SurreyDepartment of ComputingINSTANCES OF GENERIC VIRTUAL USER MODELSINSTANCES OF GENERIC VIRTUAL USER MODELS• A specific virtual user with specific disability related parameters including:– disabilities– affected primitive tasksUser ID DisabilitycategoryDisability Affected primitivetasksAffected primitive tasks’ parametersGrasp The user is able to grasp objects, withsize <= 2.5cm x 2.5cm x 2.5cmPull The user can pull an object withmax_Force: 3NGait velocity : 0.9 m/sUser 1 Motor HemiplegiaWalkAbnormal step rhythm
    39. 39. University of SurreyDepartment of ComputingINSTANCES OF GENERIC VIRTUAL USER MODELS - IMPLEMENTATIONINSTANCES OF GENERIC VIRTUAL USER MODELS - IMPLEMENTATION<?xml version="1.0" encoding="UTF-8"?><userModel xmlns="http://www.usixml.org" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.usixml.org/spec/UsiXML-ui_model.xsd"id="User_Model" name="Adam Brown" creationDate="2013/02/0220:31:06" schemaVersion="1.8.0"><head><version modifDate="2010/10/17 20:31:06">0.0</version><authorName>Automatically generated byethe User Model Generator</authorName><comment>This model has been generated using the User Model Generator</comment></head><disabilityModel><disability name="Spinal cord injuries" ageRelated="true" typesICF="b147, b760,b7603, b780"><disabilityDetails>Spinal cord injuries cause myelopathy or damage tonerve roots or myelinated fiber tracts that carry signals to and fromthe brain. Depending on its classification and severity, this type oftraumatic injury could also damage the grey matter in the central partof the cord, causing segmental losses of interneurons and motorneurons.</disabilityDetails><affectedTasks><affectedTask idTask="walking_ID" type="motor" name="walking"details="inability to effectively transfer weightbetweenlegs, abnormal step rhythm, excessive plantar flexion during swingphase, falling during activities" /><affectedTask idTask="driving_ID" type="motor"name="driving"details="automatic transmission and buttons on thesteering wheelinstead of pedals for brake and accelerator needed" /></affectedTasks></disability></disabilityModel><capabilityModel><general><gender>male</gender><ageGroup>30-49</ageGroup></general><generalPreferences /><anthropometric><weight measureUnits="kgr" value="81.023056"/><stature measureUnits="cm" value="174.586060"/><headLength measureUnits="cm" value="19.533623"/>…GVUM: range of valuesInstance of a GVUM: absolute valueswhile…
    40. 40. University of SurreyDepartment of ComputingCOMPARISON BETWEEN THE PROPOSED VUM & EXISTING MODELSCOMPARISON BETWEEN THE PROPOSED VUM & EXISTING MODELSNumber of variablesof the proposed VUMNumber of variablesof the GUIDE VUMNumber of variablesof the MyUI VUMNumber of variablesof the VICON VUM• total: over 250 • total: 27• included in theproposed VUM: 21• total: 30• included in theproposed VUM: 10• total: 33• included in theproposed VUM: 11Why some variables of the others VUMs do not exist in the proposedVUM?• they describe qualitative (not quantitative) characteristics and, thus,cannot be directly simulated (e.g., ability to move the hand precisely)• describe the result of the simulation process (e.g., number of successfulinteractions with the system)• are not actually user characteristics (e.g., amount of ambient light at theusers place)
    41. 41. University of SurreyDepartment of Computing• The scenario to be followed during the simulation process• A Simulation Model may include three different types of tasks:– a) abstract tasks– b) interaction tasks– c) application tasksSIMULATION MODELSSIMULATION MODELSScenario Main tasks SubtasksPull handbrakeUse handbrakeRelease handbrakeOpen storagecompartmentAutomotive simulation:assess the accessibility of thehandbrake and the storagecompartment Use storagecompartmentClose storagecompartment
