Search

SocInfo 2018: Social Informatics: 10th International Conference on Social Informatics (St. Petersburg, Russia, September 25-28, 2018), Item does not contain fulltext, Human evacuation experiments can trigger distress, be unethical and present high costs. As a solution, computer simulations can predict the effectiveness of new emergency management procedures. This paper applies multi-agent simulation to measure the influence of staff members with diverse training levels on evacuation time. A previously developed and validated model was extended with explicit mechanisms to simulate staff members helping people to egress. The majority of parameter settings have been based on empirical data acquired in earlier studies. Therefore, simulation results are expected to be realistic. Results show that staff are more effective in complex environments, especially when trained. Not only specialised security professionals but, especially, regular workers of shopping facilities and offices play a significant role in evacuation processes when adequately trained. These results can inform policy makers and crowd managers on new emergency management procedures.

Contains fulltext : 191881.pdf (publisher's version ) (Closed access), The rapid development and increasing complexity of modern socio-technical systems suggest an urgent need for systemic safety analysis approaches because traditional linear models cannot cope with this complexity. In the aviation safety literature, among systemic accident and incident analysis methods, Systems Theoretic Accident Modelling and Processes (STAMP) and Agent-based modelling (ABM) are the most cited ones. STAMP is a qualitative analysis approach known for its thoroughness and comprehensiveness. Computational ABM approach is a formal quantitative method which proved to be suitable for modelling complex flexible systems. In addition, from a legal point of view, formal systemic institutional modelling potentially provides an interesting contribution to accident and incident analysis. The current work compares three systemic modelling approaches: STAMP, ABM and institutional modelling applied to a case study in an aviation domain.

MA3HMI 2018: Fourth International Workshop on Multimodal Analysis enabling Artificial Agents in Human-Agent Interaction, Boulder, CO, USA, October 16 - 20, 2018, Item does not contain fulltext, The role of emotions and other affective states within Human-Computer Interaction (HCI) is gaining importance. Introducing affect into computer applications typically makes these systems more efficient, effective and enjoyable. This paper presents a model that is able to extract interpersonal stance from vocal signals. To achieve this, a dataset of 3840 sentences spoken by 20 semi-professional actors was built and was used to train and test a model based on Support Vector Machines (SVM). An analysis of the results indicates that there is much variation in the way people express interpersonal stance, which makes it difficult to build a generic model. Instead, the model shows good performance on the individual level (with accuracy above 80%). The implications of these findings for HCI systems are discussed.

BNAIC 2018: 30th Benelux Conference on Artificial Intelligence, 's Hertogenbosch, The Netherlands, 8-9 November 2018, Item does not contain fulltext, To train public transport employees to deal with aggressive passengers, simulation-based training can be a cost-effective instrument. Within this research a virtual reality training application for bus drivers has been developed. The application consists of 10 different scenarios that can be used to practice how to act during confrontations with aggressive passengers. During each scenario, the user will have a conversation through text and pre-recorded sentences with an intelligent virtual agent. In addition, during the training the user will have to determine the kind of aggression that is shown by the passenger, choose the right communication style and decide whether or not to report the incident at the end of the scenario. This paper describes the full development of the application and the scenarios, as well as the first experiences reported during a pilot study. Suggestions for improvement and future work are discussed.

17th International Conference on Autonomous Agents and MultiAgent Systems (Stockholm, Sweden - July 10 - 15, 2018), Item does not contain fulltext, In this paper, an empirical study conducted with Amazon Mechanical Turk workers is presented that aims to make a correspondence between messages about stressful situations, which were shared via Twitter, and 5 different types of supportive replies to them. Around 10.00 tweets were collected and analyzed in the terms previously described. We performed statistical tests to determine possible correlations between causes of stress and supportive response strategies. We are about to use these findings to improve a previously implemented algorithm that automatically generates supportive messages to stressed users. This algorithm is the core of an agent, in the form of a chatbot, that would be able to interact with such stressed users as if it would belong to their social networks.

