Nicole Karpinsky defended her thesis which examined how spatial memory load affects compliance with an imperfect automated decision aid during a visual detection task. The study found that while the aid improved performance, spatial memory load reduced compliance compared to a no load condition. Compliance was higher for a near-perfect aid than a more fallible aid. The results provide insights into how workload influences human-automation interaction, though have some differences from prior studies that examined the impact of distracted conditions on compliance. Addressing limitations and exploring attentional demands could help characterize automation use in attention-demanding environments.
From Engineer to Alchemist, There and Back Again: An Alchemist TaleDanilo Pianini
When computer science meets (bio)chemistry, a new world of possibilities emerges. Taking inspiration from mechanisms that exist in nature, and empowering them with the magic of computation, engineers become alchemists: they create new worlds, ruled by their own (bio)chemical rules. We will call such virtual worlds"computational ecosystems". In this talk we introduce Alchemist, a simulator developed by our research group in order to let us experiment with this complexity. Alchemist takes inspiration from chemical simulators and ABMs, trying to grab the best of both worlds by providing a fast, reliable and extensible framework. We will have a flyby of the simulator features, showing some case studies: crowd steering, crowd evacuation, morphogenesis, anticipative adaptation. The final part of the talk will frame the future activities, offering a pool of opportunities to those interested in this research area
From Engineer to Alchemist, There and Back Again: An Alchemist TaleDanilo Pianini
When computer science meets (bio)chemistry, a new world of possibilities emerges. Taking inspiration from mechanisms that exist in nature, and empowering them with the magic of computation, engineers become alchemists: they create new worlds, ruled by their own (bio)chemical rules. We will call such virtual worlds"computational ecosystems". In this talk we introduce Alchemist, a simulator developed by our research group in order to let us experiment with this complexity. Alchemist takes inspiration from chemical simulators and ABMs, trying to grab the best of both worlds by providing a fast, reliable and extensible framework. We will have a flyby of the simulator features, showing some case studies: crowd steering, crowd evacuation, morphogenesis, anticipative adaptation. The final part of the talk will frame the future activities, offering a pool of opportunities to those interested in this research area
Influence of Timeline and Named-entity Components on User Engagement Roi Blanco
Nowadays, successful applications are those which contain features that captivate and engage users. Using an interactive news retrieval system as a use case, in this paper we study the effect of timeline and named-entity components on user engagement. This is in contrast with previous studies where the importance of these components were studied from a retrieval effectiveness point of view. Our experimental results show significant improvements in user engagement when named-entity and timeline components were installed. Further, we investigate if we can predict user-centred metrics through user's interaction with the system. Results show that we can successfully learn a model that predicts all dimensions of user engagement and whether users will like the system or not. These findings might steer systems that apply a more personalised user experience, tailored to the user's preferences.
Automated software testing cases generation framework to ensure the efficienc...Sheikh Monirul Hasan
Automated Software Testing Cases Generation Framework to Ensure the Efficiency of the Gesture Recognition Systems by Sheikh Monirul Hasan. This is a research work for creating a standard or benchmark for testing gesture recognition system software. sheikh monirul Hasan is the first author of the research paper, the complete work summary of the research we tried to discuss in the presentation slide. so there have many kinds of software engineering think and how we get a quality product specially for gesture recognition system.
This slide gives brief overview of supervised, unsupervised and reinforcement learning. Algorithms discussed are Naive Bayes, K nearest neighbour, SVM,decision tree, Markov model.
Difference between regression and classification. difference between supervised and reinforcement, iterative functioning of Markov model and machine learning applications.
Audio Music Similarity is a task within Music Information Retrieval that deals with systems that retrieve songs musically similar to a query song according to their audio content. Evaluation experiments are the main scientific tool in Information Retrieval to determine what systems work better and advance the state of the art accordingly. It is therefore essential that the conclusions drawn from these experiments are both valid and reliable, and that we can reach them at a low cost. This dissertation studies these three aspects of evaluation experiments for the particular case of Audio Music Similarity, with the general goal of improving how these systems are evaluated. The traditional paradigm for Information Retrieval evaluation based on test collections is approached as an statistical estimator of certain probability distributions that characterize how users employ systems. In terms of validity, we study how well the measured system distributions correspond to the target user distributions, and how this correspondence affects the conclusions we draw from an experiment. In terms of reliability, we study the optimal characteristics of test collections and statistical procedures, and in terms of efficiency we study models and methods to greatly reduce the cost of running an evaluation experiment.