    42. 42. University of SurreyDepartment of Computing• UsiXML’s taskModelSIMULATION MODELS - IMPLEMENTATIONSIMULATION MODELS - IMPLEMENTATION
    43. 43. University of SurreyDepartment of Computing• Introduction•An innovative usermodelingmethodology focusedon the elderly anddisabled• A methodology forcreating virtual usermodels representingspecific populationgroups• The proposed usermodeling methodologyin practice• Conclusions• The proposed methodology offers• detailed definition of• user (including disabilities)• user tasks• simulation scenarios• Mapping/linkage between all thesedefinitions
    44. 44. University of SurreyDepartment of Computing• Introduction•An innovative usermodeling methodologyfocused on the elderlyand disabled• A methodology forcreating virtual usermodels representingspecific populationgroups• The proposed usermodeling methodologyin practice• ConclusionsA methodology for creatingvirtual user modelsrepresenting specificpopulation groups
    45. 45. University of SurreyDepartment of Computing1. Selection of disability parameters– trying to model the most characteristic functional limitations (Abstract UserModel) caused by each disability• e.g., disability parameters for Gonarthrosis (osteoarthritis of the knee) include: kneeflexion/extension and gait parameters, like cadence, step length, etc.2. Acquire raw data (real measurements)3. Estimate the probability density function (pdf) of each disabilityparameter• Parametric regression• Non-parametric regression• An innovative hybrid regression method able to handle small sample sizes• All the above regression methods are applied and the one with the smallest MeanSquared Error (MSE) is chosen as the most suitable4. Validation and optimization of the regression model– Bootstrap method (Adèr et al., 2008)MODELING A DISABILITY – GENERAL STEPSMODELING A DISABILITY – GENERAL STEPS
    46. 46. University of SurreyDepartment of ComputingSTEP 2: MEASUREMENTS CONDUCTEDSTEP 2: MEASUREMENTS CONDUCTED• Measurements were obtained from total 164 subjects• Parkinson’s disease (17 subjects)• Stroke (36 subjects)• Multiple Sclerosis (20 subjects)• Cerebral Palsy (17 subjects)• Coxarthrosis (7 subjects)• Gonarthrosis (5 subjects)• Lower Limb Amputation (3 subjects)• Elderly people (59 subjects)
    47. 47. University of SurreyDepartment of ComputingSTEP3: PARAMETRIC REGRESSIONSTEP3: PARAMETRIC REGRESSION222)(221)( σµπσ−−=xexfDistribution Type PDF ExampleGaussian Hip flexion ofthe elderlyDistribution Type PDF ExampleGaussian mixture Fingers’ extensionof people withstrokeDistributionTypePDF ExampleExponential Gait cycleof strokepatientsDistributionTypePDF ExampleLévy Double supportof strokepatientsDistributionTypePDF ExampleUniform Gait velocity ofpatients withcerebral palsy∑=−−⋅=nixiiiiewxf12)(22221)( σµπσθθxkexkkxf−−Γ= 1)()(1)(23)(2)(2)(µπµ−=−−xecxfxc∈−=elsewherebaxabxf,0],[,1)(
    48. 48. University of SurreyDepartment of Computing• PDF is unknownSTEP3: NON-PARAMETRIC REGRESSIONSTEP3: NON-PARAMETRIC REGRESSIONKernel regressionKnee angular velocity of the stroke patientsLocal polynomial multiple regressionStep width of the elderly•Seems to follow a Lévy distribution.•This can be proved by examining the MSE resultingfrom parametric regression:•Lévy: MSE: 0.00204•Gaussian: MSE: 0.00256•Gaussian mixture: MSE: 0.00290•Uniform: MSE: 0.00367•Exponential: MSE: 0.07297•Seems to follow a Gaussian distribution.•This can be proved by examining the MSE resultingfrom parametric regression:•Gaussian: MSE: 0.94464•Gaussian mixture: MSE: 1.02365•Uniform: MSE: 1.26263•Exponential: MSE: 1.29327•Lévy: MSE: 2.69961
