Assessing user bias in affect detection within context-based spoken dialog systems

This paper presents an empirical evidence of user bias within a laboratory-oriented evaluation of a Spoken Dialog System. Specifically, we addressed user bias in their satisfaction judgements. We question the reliability of this data for modeling user emotion, focusing on contentment and frustration in a spoken dialog system. This bias is detected through machine learning experiments that were conducted on two datasets, users and annotators, which were then compared in order to assess the reliability of these datasets. The target used was the satisfaction rating and the predictors were conversational/dialog features. Our results indicated that standard classifiers were significantly more successful in discriminating frustration and contentment and the intensities of these emotions (reflected by user satisfaction ratings) from annotator data than from user data. Indirectly, the results showed that conversational features are reliable predictors of the two abovementioned emotions.

Personality and the healthy lifestyle as predictors of physical health: Can the healthy lifestyle be explained by personality?

In health and epidemiological research, the Healthy Lifestyle (HLS) is often invoked as an explanation for inconsistent effects. Modifiable components of the HLS are advocated as a panacea for the most common threats to public health. Biases resulting from the HLS are theorized to result from covariance among its components. This covariance has not yet been formally modeled. Furthermore, no mechanism has been proposed to explain this covariance among these factors. Using three large nationally representative samples, I evaluated the HLS as a latent variable. Using structural equation modeling (SEM) I evaluated the degree to which the shared variance of HLS components is accounted for by personality traits, and tested the HLS as a mediator of the personality health relationship. Across all three samples, the HLS fits well as a latent variable, is partially accounted for by personality traits, and mediates the effects of personality traits on health. In all cases personality traits have direct effects on health independent of the HLS. These results suggest that the utility of personality traits as predictors of health exceeds that provided by commonly used lifestyle predictors.

Precise determination of aquatic plant wet mass using a salad spinner

The reliable assessment of macrophyte biomass is fundamental for ecological research and management of freshwater ecosystems. While dry mass is routinely used to determine aquatic plant biomass, wet (fresh) mass can be more practical. We tested the accuracy and precision of wet mass measurements by using a salad spinner to remove surface water from four macrophyte species differing in growth form and architectural complexity. The salad spinner aided in making precise and accurate wet mass with less than 3% error. There was also little difference between operators, with a user bias estimated to be below 5%. To achieve this level of precision, only 10–20 turns of the salad spinner are needed. Therefore, wet mass of a sample can be determined in less than 1 min. We demonstrated that a salad spinner is a rapid and economical technique to enable precise and accurate macrophyte wet mass measurements and is particularly suitable for experimental work. The method will also be useful for fieldwork in situations when sample sizes are not overly large.

Design Optimisation for Compression Moulded Sheet Moulded Compounds

This paper builds on previous work to show how using holistic and iterative design optimisation tools can be used to produce a commercially viable product that reduces a costly assembly into a single moulded structure. An assembly consisting of a structural metallic support and compression moulding outer shell undergo design optimisation and analysis to remove the support from the assembly process in favour of a structural moulding. The support is analysed and a sheet moulded compound (SMC) alternative is presented, this is then combined into a manufacturable shell design which is then assessed on viability as an alternative to the original.
Alongside this a robust material selection system is implemented that removes user bias towards materials for designs. This system builds on work set out by the Cambridge Material Selector and Boothroyd and Dewhurst, while using a selection of applicable materials currently available for the compression moulding process. This material selection process has been linked into the design and analysis stage, via scripts for use in the finite element environment. This builds towards an analysis toolkit that is suggested to develop and enhance manufacturability of design studies.

Interactive Segmentation of MR Images from Brain Tumor Patients

Medical doctors often do not trust the result of fully automatic segmentations because they have no possibility to make corrections if necessary. On the other hand, manual corrections can introduce a user bias. In this work, we propose to integrate the possibility for quick manual corrections into a fully automatic segmentation method for brain tumor images. This allows for necessary corrections while maintaining a high objectiveness. The underlying idea is similar to the well-known Grab-Cut algorithm, but here we combine decision forest classification with conditional random field regularization for interactive segmentation of 3D medical images. The approach has been evaluated by two different users on the BraTS2012 dataset. Accuracy and robustness improved compared to a fully automatic method and our interactive approach was ranked among the top performing methods. Time for computation including manual interaction was less than 10 minutes per patient, which makes it attractive for clinical use.