International Workshops of PAAMS 2018 (Toledo, Spain, June 20-22, 2018), Item does not contain fulltext, Doorstep scams are scams, often happening at the front door, in which a con artist has a convincing, but fraudulent, story with the purpose of coming into your house and/or stealing money. Various campaigns to educate people exist, but they do not focus on the verbal skills people can use to prevent themselves from becoming a victim. This paper describes the conceptual design of a proposed training application. This application will provide an agent-based learning environment for high-risk doorstep scam victims. In order to create a training application, field research has been done to the content and progress of doorstep scams, which is used to create interactive scenarios.

29th Benelux Conference, BNAIC 2017 (Groningen, The Netherlands, November 8-9, 2017), Revised Selected Papers, Item does not contain fulltext, Intelligent Virtual Agents (IVAs) become widely used for numerous applications, varying from healthcare decision support to communication training. In several of such applications, it is useful if IVAs have the ability to take a negative stance towards the user, for instance for anti-bullying or conflict management training. However, the believability of such 'virtual bad guys' is often limited, since they are non-consequential, i.e., are unable to apply serious sanctions to users. To improve this situation, this research explores the potential of endowing IVAs with the ability to provide haptic feedback. This was realized by conducting an experiment in which users interact with a virtual agent that is able to physically 'touch' the user via a haptic gaming vest. The effect on the loudness of the speech and the subjective experience of the participants was measured. Results of the experiment suggest there might be an effect on the subjective experience of the participants and the loudness of their speech. Statistical analysis, however, shows no significant effect but due to the relatively small sample size it is advisable to further look into these aspects.

01 december 2018, Item does not contain fulltext, BACKGROUND: Non-invasive vulvar Paget disease (VPD) is a rare skin disorder mainly affecting elderly women. Recently, the immune modulator imiquimod was reported as an effective treatment option. Knowledge about the immune microenvironment of VPD is lacking. METHODS: This study investigates the basic characteristics of the immune infiltrate in VPD (n=10); moreover the influence of imiquimod was studied (n=6). Immunohistochemistry for CD4, CD8, CD14, CD20, CD56 and FoxP3 was performed. The infiltrates of VPD were compared to vulvar high-grade squamous cell intraepithelial lesions (HSIL) (n=43), a HPV induced vulvar premalignancy with known response to imiquimod cream, and healthy controls (n=30). Immune cell counts in samples taken before and after treatment were compared. RESULTS: The microenvironment in VPD differs from the healthy vulvar skin and vulvar HSIL. VPD is characterized by a decrease in immune cells in the epithelium and an abundant number of immune cells in the stroma, consisting predominantly of T cells. The intraepithelial CD8+/Foxp3+ ratio and number of CD56+ increased after imiquimod therapy, whereas the numbers of CD14+ cells decreased which may point to a treatment-induced type 1 immune response. CONCLUSIONS: The epithelium in VPD contains less immune cells, but a dense stromal immune infiltrate. Changes in immune cell counts after immune modulation in relation to clinical responses should be further investigated.

17th International Conference on Autonomous Agents and MultiAgent Systems (Stockholm, Sweden - July 10 - 15, 2018), Item does not contain fulltext, This paper introduces the concept of "virtual bad guys": intelligent virtual agents that take a negative or even aggressive stance towards the user. Although they pave the way to various interesting applications, it is hard to create virtual bad guys that are taken seriously by the user, since they are typically unable to apply serious sanctions. To address this issue, this study experimentally investigated the effect of "consequential" agents that are able to physically threaten their human interlocutors. A consequential agent was developed by equipping users with a (non-functioning) device, through which they were made to believe the agent could mildly shock them. Effects on participants' levels of anxiety and (physiological and self-reported) stress were measured, and the role of presence and perceived believability of the virtual agent was assessed. The consequential agent triggered a stronger physiological stress response than the non-consequential agent, whereas self-reported levels of anxiety and stress did not significantly differ. Furthermore, while presence and believability were substantially associated with users' stress response, both states did not mediate or explain the effect of a consequential vs. non-consequential agent on stress, as they did not significantly differ between conditions. Implications of these findings and suggestions for follow-up studies on "virtual bad guys" are discussed.