    49. 49. University of SurreyDepartment of Computing• Sample size is important!• BUT…the task of taking real measurements from disabled people is– very difficult– needs much resources• Estimating the PDF of a disability parameter using a limited set ofsubjects could result in great variance.We introduce the concept of hybrid regression that exploits– information found on the literature– real measurements– virtual measurementsSTEP3: HYBRID REGRESSIONSTEP3: HYBRID REGRESSIONSolution
    50. 50. University of SurreyDepartment of Computing• The probability distribution followed by a disability parameter isdefined using an ε-contaminated class of priors, namely Γ.– The contaminated class controls the balance of the determinatepriors, , and other possible priors, , through an ε-level of weight.• The ε-contaminated class is formulated as:– where D is our initial data set and– ε denotes the level of uncertaintyHYBRID REGRESSION –HYBRID REGRESSION – Ε-Ε-CONTAMINATED CLASS OF PRIORSCONTAMINATED CLASS OF PRIORSs0πParametric regressionon the original sampleUser definedWe tested the following values:0, 0.01, 0.05, 0.10, 0.20Gaussian
    51. 51. University of SurreyDepartment of Computing• Let variable Y represent the real measurements• Assumption: Y is be the response of the simple linear regressionmodel– where is the predictor value– is the intercept,– is the slope and– are independent and identically distributed random variables fromHYBRID REGRESSION – LINEARITY ASSUMPTIONHYBRID REGRESSION – LINEARITY ASSUMPTIONiii XY εββο ++= 1iX0β1β
    52. 52. University of SurreyDepartment of Computing• Our initial data set consists of measurements , where:– each is a real measurement– ‘s are calculated as follows:• from Gaussian pdf found in the literature– to keep a relevance between ‘s and ‘s, we get random samples ( ‘s) that meet• When there is no available information in the literature, we calculate the ‘s usingthe following formula:With this process, ‘s are actually coming from a random movement (within aspecific range) of the initial samples ‘s.HYBRID REGRESSION – INITIAL DATA SETHYBRID REGRESSION – INITIAL DATA SET),( 2litlitN σµiY iX iX410−≤−−− YYX litii µiXkYkYYXii⋅⋅−=iXiYRandom valuewithin [0.95, 1.05]
    53. 53. University of SurreyDepartment of Computing• A virtual data set consisting of m pairs is constructedas follows:– ‘s are random samples from the pdf of ‘s.– ‘s are random samples of , assuming that is a Gaussiandistribution withandwhere andHYBRID REGRESSION – VIRTUAL SAMPLESHYBRID REGRESSION – VIRTUAL SAMPLESADAjX iXAjY Γ 0πAAYXYA0100 ββµ +== ∑=−−==niiiYxynDMSEA1201002)(1)( ββσ∑ ∑∑∑∑= ====−−=niniiiniiniiniiiXnXYXnYX12121110111β XY 0100 ββ −=),( AjAj XYUser-selected valueWe tested the following values:0, 0.5n, n, 2n, 3n
    54. 54. University of SurreyDepartment of Computing• At this stage, we have available:– The initial data set– A virtual data set• Both data sets are used for the fine tuning of the parameters of, which describes the distribution of disability parameter .• Now, for the definition of , we perform Gaussian parametricregression on the real measurements (Yi’s).HYBRID REGRESSION – FINE TUNING OF PARAMETERS OFHYBRID REGRESSION – FINE TUNING OF PARAMETERS OF ΓΓADDΓ y0πThis is what we aresearching for…
    55. 55. University of SurreyDepartment of Computing• The maximum likelihood estimate taken into account both theoriginal and the virtual sample sets ( and , correspondingly)is given by:• Using the EM algorithm for parameter estimation concludes afterk iterations with and , which are used for the optimisationof the Gaussian distribution .– More specifically,• now becomes• and becomesHYBRID REGRESSION – EM ALGORITHMHYBRID REGRESSION – EM ALGORITHMD AD∏∏++===mnniAiniiMLE DyDy11, )|()|(maxarg ππββπk0β k1β0πAAYXYA0100 ββµ +== AkkAYXYA 10 ββµ +==∑=−−==niiiYxynDMSEA1201002)(1)( ββσ ∑=−−==niikkiYxynDMSEA12102)(1)( ββσ