A major goal of this study is to address the use and functionality of the impaired arm through specific assessment, health application and wearable. So far Rehabilitation Gaming System Wearable (RGS-wear) has focused on the amount of movement. In this thesis, the aim is to enhance the current state and establish a novel measurement providing qualitative assessment of movement. Once understanding the rationale for motor learning, impairments and motor control I further developed, and validated features for the rehabilitation applied technology RGS-wear. The execution of this project was divided into three main stages. The first step included kinesthetic data acquisition and assessment, through the use of wearable sensors. Secondly, I performed motion evaluation, analyzed and compared non-dominant and dominant hand movement, in natural and constrained settings, studied patterns and extracted measures of motor function. Thirdly, I studied the functionalities of the wearable and evaluated the acceptability of the wearable as an evaluation tool. The goal of this project was to design and implement appropriate system features and strategies that can augment current rehabilitation protocols. The outcome I believe carries the potential to lead to new guidelines and recommendations for the development of wearable technologies for clinical practices especially in context of motor function.
Oplægget blev holdt ved InfinIT-arrangementet "Temadag om Remote Usability Testing" afholdt den 17. januar 2012. Læs mere på http://www.infinit.dk/dk/hvad_kan_vi_goere_for_dig/viden/reportager/mere_tid_prioritering_og_penge_til_brugbarhedstests.htm
Machine Learning Challenges For Automated Prompting In Smart HomesBarnan Das
As the world's population ages, there is an increased prevalence of diseases related to aging, such as dementia. Caring for individuals with dementia is frequently associated with extreme physical and emotional stress, which often leads to depression. Smart home technology and advances in machine learning techniques can provide innovative solutions to reduce caregiver burden. One key service that caregivers provide is prompting individuals with memory limitations to initiate and complete daily activities. We hypothesize that sensor technologies combined with machine learning techniques can automate the process of providing reminder-based interventions or prompts. This dissertation focuses on addressing machine learning challenges that arise while devising an effective automated prompting system.
Our first goal is to emulate natural interventions provided by a caregiver to individuals with memory impairments, by using a supervised machine learning approach to classify pre-segmented activity steps into prompt or no-prompt classes. However, the lack of training examples representing prompt situations causes imbalanced class distribution. We proposed two probabilistic oversampling techniques, RACOG and wRACOG, that help in better learning of the``prompt'' class. Moreover, there are certain prompt situations where the sensor triggering signature is quite similar to the situations when the participant would probably need no prompt. The absence of sufficient data attributes to differentiate between prompt and no-prompt classes causes class overlap. We propose ClusBUS, a clustering-based under-sampling technique that identifies ambiguous data regions. ClusBUS preprocesses the data in order to give more importance to the minority class during classification.
Our second goal is to automatically detect activity errors in real time, while an individual performs an activity. We propose a collection of one-class classification-based algorithms, known as DERT, that learns only from the normal activity patterns and without using any training examples for the activity errors. When evaluated on unseen activity data, DERT is able to identify abnormalities or errors, which can be potential prompt situations.
We validate the effectiveness of the proposed algorithms in predicting potential prompt situations on the sensor data of ten activities of daily living, collected from 580 participants, who were part of two smart home studies.
Evaluation in (Music) Information Retrieval through the Audio Music Similarit...Julián Urbano
Test-collection based evaluation in (Music) Information Retrieval has been used for half a century now as the means to evaluate and compare retrieval techniques and advance the state of the art. However, this paradigm makes certain assumptions that remain a research problem and that may invalidate our experimental results. In this talk I will approach this paradigm as an estimator of certain probability distributions that describe the final user experience. These distributions are estimated with a test collection, computing system-related distributions assumed to reliably correlate with the target user-related distributions.
