Comparing segmentation strategies for efficient video passage retrieval

We compare the effect of different text segmentation strategies on speech based passage retrieval of video. Passage retrieval has mainly been studied to improve document retrieval and to enable question answering. In these domains best results were obtained using passages defined by the paragraph structure of the source documents or by using arbitrary overlapping passages. For the retrieval of relevant passages in a video, using speech transcripts, no author defined segmentation is available. We compare retrieval results from 4 different types of segments based on the speech channel of the video: fixed length segments, a sliding window, semantically coherent segments and prosodic segments. We evaluated the methods on the corpus of the MediaEval 2011 Rich Speech Retrieval task. Our main conclusion is that the retrieval results highly depend on the right choice for the segment length. However, results using the segmentation into semantically coherent parts depend much less on the segment length. Especially, the quality of fixed length and sliding window segmentation drops fast when the segment length increases, while quality of the semantically coherent segments is much more stable. Thus, if coherent segments are defined, longer segments can be used and consequently less segments have to be considered at retrieval time.

Are Deep Learning Approaches Suitable for Natural Language Processing?

In recent years, Deep Learning (DL) techniques have gained much at-tention from Artificial Intelligence (AI) and Natural Language Processing (NLP) research communities because these approaches can often learn features from data without the need for human design or engineering interventions. In addition, DL approaches have achieved some remarkable results. In this paper, we have surveyed major recent contributions that use DL techniques for NLP tasks. All these reviewed topics have been limited to show contributions to text understand-ing, such as sentence modelling, sentiment classification, semantic role labelling, question answering, etc. We provide an overview of deep learning architectures based on Artificial Neural Networks (ANNs), Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and Recursive Neural Networks (RNNs).

How level and type of deafness affects user perception of multimedia video clips

Our research investigates the impact that hearing has on the perception of digital video clips, with and without captions, by discussing how hearing loss, captions and deafness type affects user QoP (Quality of Perception). QoP encompasses not only a user's satisfaction with the quality of a multimedia presentation, but also their ability to analyse, synthesise and assimilate informational content of multimedia . Results show that hearing has a significant effect on participants’ ability to assimilate information, independent of video type and use of captions. It is shown that captions do not necessarily provide deaf users with a ‘greater level of information’ from video, but cause a change in user QoP, depending on deafness type, which provides a ‘greater level of context of the video’. It is also shown that post-lingual mild and moderately deaf participants predict less accurately their level of information assimilation than post-lingual profoundly deaf participants, despite residual hearing. A positive correlation was identified between level of enjoyment (LOE) and self-predicted level of information assimilation (PIA), independent of hearing level or hearing type. When this is considered in a QoP quality framework, it puts into question how the user perceives certain factors, such as ‘informative’ and ‘quality’.

En jämförelse av molnbaserade realtidsanalystjänster : Skillnader i användbarhet mellan multi-tenant och single-tenant arkitekturer

Internet of Things är ett samlingsbegrepp för den utveckling som innebär att olika typer av enheter kan förses med sensorer och datachip som är uppkopplade mot internet. En ökad mängd data innebär en ökad förfrågan på lösningar som kan lagra, spåra, analysera och bearbeta data. Ett sätt att möta denna förfrågan är att använda sig av molnbaserade realtidsanalystjänster. Multi-tenant och single-tenant är två typer av arkitekturer för molnbaserade realtidsanalystjänster som kan användas för att lösa problemen med hanteringen av de ökade datamängderna. Dessa arkitekturer skiljer sig åt när det gäller komplexitet i utvecklingen. I detta arbete representerar Azure Stream Analytics en multi-tenant arkitektur och HDInsight/Storm representerar en single-tenant arkitektur. För att kunna göra en jämförelse av molnbaserade realtidsanalystjänster med olika arkitekturer, har vi valt att använda oss av användbarhetskriterierna: effektivitet, ändamålsenlighet och användarnöjdhet. Vi kom fram till att vi ville ha svar på följande frågor relaterade till ovannämnda tre användbarhetskriterier: • Vilka likheter och skillnader kan vi se i utvecklingstider? • Kan vi identifiera skillnader i funktionalitet? • Hur upplever utvecklare de olika analystjänsterna? Vi har använt en design and creation strategi för att utveckla två Proof of Concept prototyper och samlat in data genom att använda flera datainsamlingsmetoder. Proof of Concept prototyperna inkluderade två artefakter, en för Azure Stream Analytics och en för HDInsight/Storm. Vi utvärderade dessa genom att utföra fem olika scenarier som var för sig hade 2-5 delmål. Vi simulerade strömmande data genom att låta en applikation kontinuerligt slumpa fram data som vi analyserade med hjälp av de två realtidsanalystjänsterna. Vi har använt oss av observationer för att dokumentera hur vi arbetade med utvecklingen av analystjänsterna samt för att mäta utvecklingstider och identifiera skillnader i funktionalitet. Vi har även använt oss av frågeformulär för att ta reda på vad användare tyckte om analystjänsterna. Vi kom fram till att Azure Stream Analytics initialt var mer användbart än HDInsight/Storm men att skillnaderna minskade efter hand. Azure Stream Analytics var lättare att arbeta med vid simplare analyser medan HDInsight/Storm hade ett bredare val av funktionalitet.