    56. 56. University of SurreyDepartment of Computing• The combination of– ε (uncertainty) and– m (number of virtual samples)that results to the minimum MSE is used to formthe final Γ.HUBRID REGRESSION – RESULTEDHUBRID REGRESSION – RESULTED ΓΓ
    57. 57. University of SurreyDepartment of ComputingHYBRID REGRESSION EXAMPLE - STEP WIDTH OF THE ELDERLYHYBRID REGRESSION EXAMPLE - STEP WIDTH OF THE ELDERLYPurple line: data taken from the MSPGreen line: virtual samples generated considering the MSP data.Yellow line: virtual samples generated considering the data coming from the literatureLight blue line: pdf reported in the literatureDark blue line: represents all the samples (both real and virtual)Red line: final estimated pdfRegression type MSEParametric – Gaussian 0.74377Parametric – Lévy 2.73860Parametric – Gaussianmixture0.91292Parametric – Exponential 1.46577Parametric – Uniform 1.13034Nonparametric – Kernel 0.5102Nonparametric –Polynomial0.6414Hybrid 0.00642
    58. 58. University of SurreyDepartment of Computing• The definition of , enables the calculation ofdisability parameter values for different populationgroups.– e.g. , to find the value of wrist flexion that corresponds to the90% of people with arthritis, we have:CALCULATING THE DISABILITY PARAMETER VALUES FROM THE ESTIMATED PDFCALCULATING THE DISABILITY PARAMETER VALUES FROM THE ESTIMATED PDF∫ −=bajijiijijji aFbFdxxf )()()( ,,,jiF ,9.0)0()( ,, =− onwristFlexiarthritisonwristFlexiarthritis FkFPDF of parameter j ofdisability iFinite integral:The value of wrist flexioncorresponding to the 90% ofpeople with arthritis
    59. 59. University of SurreyDepartment of ComputingREGRESSION MODELS - INDICATIVE RESULTS (STROKE PATIENTS)REGRESSION MODELS - INDICATIVE RESULTS (STROKE PATIENTS)ParameterNameBefore validation & optimisation After optimisationRegressionTypeDistribution TypePDFParametersDistribution TypePDFParametersGait Cycle ParametricLevyμ = 0,918c = 0,29893Gaussianμ = 1,81818σ = 0,609692PushForceHybridΠ0(Gaussian):μ=65,47σ=67,54q(Exponential):λ=0,02ε = 0,05Gaussianμ = 78,5959σ = 61,9811HipExtensionRightParametricGaussian Mixtureμ1=32,81111σ1=5,10935w1=0,34615μ2=15,82353σ2=4,21936w2=0,65385Gaussian Mixtureμ1 = 32,597σ1 = 5,682615w1 = 0,343074μ2 = 15,86191σ2 = 4,316446w2 = 0,656926
    60. 60. University of SurreyDepartment of Computing• Introduction•An innovative usermodelingmethodology focusedon the elderly anddisabled• A methodology forcreating virtual usermodels representingspecific populationgroups• The proposed usermodelingmethodology inpractice• ConclusionsThe proposed user modelingmethodology in practice
    61. 61. University of SurreyDepartment of Computing• VERITAS simulationplatform• Automatic simulatedaccessibility andergonomy assessmentof designs• Its development wasbased on the proposeduser modelingmethodologyACCESSIBILITY AND ERGONOMY EVALUATION OF DESIGNSACCESSIBILITY AND ERGONOMY EVALUATION OF DESIGNSPhysical characteristics Normal values Elderly (60 to 84) Rheumatoid arthritisHand maximum pull force (N) 335 76.8Wrist extension (°) 0 - 60Wrist radial deviation (°) 0 - 27.5 0 – 19Wrist ulnar deviation (°) 0 – 35 0 – 26Forearm supination (°) 0 – 85 0 – 74Forearm pronation (°) 0 – 85 0 – 71Elbow flexion (°) 0 – 142.5Elbow hyper-extension (°) 0 – 10 0 – 4Shoulder flexion (°) 0 – 160 0 – 10Shoulder abduction (°) 0 – 85 0 – 67 0 – 15Shoulder internal rotation (°) 0 – 80 0 – 63Shoulder external rotation (°) 0 – 45 0 - 15Spinal column flexion (°) 0 - 90 0 – 23.6Spinal column extension (°) 0 – 30 0 - 17Spinal column left lateral flexion (°) 0 – 25 0 – 19Spinal column right lateral flexion (°) 0 – 25 0 - 20