Using the Audio Music Similarity task as an example, I will talk about issues with our current evaluation methods, the degree to which they are problematic, how to analyze them and improve the situation. In terms of validity, we will see how the measured system distributions correspond to the target user distributions, and how this correspondence affects the conclusions we draw from an experiment. In terms of reliability, we will discuss optimal characteristics of test collections and statistical procedures. In terms of efficiency, we discuss models and methods to greatly reduce the annotation cost of an evaluation experiment.
A SYSTEM is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal
This exhaustive and vibrant PowerPoint has around 90 slides and explains in detail all the must know concepts of Management in Healthcare. These slides have enough information to use it for 3 hour seminar (2 sessions) on Modern Management Techniques and its application in Healthcare. The session can be further extended if the concepts are explained with appropriate examples.
The effects of visual realism on search tasks in mixed reality simulations-IE...Yadhu Kiran
Abstract—In this paper, we investigate the validity of Mixed Reality (MR) Simulation by conducting an experiment studying the effects of the visual realism of the simulated environment on various search tasks in Augmented Reality (AR). MR Simulation is a practical approach to conducting controlled and repeatable user experiments in MR, including AR. This approach uses a high-fidelity Virtual
Reality (VR) display system to simulate a wide range of equal or lower fidelity displays from the MR continuum, for the express purpose of conducting user experiments. For the experiment, we created three virtual models of a real-world location, each with a different perceived level of visual realism. We designed and executed an AR experiment using the real-world location and repeated
the experiment within VR using the three virtual models we created. The experiment looked into how fast users could search for both physical and virtual information that was present in the scene. Our experiment demonstrates the usefulness of MR Simulation and provides early evidence for the validity of MR Simulation with respect to AR search tasks performed in immersive VR.
Influence of Timeline and Named-entity Components on User Engagement Roi Blanco
Nowadays, successful applications are those which contain features that captivate and engage users. Using an interactive news retrieval system as a use case, in this paper we study the effect of timeline and named-entity components on user engagement. This is in contrast with previous studies where the importance of these components were studied from a retrieval effectiveness point of view. Our experimental results show significant improvements in user engagement when named-entity and timeline components were installed. Further, we investigate if we can predict user-centred metrics through user's interaction with the system. Results show that we can successfully learn a model that predicts all dimensions of user engagement and whether users will like the system or not. These findings might steer systems that apply a more personalised user experience, tailored to the user's preferences.
Automated software testing cases generation framework to ensure the efficienc...Sheikh Monirul Hasan
Automated Software Testing Cases Generation Framework to Ensure the Efficiency of the Gesture Recognition Systems by Sheikh Monirul Hasan. This is a research work for creating a standard or benchmark for testing gesture recognition system software. sheikh monirul Hasan is the first author of the research paper, the complete work summary of the research we tried to discuss in the presentation slide. so there have many kinds of software engineering think and how we get a quality product specially for gesture recognition system.
This slide gives brief overview of supervised, unsupervised and reinforcement learning. Algorithms discussed are Naive Bayes, K nearest neighbour, SVM,decision tree, Markov model.
Difference between regression and classification. difference between supervised and reinforcement, iterative functioning of Markov model and machine learning applications.
Audio Music Similarity is a task within Music Information Retrieval that deals with systems that retrieve songs musically similar to a query song according to their audio content. Evaluation experiments are the main scientific tool in Information Retrieval to determine what systems work better and advance the state of the art accordingly. It is therefore essential that the conclusions drawn from these experiments are both valid and reliable, and that we can reach them at a low cost. This dissertation studies these three aspects of evaluation experiments for the particular case of Audio Music Similarity, with the general goal of improving how these systems are evaluated. The traditional paradigm for Information Retrieval evaluation based on test collections is approached as an statistical estimator of certain probability distributions that characterize how users employ systems. In terms of validity, we study how well the measured system distributions correspond to the target user distributions, and how this correspondence affects the conclusions we draw from an experiment. In terms of reliability, we study the optimal characteristics of test collections and statistical procedures, and in terms of efficiency we study models and methods to greatly reduce the cost of running an evaluation experiment.