International Workshops of PAAMS 2018 (Toledo, Spain, June 20-22, 2018), Contains fulltext : 194171.pdf (Publisher’s version ) (Closed access), Human support is thought to increase adherence to internet-based interventions for common mental health disorders, but can be costly and reduce treatment accessibility. Embodied virtual agents may be used to deliver automated support, but while many solutions have been shown to be feasible, there is still little controlled research that empirically validates their clinical effectiveness in this context. This study uses a controlled and randomized paradigm to investigate whether feedback from an embodied virtual agent can increase adherence. In a three-week ecological momentary assessment smartphone study, 68 participants were asked to report their mood three times a day. An embodied virtual agent could mirror participant-reported mood states when thanking them for their answers. A two-stage randomization into a text and personalized visual feedback group, versus a text-only control group, was applied to control for individual differences (study onset) and feedback history (after two weeks). Results indicate that while personalized visual feedback did not increase adherence, it did manage to keep adherence constant over a three-week period, whereas fluctuations in adherence could be observed in the text-only control group. Although this was a pilot study, and its results should be interpreted with some caution, this paper shows how virtual agent feedback may have a stabilizing effect on adherence, how controlled experiments on the relationship between virtual agent support and clinically relevant measures such as adherence can be conducted, and how results may be analyzed.

BNAIC 2018: 30th Benelux Conference on Artificial Intelligence, 's Hertogenbosch, The Netherlands, 8-9 November 2018, Item does not contain fulltext

SocInfo 2018: Social Informatics: 10th International Conference on Social Informatics (St. Petersburg, Russia, September 25-28, 2018), Item does not contain fulltext, In this paper, a multi-agent simulation model is presented to investigate the dynamics of 'mutual-support networks': online social networks consisting of both humans and artificial agents. Via such networks, human users who are coping with stress can share their problems, via text messages, with human peers as well as 'artificial friends'. Even though not everybody feels comfortable sharing personal problems with artificial agents, a bot is always available to help human users, and does not face any negative consequence of providing help to stressed peers. Using the simulation model, the dynamics of social networks consisting of an arbitrary combination of humans and agents have been explored under various circumstances. This exploration resulted in several insights that are useful for shaping our vision on artificial friends: (1) humans can provide less emotional support than artificial agents because they have limited emotional resources, (2) the type of support that is provided has a large impact on the human’s stress level, and (3) the more open users are to receiving automated support, the more effective the support is in reducing their stress level. The model was internally validated by means of a mathematical verification.

Contains fulltext : 191917.pdf (publisher's version ) (Open Access), In this study, we investigate if a digital coach for low-literate learners that provides cognitive learning support based on scaffolding can be improved by adding affective learning support based on motivational interviewing, and social learning support based on small talk. Several knowledge gaps are identified: motivational interviewing and small talk must be translated to control rules for this coach, a formal model of participant emotional states is needed to allow the coach to parse the learner's emotional state, and various sensors must be used to let the coach detect and act on this state. We use the situated Cognitive Engineering (sCE) method to update an existing foundation of knowledge with emotional models, motivational interviewing, and small talk theory, technology, and a new exercise in the volunteer work domain. We use this foundation to create a design specification for an Embodied Conversational Agent (ECA) coach that provides cognitive, affective, and social learning support for this exercise. A prototype is created, and compared to a prototype that only provides cognitive support in a within- and between-subjects experiment. Results show that both prototypes work as expected: learners interact with the coach and complete all exercises. Almost no significant differences are found between the two prototypes, indicating that the affective and social support were not effective as designed. Potential improvements are provided for future work. Results also show significant differences between two subgroups of low-literate participants, and between men and women, reinforcing the importance of using individualized support measures with this demographic.

16th International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2018 (Toledo, Spain, June 20–22, 2018), Item does not contain fulltext, This exploratory study discusses how biofeedback can be displayed and used in the context of virtual agent-based social skills training. Using Virtual Reality, a prototype of an aggression de-escalation training application has been developed. The application simulates an adaptive scenario in which the user interacts with a threatening virtual conversation partner. The stress level of the user, which is displayed at run-time, influences the development of the scenario. A pilot experiment is conducted to test the impact of the application on users’ subjective experience (measured through a questionnaire) and physiological response (measured through skin conductance). Although preliminary, the results seem to indicate that biofeedback potentially contributes to enhancing the user’s awareness of his/her own emotional state. Implications for the use of biofeedback within virtual training applications are discussed.