A Satisfaction-based Model for Affect Recognition from Conversational Features in Spoken Dialog Systems

Detecting user affect automatically during real-time conversation is the main challenge towards our greater aim of infusing social intelligence into a natural-language mixed-initiative High-Fidelity (Hi-Fi) audio control spoken dialog agent. In recent years, studies on affect detection from voice have moved on to using realistic, non-acted data, which is subtler. However, it is more challenging to perceive subtler emotions and this is demonstrated in tasks such as labelling and machine prediction. This paper attempts to address part of this challenge by considering the role of user satisfaction ratings and also conversational/dialog features in discriminating contentment and frustration, two types of emotions that are known to be prevalent within spoken human-computer interaction. However, given the laboratory constraints, users might be positively biased when rating the system, indirectly making the reliability of the satisfaction data questionable. Machine learning experiments were conducted on two datasets, users and annotators, which were then compared in order to assess the reliability of these datasets. Our results indicated that standard classifiers were significantly more successful in discriminating the abovementioned emotions and their intensities (reflected by user satisfaction ratings) from annotator data than from user data. These results corroborated that: first, satisfaction data could be used directly as an alternative target variable to model affect, and that they could be predicted exclusively by dialog features. Second, these were only true when trying to predict the abovementioned emotions using annotator?s data, suggesting that user bias does exist in a laboratory-led evaluation.

Smart buildings for healthy and sustainable workplace: scoping study report

Deficiencies in the design and operation of office buildings can give rise to high social, environmental and economic (triple bottom line) costs. As a result, there are significant pressures and incentives to develop ‘smart building’ technologies that can facilitate improved indoor environment quality (IEQ), and more energy efficient operation of office buildings. IEQ indicators include lighting, ventilation, thermal comfort, indoor air quality and noise. In response to this, the CRC for Construction Innovation commissioned a six-month scoping study (Project no. 2002-043) to examine how different technologies could be used to improve the ‘triple bottom line’ for office buildings. The study was supported by three industry partners, Bovis Lend Lease, Arup, and The Queensland Department of Public Works. The objective of the study was to look at the history, trends, drivers, new technologies and potential application areas related to the operation of healthy and efficient office buildings. The key output from the study was a recommendation for a prototype system for intelligent monitoring and control of an office environment, based on identified market, technical and user requirements and constraints.

Urban informatics, ubiquitous computing and social media for healthy cities

The increasing ubiquity of digital technology, internet services and location-aware applications in our everyday lives allows for a seamless transitioning between the visible and the invisible infrastructure of cities: road systems, building complexes, information and communication technology, and people networks create a buzzing environment that is alive and exciting. Driven by curiosity, initiative and interdisciplinary exchange, the Urban Informatics Research Lab at Queensland University of Technology (QUT), Brisbane, Australia, is an emerging cluster of people interested in research and development at the intersection of people, place and technology with a focus on cities, locative media and mobile technology. This paper introduces urban informatics as a transdisciplinary practice across people, place and technology that can aid local governments, urban designers and planners in creating responsive and inclusive urban spaces and nurturing healthy cities. Three challenges are being discussed. First, people, and the challenge of creativity explores the opportunities and challenges of urban informatics that can lead to the design and development of new tools, methods and applications fostering participation, the democratisation of knowledge, and new creative practices. Second, technology, and the challenge of innovation examines how urban informatics can be applied to support user-led innovation with a view to promote entrepreneurial ideas and creative industries. Third, place, and the challenge of engagement discusses the potential to establish places within cities that are dedicated to place-based applications of urban informatics with a view to deliver community and civic engagement strategies.