A major goal of this study is to address the use and functionality of the impaired arm through specific assessment, health application and wearable. So far Rehabilitation Gaming System Wearable (RGS-wear) has focused on the amount of movement. In this thesis, the aim is to enhance the current state and establish a novel measurement providing qualitative assessment of movement. Once understanding the rationale for motor learning, impairments and motor control I further developed, and validated features for the rehabilitation applied technology RGS-wear. The execution of this project was divided into three main stages. The first step included kinesthetic data acquisition and assessment, through the use of wearable sensors. Secondly, I performed motion evaluation, analyzed and compared non-dominant and dominant hand movement, in natural and constrained settings, studied patterns and extracted measures of motor function. Thirdly, I studied the functionalities of the wearable and evaluated the acceptability of the wearable as an evaluation tool. The goal of this project was to design and implement appropriate system features and strategies that can augment current rehabilitation protocols. The outcome I believe carries the potential to lead to new guidelines and recommendations for the development of wearable technologies for clinical practices especially in context of motor function.
Oplægget blev holdt ved InfinIT-arrangementet "Temadag om Remote Usability Testing" afholdt den 17. januar 2012. Læs mere på http://www.infinit.dk/dk/hvad_kan_vi_goere_for_dig/viden/reportager/mere_tid_prioritering_og_penge_til_brugbarhedstests.htm
Machine Learning Challenges For Automated Prompting In Smart HomesBarnan Das
As the world's population ages, there is an increased prevalence of diseases related to aging, such as dementia. Caring for individuals with dementia is frequently associated with extreme physical and emotional stress, which often leads to depression. Smart home technology and advances in machine learning techniques can provide innovative solutions to reduce caregiver burden. One key service that caregivers provide is prompting individuals with memory limitations to initiate and complete daily activities. We hypothesize that sensor technologies combined with machine learning techniques can automate the process of providing reminder-based interventions or prompts. This dissertation focuses on addressing machine learning challenges that arise while devising an effective automated prompting system.
Our first goal is to emulate natural interventions provided by a caregiver to individuals with memory impairments, by using a supervised machine learning approach to classify pre-segmented activity steps into prompt or no-prompt classes. However, the lack of training examples representing prompt situations causes imbalanced class distribution. We proposed two probabilistic oversampling techniques, RACOG and wRACOG, that help in better learning of the``prompt'' class. Moreover, there are certain prompt situations where the sensor triggering signature is quite similar to the situations when the participant would probably need no prompt. The absence of sufficient data attributes to differentiate between prompt and no-prompt classes causes class overlap. We propose ClusBUS, a clustering-based under-sampling technique that identifies ambiguous data regions. ClusBUS preprocesses the data in order to give more importance to the minority class during classification.
Our second goal is to automatically detect activity errors in real time, while an individual performs an activity. We propose a collection of one-class classification-based algorithms, known as DERT, that learns only from the normal activity patterns and without using any training examples for the activity errors. When evaluated on unseen activity data, DERT is able to identify abnormalities or errors, which can be potential prompt situations.
We validate the effectiveness of the proposed algorithms in predicting potential prompt situations on the sensor data of ten activities of daily living, collected from 580 participants, who were part of two smart home studies.
Evaluation in (Music) Information Retrieval through the Audio Music Similarit...Julián Urbano
Test-collection based evaluation in (Music) Information Retrieval has been used for half a century now as the means to evaluate and compare retrieval techniques and advance the state of the art. However, this paradigm makes certain assumptions that remain a research problem and that may invalidate our experimental results. In this talk I will approach this paradigm as an estimator of certain probability distributions that describe the final user experience. These distributions are estimated with a test collection, computing system-related distributions assumed to reliably correlate with the target user-related distributions.