International Workshops of PAAMS 2018 (Toledo, Spain, June 20-22, 2018), Item does not contain fulltext, This paper describes several steps towards the development of an online agent (or socialbot) that provides emotional support to stressed users. In particular, we present an empirical study that was conducted with the aim to investigate how people help each other to cope with stressful situations via online social networks. To this end, around 10.000 tweets about stressful situations were collected. Then, using crowdsourcing, these tweets were classified into stress categories, and supportive replies to them were collected, which were also classified into categories. Contingency tables were constructed in order to explore which types of support were most frequently used in which circumstances. The resulting values can be used as parameters for a previously developed algorithm that automatically constructs support messages. This allows our agent to generate supportive messages that are more similar to the support messages that human beings send via social media.

BNAIC 2018: 30th Benelux Conference on Artificial Intelligence, 's Hertogenbosch, The Netherlands, 8-9 November 2018, Item does not contain fulltext

Contains fulltext : 191982.pdf (publisher's version ) (Open Access), Serious games are increasingly being used for training of social skills. The main idea is to create a virtual environment in which a trainee can interact with graphically embodied virtual characters. By designing scenarios in such way that the character’s behaviour provides direct feedback on the correctness of the trainee’s choices, an interactive learning experience is created. This paper explores the potential of this approach in the domain of law enforcement. A prototype has been developed of a serious game that enables police academy students to train their communicative skills. A pilot study with 41 students has been conducted. The results show that this is a promising instrument for education in this domain, but also point out several suggestions for improvement.

Background: PORTEC-2 was a randomised trial for women with high-intermediate risk (HIR) endometrial cancer, comparing pelvic external beam radiotherapy (EBRT) with vaginal brachytherapy (VBT). We evaluated long-term outcomes combined with the results of pathology review and molecular analysis. Methods: 427 women with HIR endometrial cancer were randomised between 2002–2006 to VBT or EBRT. Primary endpoint was vaginal recurrence (VR). Pathology review was done in 97.4%, combined with molecular analysis. Results: Median follow-up was 116 months; 10-year VR was 3.4% versus 2.4% for VBT vs. EBRT (p = 0.55). Ten-year pelvic recurrence (PR) was more frequent in the VBT group (6.3% vs. 0.9%, p = 0.004), mostly combined with distant metastases (DM). Ten-year isolated PR was 2.5% vs. 0.5%, p = 0.10, and DM 10.4 vs. 8.9% (p = 0.45). Overall survival for VBT vs. EBRT was 69.5% vs. 67.6% at 10 years (p = 0.72). L1CAM and p53-mutant expression and substantial lymph-vascular space invasion were risk factors for PR and DM. EBRT reduced PR in cases with these risk factors. Conclusion: Long-term results of the PORTEC-2 trial confirm VBT as standard adjuvant treatment for HIR endometrial cancer. Molecular risk assessment has the potential to guide adjuvant therapy. EBRT provided better pelvic control in patients with unfavourable risk factors.

BRCA1/2 variant analysis in tumor tissue could streamline the referral of patients with epithelial ovarian, fallopian tube, or primary peritoneal cancer to genetic counselors and select patients who benefit most from targeted treatment. We investigated the sensitivity of BRCA1/2 variant analysis in formalin-fixed, paraffin-embedded tumor tissue using a combination of next-generation sequencing and copy number variant multiplex ligation-dependent probe amplification. After optimization using a training cohort of known BRCA1/2 mutation carriers, validation was performed in a prospective cohort in which screening of BRCA1/2 tumor DNA and leukocyte germline DNA was performed in parallel. BRCA1 promoter hypermethylation and pedigree analysis were also performed. In the training cohort, 45 of 46 germline BRCA1/2 variants were detected (sensitivity, 98%). In the prospective cohort (n = 62), all six germline variants were identified (sensitivity, 100%), together with five somatic BRCA1/2 variants and eight cases with BRCA1 promoter hypermethylation. In four BRCA1/2 variant–negative patients, surveillance or prophylactic management options were offered on the basis of positive family histories. We conclude that BRCA1/2 formalin-fixed, paraffin-embedded tumor tissue analysis reliably detects BRCA1/2 variants. When taking family history of BRCA1/2 variant–negative patients into account, tumor BRCA1/2 variant screening allows more efficient selection of epithelial ovarian cancer patients for genetic counseling and simultaneously selects patients who benefit most from targeted treatment.