    62. 62. University of SurreyDepartment of ComputingACCESSIBILITY AND ERGONOMY EVALUATION OF DESIGNS – STAPLER USEACCESSIBILITY AND ERGONOMY EVALUATION OF DESIGNS – STAPLER USEStapler torque-resistance (Nm) Normal Elderly Rheumatoid arthritis2.5 Pass Pass Pass5.0 Pass Pass Pass10 Pass Pass Pass15 Pass Fail Pass20 Pass Fail Fail25 Pass Fail Fail30 Fail Fail Fail
    63. 63. University of SurreyDepartment of Computing• DIAS tool: Impairmentssimulator• It has been extended tosupport the proposed VUMs.IMPAIRMENT SIMULATION OVER JAVA, MOBILE AND WEB APPLICATIONSIMPAIRMENT SIMULATION OVER JAVA, MOBILE AND WEB APPLICATIONS
    64. 64. University of SurreyDepartment of Computing• ACT-R: a cognitivearchitecture showing howhuman cognition works.• The proposed VUM includesACT-R parameters.• Reading task example– a non-stressed user• visual-attention-latency: 0.085 sec• reading task – total time: 3.57 sec– a user with acute stress• visual-attention-latency: 0.0978 sec• reading task – total time: 3.726 secCOGNITIVE SIMULATION – THE PROPOSED VUMs WITHIN ACT-RCOGNITIVE SIMULATION – THE PROPOSED VUMs WITHIN ACT-R
    65. 65. University of SurreyDepartment of Computing• Introduction•A new User Modelable to describeelderly and disabledpeople•An innovative usermodeling methodologyfocused on the elderlyand disabled• A methodology forcreating virtual usermodels representingspecific populationgroups• The proposed usermodeling methodologyin practice• ConclusionsConclusions – Future Work
    66. 66. University of SurreyDepartment of Computing1. A new Virtual User Model was introduced, able to describe a variety ofuser characteristics, including physical, cognitive and behavioural.2. This Virtual User Model is a part of an innovative coupled user and taskmodeling methodology that aims to be used mainly in simulationframeworks performing accessibility assessment of virtual designs.3. A methodology for creating virtual user models able to representspecific disabled population groups4. Tools for putting the proposed user modeling methodology into practiceCONCLUSIONSCONCLUSIONS
    67. 67. University of SurreyDepartment of Computing• The proposed VUM goes one step beyond the state-of-the-art, as itenables the detailed description of the functioning of the elderly anddisabled• The proposed coupled user & task modeling methodology can be avaluable tool for the designers/developers towards the development ofaccessible products• The proposed methodology for creating virtual user models able torepresent specific disabled population groups offers designers withthe opportunity to define an accessibility threshold for their designs.IMPORTANCE OF THE RESEARCH CONTRIBUTIONIMPORTANCE OF THE RESEARCH CONTRIBUTION
    68. 68. University of SurreyDepartment of Computing• The accuracy of a VUM representing a specific population groupdepends on the sample size.• However, the proposed methodology can result to the productionof very accurate personas.CRITICAL DISCUSSIONCRITICAL DISCUSSION
    69. 69. University of SurreyDepartment of Computing• Fine-tuning of the resulted regression models using• more data coming from the literature• more real measurements• More disabilities could be modeled• A coupled physical-cognitive user modeling approach could beconsidered• Continue the dissemination through– the VUMS cluster of projects (http://www.veritas-project.eu/vums/)– the activities of the W3C MBUI Working GroupFUTURE WORKFUTURE WORK