Using the Audio Music Similarity task as an example, I will talk about issues with our current evaluation methods, the degree to which they are problematic, how to analyze them and improve the situation. In terms of validity, we will see how the measured system distributions correspond to the target user distributions, and how this correspondence affects the conclusions we draw from an experiment. In terms of reliability, we will discuss optimal characteristics of test collections and statistical procedures. In terms of efficiency, we discuss models and methods to greatly reduce the annotation cost of an evaluation experiment.
A SYSTEM is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal
This exhaustive and vibrant PowerPoint has around 90 slides and explains in detail all the must know concepts of Management in Healthcare. These slides have enough information to use it for 3 hour seminar (2 sessions) on Modern Management Techniques and its application in Healthcare. The session can be further extended if the concepts are explained with appropriate examples.
The effects of visual realism on search tasks in mixed reality simulations-IE...Yadhu Kiran
Abstract—In this paper, we investigate the validity of Mixed Reality (MR) Simulation by conducting an experiment studying the effects of the visual realism of the simulated environment on various search tasks in Augmented Reality (AR). MR Simulation is a practical approach to conducting controlled and repeatable user experiments in MR, including AR. This approach uses a high-fidelity Virtual
Reality (VR) display system to simulate a wide range of equal or lower fidelity displays from the MR continuum, for the express purpose of conducting user experiments. For the experiment, we created three virtual models of a real-world location, each with a different perceived level of visual realism. We designed and executed an AR experiment using the real-world location and repeated
the experiment within VR using the three virtual models we created. The experiment looked into how fast users could search for both physical and virtual information that was present in the scene. Our experiment demonstrates the usefulness of MR Simulation and provides early evidence for the validity of MR Simulation with respect to AR search tasks performed in immersive VR.
The effects of visual realism on search tasks in mixed reality simulations-IE...
Thesis Defense_Karpinsky
1. Thesis Defense
Nicole D. Karpinsky
September 29, 2016
The effect of concurrent visuospatial
memory demand on automation use
in a visual detection task
2. Automation can reduce workload, improve efficiency, and
increase safety for human operators (https://google.com/)
3. Automated Aids and Human Performance
Sensory
Processing
Perception
Decision
Making
Response
Selection &
Execution
Information
Acquisition
Information
Analysis
Decision
Selection
Action
Implementation
Human-Information
Processing Model
(Wickens et al., 2015)
Stages/Levels of
Automation
(Parasuraman, Sheridan, & Wickens, 2000)
4. HIGH
LOW
10. The computer decides everything, acts autonomously, ignoring the human
9. Informs the human only if it, the computer, decides to
8. Informs the human only if asked, or
7. Executes automatically, then necessarily informs the human, and
6. allows the human a restricted time to veto before automatic execution, or
5. Executes that suggestion if the human approves, or
4. Suggests one alternative
3. Narrows the selection down to a few, or
2. The computer offers a complete set of decision/action alternatives, or
1. The computer offers no assistance: human must take all decisions and actions.
LOA of Decision and Action Selection
(Sheridan & Verplank, 1978)
5. Automation Use
• Appropriate automation usage
allows operator to allot tasks to
automation to increase safety
and performance (Lee, 2008; Lee & See, 2004)
• Unfortunately, operators do not
always use automation as
prescribed
• misuse and disuse (Parasuraman & Riley, 1997)
8. Signal Detection Theory (SDT)
• Statistical theory of decision
making in uncertain situations
• Characterizes operators’
responses toward imperfect
automated decision aids
• Isolates sensitivity and
response bias towards these
decisions
9. Signal Detection Theory (SDT)
Sensitivity (d’)
• Measures ability to discriminate
a stimulus from noise
• Represented as distance
between the means of signal
and noise distributions
Response bias (c)
• Indicates amount of evidence
needed to accumulate to make
a response
• Yes-No responses
10. Imperfect Automation Systems
• Compliance: Degree to which the operator takes action when the
system indicates that there is a signal
(Meyer et al., 2004)
Compliance = C Control – C Alert
11. Automation Use and Workload
• Previous studies examined the effects of increased cognitive load
on automation use
• In an automated warehouse-management system control task,
operators’ compliance increased with higher levels of workload (McBride
et al., 2011)
• Competition for attentional resources did not interfere with
response choices toward imperfect automation, resulting in no
compliance (Botzer et al., 2013)
12. Research Question
How does spatial memory load modulate
operators’ compliance towards an imperfect
decision aid?