BACKGROUND: While it has been challenging to establish prostate cancer patient-derived xenografts (PDXs), with a take rate of 10-40% and long latency time, multiple groups throughout the world have developed methods for the successful establishment of serially transplantable human prostate cancer PDXs using a variety of immune deficient mice. In 2014, the Movember Foundation launched a Global Action Plan 1 (GAP1) project to support an international collaborative prostate cancer PDX program involving eleven groups. Between these Movember consortium members, a total of 98 authenticated human prostate cancer PDXs were available for characterization. Eighty three of these were derived directly from patient material, and 15 were derived as variants of patient-derived material via serial passage in androgen deprived hosts. A major goal of the Movember GAP1 PDX project was to provide the prostate cancer research community with a summary of both the basic characteristics of the 98 available authenticated serially transplantable human prostate cancer PDX models and the appropriate contact information for collaborations. Herein, we report a summary of these PDX models. METHODS: PDX models were established in immunocompromised mice via subcutaneous or subrenal-capsule implantation. Dual-label species (ie, human vs mouse) specific centromere and telomere Fluorescence In Situ Hybridization (FISH) and immuno-histochemical (IHC) staining of tissue microarrays (TMAs) containing replicates of the PDX models were used for characterization of expression of a number of phenotypic markers important for prostate cancer including AR (assessed by IHC and FISH), Ki67, vimentin, RB1, P-Akt, chromogranin A (CgA), p53, ERG, PTEN, PSMA, and epithelial cytokeratins. RESULTS: Within this series of PDX models, the full spectrum of clinical disease stages is represented, including androgen-sensitive and castration-resistant primary and metastatic prostate adenocarcinomas as well as prostate carcino

30th International Conference on Industrial Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2017, Arras, France, June 27-30, 2017, Item does not contain fulltext, A fire incident at a transport hub can cost many lives. To save lives, effective crisis management and prevention measures need to be taken. In this project, the effect of cultural factors in managing and preventing emergencies in public transport systems is analysed. An agent–based model of an evacuating crowd was created. Socio-cultural factors that were modelled are: familiarity with environment, response time and social contagion of fear and beliefs about the situation. Simulation results show that (1) familiarity and social contagion decrease evacuation time, while increasing the number of falls; (2) crowd density and social contagion increase evacuation time and falls. All three factors show different effects on the response time. This model will be used by transport operators to estimate the effect of these socio-cultural factors and prepare for emergencies.

19th International Conference on Human-Computer Interaction, HCI International 2017, Vancouver, BC, Canada, July 9-14, 2017, Item does not contain fulltext, This paper presents an algorithm to automatically detect interpersonal stance in vocal signals. The focus is on two stances (referred to as 'Dominant' and 'Empathic') that play a crucial role in aggression de-escalation. To develop the algorithm, first a database was created with more than 1000 samples from 8 speakers from different countries. In addition to creating the algorithm, a detailed analysis of the samples was performed, in an attempt to relate interpersonal stance to emotional state. Finally, by means of an experiment via Mechanical Turk, the performance of the algorithm was compared with the performance of human beings. The resulting algorithm provides a useful basis to develop computer-based support for interpersonal skills training.

Transactions on Computational Collective Intelligence XXVII, Item does not contain fulltext, In this research, the effects of culture, cognitions, and emotions on crisis management and prevention are analysed. An agent-based crowd evacuation simulation model was created, named IMPACT, to study the evacuation process from a transport hub. To extend previous research, various socio-cultural, cognitive, and emotional factors were modelled, including: language, gender, familiarity with the environment, emotional contagion, prosocial behaviour, falls, group decision making, and compliance. The IMPACT model was validated against data from an evacuation drill using the existing EXODUS evacuation model. Results show that on all measures, the IMPACT model is within or close to the prescribed boundaries, thereby establishing its validity. Structured simulations with the validated model revealed important findings, including: the effect of doors as bottlenecks, social contagion speeding up evacuation time, falling behaviour not affecting evacuation time significantly, and travelling in groups being more beneficial for evacuation time than travelling alone. This research has important practical applications for crowd management professionals, including transport hub operators, first responders, and risk assessors.