    70. 70. University of SurreyDepartment of Computing1. Kaklanis, N., Moschonas, P., Moustakas, K., Tzovaras, D., “Virtual User Models for the elderly and disabledfor automatic simulated accessibility and ergonomy evaluation of designs”, Universal Access in theInformation Society, Special Issue: Accessibility aspects in UIDLs, Springer2. Giakoumis, D., Kaklanis, N., Votis, K., Tzovaras, D., “Enabling user interface developers to understandaccessibility limitations through visual, hearing, physical and cognitive impairment simulation”, UniversalAccess in the Information Society, Springer3. Kaklanis, N., Votis, K., Tzovaras, D., ”Open Touch/Sound Maps: A system to convey street data throughhaptic and auditory feedback”, Computers & Geosciences, Elsevier (accepted for publication)4. Kaklanis, N., Votis, K., Tzovaras, D., “An online multimodal audio-haptic framework for visually impairedusers accessing OpenStreetMap data”, Universal Access in the Information Society, Special Issue: 3rdgeneration accessibility: Information and Communication Technologies towards universal access, Springer(under review)5. Kaklanis, N., Biswas, P., Mohamad, Y., Gonzalez, M.F., Peissner, M., Langdon, P., Tzovaras, D., Jung, C,“Towards Standardization of User Models for Simulation and Adaptation Purposes”, Universal Access in theInformation Society, Special Issue: 3rd generation accessibility: Information and CommunicationTechnologies towards universal access, Springer (under review)6. Koutkias, V., Kaklanis, N., Votis, K., Tzovaras, D., Maglaveras, N., “An Integrated Semantic FrameworkSupporting Universal Accessibility to ICT”, Universal Access in the Information Society, Special Issue: 3rdgeneration accessibility: Information and Communication Technologies towards universal access, Springer(under review)7. Kaklanis, N., Stavropoulos, G., Tzovaras, D., “Modeling disabilities through regression analysis for thedevelopment of accurate virtual user models”, User Modeling and User-Adapted Interaction, The Journal ofPersonalization Research (under review)PUBLICATIONS – SCIENTIFIC JOURNALSPUBLICATIONS – SCIENTIFIC JOURNALS
    71. 71. University of SurreyDepartment of Computing8. Kaklanis, N., Votis, K., Moustakas, K., Tzovaras, D. 3D HapticWebBrowser: Towards Universal WebNavigation for the Visually Impaired, 7th International Conference on Web Accessibility, W4A, 26-27 April2010, Raleigh, North Carolina, USAThis paper won the "Judges Award” at the “Web Accessibility Challenge” (sponsored by Microsoft)9. Kaklanis, N., Moustakas, K., Votis, K., Tzovaras, D. A framework for accessibility testing of virtualenvironments based on UsiXML”, ACM SIGCHI Symposium on Engineering Interactive ComputingSystems, UsiXML-EICS 2010, Berlin, June 201010. Kaklanis, N. Moschonas, P., Moustakas, K., Tzovaras, D., Enforcing accessible design of products andservices through simulated accessibility evaluation, CONFIDENCE 2010, Jyväskylä, Finland, 9-10 December201011. Kaklanis, N., Moustakas, K., Tzovaras, D. A framework for automatic simulated accessibility assessment invirtual environments, HCI International 2011, Orlando, Florida, USA, 9-14 July 201112. Kaklanis, N., Votis, K., Moschonas, P., Tzovaras, D. HapticRiaMaps: Towards Interactive exploration ofWeb World maps for the Visually Impaired”, W4A 2011, Hyderabad, India, March 201113. Oikonomou, T., Kaklanis, N., Votis, K., Kastori, G.E., Partarakis, N., Tzovaras, D. WaaT: Personalised WebAccessibility Evaluation Tool, W4A 2011, Hyderabad, India, March 201114. Oikonomou, T., Kaklanis, N., Votis, K., Tzovaras, D. An Accessibility Assessment Framework forImproving Designers Experience in Web Applications, HCII 2011, Orlando, Florida, USA, July 201115. Partarakis, N., Doulgeraki, C., Antona, M., Oikonomou, T., Kaklanis, N., Votis, K., Kastori, G.E., Tzovaras,D. A Unified Environment for Accessing a Suite of Accessibility Evaluation Facilities, HCII 2011, Orlando,Florida, USA, July 201116. Moschonas, P., Kaklanis, N., Tsakiris, A., Moustakas, K., Tzovaras, D., “An open simulation framework forautomated and immersive accessibility engineering”, 6th Cambridge workshop on universal access andassistive technology, CWUAAT 2012, Fitzwilliam College, University of Cambridge, March, 2012PUBLICATIONS – INTERNATIONAL CONFERENCESPUBLICATIONS – INTERNATIONAL CONFERENCES