13. Hypothesis 1: Main Effect of Automation
• Operators will comply with responses issued by an automated
decision aid less when performance interference occurs between
primary (aided) and secondary (unaided) tasks than when it does
not
14. Hypothesis 2: Main Effect of Load
• Operators will comply with the aid less in the spatial memory
condition than the single task condition
15. Hypothesis 3: Interaction
• Operators will comply with the aid similarly in the non-spatial
memory and the single task conditions
16. Design
• 3 × 2 × 2 mixed design
• Memory Load: NVSM, VSM, ST
• Automation Availability: Present vs. Absent
• Automation Type: FA-prone (PPV = .75 and NPV = .95) vs.
Near Perfect (PPV = .95 and NPV = .95)
17. Participants
• 30 participants from ODU participated in the study
• 23 females, (M = 24.65 years, SD = 4.92 years)
• Participants were screened for normal to corrected-to-normal visual
acuity and normal color vision
• Received research credit through the SONA system
18. Apparatus
• 23.6” LED monitor (1920 × 1080 px)
oRefresh rate of 60 Hz
• E-Prime 2.0
• Computer mouse
• Dimmed and quiet room
19. Visual Search Task
• Distractor letters Q and target
letter O, positioned at 4
random angles
• Each display contained 25
letters in 5 × 5 pattern
• Target was present half of trials
• Position and orientation were
random each trial
20. Non-Visual Spatial
Memory Task
• Solid colored squares, positioned at 4 set
locations
• Color of squares were randomly selected
• Memory test probe display contained 1
colored square at the middle of the display
• Probe square contained 1 memory stimuli
half of the trials
21. Visual Spatial Memory
• Solid black squares, positioned at 8
possible locations
• Location of squares were randomly
selected
• Memory test probe display contained 1
empty square at 1 of the 8 locations
• Probe square appeared at 1 of the
locations of the memory stimuli half of the
trials
23. Statistical Analysis
• All statistical analyses adopted standard level of alpha (.05)
• Data with incorrect responses in memory tasks were removed
• Log-linear transformations were used to correct for extreme STD
scores (Hautus, 1995)
• Greenhouse-Geisser used to correct any sphericity violations
• Bonferroni used for post-hoc analyses
24. Results: Sensitivity, d’
• Sensitivity was greater for aided than unaided conditions
• (M = 1.36 vs. .67), true for FA-Prone and NP conditions
• Sensitivity was greater for Spatial Load than No Load condition
• (M = 1.24 vs. .73)
• NP condition had greater sensitivity in Spatial Load and Non-Spatial
Load conditions
25. Results: Sensitivity, d’
• For the NP condition, sensitivity
was greater in Spatial Load
condition compared to No Load
and Non-Spatial Load condition
• Automation Availability was
more pronounced for Spatial
Load condition compared to No
Load condition
26. Results: Compliance, c
• Compliance was greater
when the aid was present in
No Load than Non-Spatial
Load condition
• Compliance was also
greater in FA-Prone than NP
system
27. Discussion
• As expected, the aid improved search performance in all conditions
• There was a cost of dual-tasking towards compliance
• Results are similar and different from previous research
• Phillips and Madhavan, 2011
• Karpinsky et al., 2016
28. Discussion
Phillips & Madhavan, 2011
• Similar: Automation improved
performance in the aided
condition compared to the
unaided conditions
• Different: Compliance was
higher in all distracted
conditions compared to the
non-distracted condition
Karpinsky et al., 2016
• Similar: Participants complied
less as performance
interference increased
• Different: Compliance was
higher in all distracted
conditions compared to the
non-distracted condition
29. Discussion
Limitations
• Memory performance was
lower in the current study
compared to previous research
(76.5% vs. 65%)
Future Considerations
• Researchers should look into
the effects of attentional
demand incurred by secondary
tasks on automation use
• Will help to further characterize
HAI in attention demanding
environments