User preferences for text message-delivered skin cancer prevention and early detection

Evidence is needed for the acceptability and user preferences of receiving skin cancer-related text messages. We prepared 27 questions to evaluate attitudes, satisfaction with program characteristics such as timing and spacing, and overall satisfaction with the Healthy Text program in young adults. Within this randomised controlled trial (age 18-42 years), 546 participants were assigned to one of three Healthy Text message groups; sun protection, skin self-examination, or attention-control. Over a 12-month period, 21 behaviour-specific text messages were sent to each group. Participants’ preferences were compared between the two interventions and control group at the 12-month follow-up telephone interview. In all three groups, participants reported the messages were easy to understand (98%), provided good suggestions or ideas (88%), and were encouraging (86%) and informative (85%) with little difference between the groups. The timing of the texts was received positively (92%); however, some suggestions for frequency or time of day the messages were received from 8% of participants. Participants in the two intervention groups found their messages more informative, and triggering behaviour change compared to control. Text messages about skin cancer prevention and early detection are novel and acceptable to induce behaviour change in young adults.

Heritability of head motion during resting state functional MRI in 462 healthy twins

Head motion (HM) is a critical confounding factor in functional MRI. Here we investigate whether HM during resting state functional MRI (RS-fMRI) is influenced by genetic factors in a sample of 462 twins (65% fema≤ 101 MZ (monozygotic) and 130 DZ (dizygotic) twin pairs; mean age: 21 (SD=3.16), range 16-29). Heritability estimates for three HM components-mean translation (MT), maximum translation (MAXT) and mean rotation (MR)-ranged from 37 to 51%. We detected a significant common genetic influence on HM variability, with about two-thirds (genetic correlations range 0.76-1.00) of the variance shared between MR, MT and MAXT. A composite metric (HM-PC1), which aggregated these three, was also moderately heritable (h2=42%). Using a sub-sample (N=35) of the twins we confirmed that mean and maximum translational and rotational motions were consistent "traits" over repeated scans (r=0.53-0.59); reliability was even higher for the composite metric (r=0.66). In addition, phenotypic and cross-trait cross-twin correlations between HM and resting state functional connectivities (RS-FCs) with Brodmann areas (BA) 44 and 45, in which RS-FCs were found to be moderately heritable (BA44: h2-=0.23 (sd=0.041), BA45: h2-=0.26 (sd=0.061)), indicated that HM might not represent a major bias in genetic studies using FCs. Even so, the HM effect on FC was not completely eliminated after regression. HM may be a valuable endophenotype whose relationship with brain disorders remains to be elucidated.

On food security and alternative food networks: Understanding and performing food security in the context of urban bias

This paper offers one explanation for the institutional basis of food insecurity in Australia, and argues that while alternative food networks and the food sovereignty movement perform a valuable function in building forms of social solidarity between urban consumers and rural producers, they currently make only a minor contribution to Australia’s food and nutrition security. The paper begins by identifying two key drivers of food security: household incomes (on the demand side) and nutrition-sensitive, ‘fair food’ agriculture (on the supply side). We focus on this second driver and argue that healthy populations require an agricultural sector that delivers dietary diversity via a fair and sustainable food system. In order to understand why nutrition-sensitive, fair food agriculture is not flourishing in Australia we introduce the development economics theory of urban bias. According to this theory, governments support capital intensive rather than labour intensive agriculture in order to deliver cheap food alongside the transfer of public revenues gained from rural agriculture to urban infrastructure, where the majority of the voting public resides. We chart the unfolding of the Urban Bias across the twentieth century and its consolidation through neo-liberal orthodoxy, and argue that agricultural policies do little to sustain, let alone revitalize, rural and regional Australia. We conclude that by observing food system dynamics through a re-spatialized lens, Urban Bias Theory is valuable in highlighting rural–urban socio-economic and political economy tensions, particularly regarding food system sustainability. It also sheds light on the cultural economy tensions for alternative food networks as they move beyond niche markets to simultaneously support urban food security and sustainable rural livelihoods.