    72. 72. University of SurreyDepartment of Computing17. Korn, P., Kaklanis, N., Votis, K., Tzovaras, D., “Using the AEGIS OAF: making an accessible RIA”, 27thAnnual International Technology and Persons with Disabilities Conference, CSUN 2012, March, 2012, SanDiego, CA18. Moschonas, P., Kaklanis, N., Tzovaras, D., “Novel Human Factors for Ergonomy Evaluation in VirtualEnvironments Using Virtual User Models”, 10th International Conference on Virtual Reality Continuum andits Applications in Industry, VRCAI’ 2011, December, 2011, Hong Kong, China19. Kaklanis, N., Moustakas, K., Tzovaras, D., "An extension of UsiXML enabling the detailed description ofusers including elderly and disabled", in International Workshop on Software Support for User InterfaceDescription Language, Interact 2011, Lisbon, September, 201120. Mohamad, Y., Biswas, P., Langdon, P., Peissner, M., Dangelmaier, M., Jung, C., Wolf P., Kaklanis, N.,“Standardisation of user models for designing and using inclusive products”, Joint Virtual Reality Conference(JVRC), 20-21 September, 2011, Nottingham, UK21. Kaklanis, N., Moustakas, K., Tzovaras, D., 2012. A methodology for generating virtual user models ofelderly and disabled for the accessibility assessment of new products. In Proceedings of the 13thinternational conference on Computers Helping People with Special Needs - Volume Part I (ICCHP12),Klaus Miesenberger, Arthur Karshmer, Petr Penaz, and Wolfgang Zagler (Eds.), Vol. Part I. Springer-Verlag,Berlin, Heidelberg, 295-302. DOI=10.1007/978-3-642-31522-0_44http://dx.doi.org/10.1007/978-3-642-31522-0_4422. Moschonas, P., Tsakiris., A., Kaklanis, N., Stavropoulos, G., Tzovaras, D., “Holistic accessibility evaluationusing VR simulation of users with special needs”, UMAP 2012, Montreal, Canada, July, 201223. Kaklanis, N., Mohamad, Y., Peissner, M., Biswas, P., Langdon, P., Tzovaras, D., ”An Interoperable andInclusive User Modelling concept for Simulation and Adaptation”, UMAP 2012, Montreal, Canada, July,201224. Kaklanis, N., Votis, K., Tzovaras, D., “Touching OpenStreetMap Data in Mobile Context for the VisuallyPUBLICATIONS – INTERNATIONAL CONFERENCESPUBLICATIONS – INTERNATIONAL CONFERENCES
    73. 73. University of SurreyDepartment of Computing25. Kaklanis, N., Votis, K., Tzovaras, D., “A Mobile Interactive Maps Application for a Visually ImpairedAudience”, The Paciello Group Web Accessibility Challenge - W4A 2013, 10th International Cross-Disciplinary Conference on Web Accessibility, 13-15 May 2013, Rio de Janeiro, Brazil (accepted)26. Tsakiris, A., Kaklanis, N., Paliokas, I., Stavropoulos G., and Tzovaras, D. Cognitive Impairments Simulationin a Holistic GUI Accessibility Assessment Framework, 12th European AAATE Conference, 19-22September 2013, Vilamoura, Algarve, Portugal (under review)PUBLICATIONS – INTERNATIONAL CONFERENCESPUBLICATIONS – INTERNATIONAL CONFERENCES
    74. 74. University of SurreyDepartment of Computing27. Kaklanis, N., Moustakas, K., Tzovaras, D. Haptic Rendering of HTML components and 2D maps included inweb pages, in G. Ghinea, F. Andres, S. Gulliver (Eds) Multiple Sensorial Media Advances and Applications:New Developments in MulSeMedia, 201028. Biswas, P., Kaklanis, N., Mohamad, Y., Peissner, M., Langdon, P., Tzovaras, D., Jung, C., “An Interoperableand Inclusive User Modeling concept for Simulation and Adaptation”, A Multimodal End-2-End Approach toAccessible Computing, SpringerPUBLICATIONS – BOOK CHAPTERSPUBLICATIONS – BOOK CHAPTERS
    75. 75. University of SurreyDepartment of ComputingTHANK YOU!Next...video presentations

    ×