Comparison of measured sleeping metabolic rate and predicted basal metabolic rate during the first year of life : evidence of a bias changing with increasing metabolic rate

Objective: To compare measurements of sleeping metabolic rate (SMR) in infancy with predicted basal metabolic rate (BMR) estimated by the equations of Schofield. Methods: Some 104 serial measurements of SMR by indirect calorimetry were performed in 43 healthy infants at 1.5, 3, 6, 9 and 12 months of age. Predicted BMR was calculated using the weight only (BMR-wo) and weight and height (BMR-wh) equations of Schofield for 0-3-y-olds. Measured SMR values were compared with both predictive values by means of the Bland-Altman statistical test. Results: The mean measured SMR was 1.48 MJ/day. The mean predicted BMR values were 1.66 and 1.47 MJ/day for the weight only and weight and height equations, respectively. The Bland-Altman analysis showed that BMR-wo equation on average overestimated SMR by 0.18 MJ/day (11%) and the BMR-wh equation underestimated SMR by 0.01 MJ/day (1%). However the 95% limits of agreement were wide: -0.64 to + 0.28 MJ/day (28%) for the former equation and -0.39 to + 0.41 MJ/day (27%) for the latter equation. Moreover there was a significant correlation between the mean of the measured and predicted metabolic rate and the difference between them. Conclusions: The wide variation seen in the difference between measured and predicted metabolic rate and the bias probably with age indicates there is a need to measure actual metabolic rate for individual clinical care in this age group.

Effects of the user model on simulation-based learning of dialogue strategies

Over the past decade, a variety of user models have been proposed for user simulation-based reinforcement-learning of dialogue strategies. However, the strategies learned with these models are rarely evaluated in actual user trials and it remains unclear how the choice of user model affects the quality of the learned strategy. In particular, the degree to which strategies learned with a user model generalise to real user populations has not be investigated. This paper presents a series of experiments that qualitatively and quantitatively examine the effect of the user model on the learned strategy. Our results show that the performance and characteristics of the strategy are in fact highly dependent on the user model. Furthermore, a policy trained with a poor user model may appear to perform well when tested with the same model, but fail when tested with a more sophisticated user model. This raises significant doubts about the current practice of learning and evaluating strategies with the same user model. The paper further investigates a new technique for testing and comparing strategies directly on real human-machine dialogues, thereby avoiding any evaluation bias introduced by the user model. © 2005 IEEE.

Polymorphims in the Interferon-y/Interleukin 26 Gene Region Contribute to Sex Bias in Susceptibility to Rheumatoid Arthritis

Objective:

To determine whether polymorphisms in the interferon-? (IFN?)/interleukin-26 (IL-26; formerly, AK155) gene cluster contribute to sex-based differential susceptibility to rheumatoid arthritis (RA).

Methods:

Four microsatellite markers, located in a 118-kb interval that contains both the IFN? and IL-26 genes on chromosome 12q15, were typed in 251 patients with RA and 198 unrelated healthy controls (all of whom lived in Northern Ireland) by means of polymerase chain reaction–based fragment analysis.

Results:

Marker D12S2510, which is located 3 kb 3' from the IL-26 gene, was significantly associated with RA in women (corrected P [Pcorr] = 0.008, 2 degrees of freedom [2 df]) but not in men (P = 0.99, 2 df). A 3-marker haplotype, IFNGCA*13;D12S2510*8;D12S2511*9, was inferred that showed significant underrepresentation in women with RA (odds ratio 0.50, 95% confidence interval 0.32–0.78; P = 0.002, Pcorr = 0.03) but not in men with RA.

Conclusion:

Our results demonstrate that common polymorphisms in the IFN?/IL-26 gene region may contribute to sex bias in susceptibility to RA, by distorting the propensity of female carriers versus male carriers to contract this disease. These results conform to our recent observations of a role for this gene cluster in sex-based differential susceptibility to another Th1-type inflammatory disease, multiple sclerosis.

Dopaminergic genotype biases spatial attention in healthy children

In everyday life, our sensory system is bombarded with visual input and we rely upon attention to select only those inputs that are relevant to behavioural goals. Typically, humans can shift their attention from one visual field to the other with little cost to perception. In cases of, unilateral neglect', however, there is a persistent bias of spatial attention towards the same side as the damaged cerebral hemisphere. We used a visual orienting task to examine the influence of functional polymorphisms of the dopamine transporter gene (DAT1) on individual differences in spatial attention in normally developing children. DAT1 genotype significantly influenced spatial bias. Healthy children who were homozygous for alleles that influence the expression of dopamine transporters in the brain displayed inattention for left-sided stimuli, whereas heterozygotes did not. Our data provide the first evidence in healthy individuals of a genetically mediated bias in spatial attention that is related to dopamine signalling.