Fundo de investimento imobiliário: metodologia para subsidiar o investidor a formar uma carteira eficiente

O presente trabalho tem por objetivo subsidiar o investidor de Fundos de Investimento Imobiliário na escolha de uma carteira de aplicação de FIIs, visando obter performance igual ou superior ao índice de referência do setor (IFIX). Tal subsídio é constituído, inicialmente, por uma metodologia que considera que o conceito de Carteira Eficiente (Risco/Retorno) preconizada por Markowitz pode trabalhar em conjunto com a dimensão do conceito das Finanças Comportamentais, liderada por Daniel Kahneman, constituindo as bases de orientação do investidor. Acrescentamos o caminho metodológico com as indicações, sugeridas por Bazerman e Moore, no processo de tomada de decisão, que reduza os efeitos de heurísticas e vieses.

Hudumia: a platform to support spontaneous volunteers' situational awareness, coordination and safety underemergency scenarios

Spontaneous volunteers always emerge under emergency scenarios and are vital to a successful community response, yet some uncertainty subsists around their role and its inherent acceptance by official entities under emergency scenarios. In our research we have identified that most of the spontaneous volunteers do have none or little support from official entities, hence they end up facing critical problems as situational awareness, safety instructions and guidance, motivation and group organization. We argue that official entities still play a crucial role and should change some of their behaviors regarding spontaneous volunteerism. We aim with this thesis to design a software architecture and a framework in order to implement a solution to support spontaneous volunteerism under emergency scenarios along with a set of guidelines for the design of open information management systems. Together with the collaboration from both citizens and emergency professionals we have been able to attain several important contributions, as the clear identification of the roles taken by both spontaneous volunteers and professionals, the importance of volunteerism in overall community response and the role which open collaborative information management systems have in the community volunteering efforts. These conclusions have directly supported the design guidelines of our software solution proposal. In what concerns to methodology, we first review literature on technologies support to emergencies and how spontaneous volunteers actually challenge these systems. Following, we have performed a field research where we have observed that the emerging of spontaneous volunteer’s efforts imposes new requirements for the design of such systems, which leaded to the creation of a cluster of design guidelines that supported our software solution proposal to address the volunteers’ requirements. Finally we have architected and developed an online open information management tool which has been evaluated via usability engineering methods, usability user tests and heuristic evaluations.

Online Shopping Websites: An Evaluation of User Experience and Interface Ergonomic Criteria from the Perspective of Older Users

This paper presents a user experience evaluation of two online shopping websites from the perspective of older users (those aged 50 and older). Two online shopping websites were evaluated using methodological procedures established in prior research [1]. The methodology consists of four steps: (1) heuristic interface evaluation using an ergonomic criteria checklist, (2) online identification and experience questionnaire, (3) evaluation of user experience and interface interaction, and (4) satisfaction questionnaire. Results of the study revealed the analyzed websites are not suitable for older users, who find it difficult to interact with these interfaces.

A feasibility study for the provision of electronic healthcare tools and services in areas of Greece, Cyprus and Italy

Background Through this paper, we present the initial steps for the creation of an integrated platform for the provision of a series of eHealth tools and services to both citizens and travelers in isolated areas of thesoutheast Mediterranean, and on board ships travelling across it. The platform was created through an INTERREG IIIB ARCHIMED project called INTERMED. Methods The support of primary healthcare, home care and the continuous education of physicians are the three major issues that the proposed platform is trying to facilitate. The proposed system is based on state-of-the-art telemedicine systems and is able to provide the following healthcare services: i) Telecollaboration and teleconsultation services between remotely located healthcare providers, ii) telemedicine services in emergencies, iii) home telecare services for "at risk" citizens such as the elderly and patients with chronic diseases, and iv) eLearning services for the continuous training through seminars of both healthcare personnel (physicians, nurses etc) and persons supporting "at risk" citizens. These systems support data transmission over simple phone lines, internet connections, integrated services digital network/digital subscriber lines, satellite links, mobile networks (GPRS/3G), and wireless local area networks. The data corresponds, among others, to voice, vital biosignals, still medical images, video, and data used by eLearning applications. The proposed platform comprises several systems, each supporting different services. These were integrated using a common data storage and exchange scheme in order to achieve system interoperability in terms of software, language and national characteristics. Results The platform has been installed and evaluated in different rural and urban sites in Greece, Cyprus and Italy. The evaluation was mainly related to technical issues and user satisfaction. The selected sites are, among others, rural health centers, ambulances, homes of "at-risk" citizens, and a ferry. Conclusions The results proved the functionality and utilization of the platform in various rural places in Greece, Cyprus and Italy. However, further actions are needed to enable the local healthcare systems and the different population groups to be familiarized with, and use in their everyday lives, mature technological solutions for the provision of healthcare services.

How do we teach usability? An Investigation of Usability Instruction in Technical Communication

This dissertation investigates the curricular implementation of usability instruction in technical communication. Though there are a plethora of publications and studies on usability in technical communication, little discussion focuses on usability instruction in the classroom or its implementation in the curriculum. Thus, this exploratory qualitative research seeks to contribute to a better understanding about technical communication students' and instructors' knowledge of and experiences with usability practices in the classroom, the challenges that impacted their usability efforts, and their recommendations on how their efforts could be improved. The study results demonstrate the need for more productive discussion on this issue and for developing more effective strategies for implementing usability in the classroom.

User-centric need-driven affect modeling for spoken conversational agents: design and evaluation

It is easy to get frustrated at spoken conversational agents (SCAs), perhaps because they seem to be callous. By and large, the quality of human-computer interaction is affected due to the inability of the SCAs to recognise and adapt to user emotional state. Now with the mass appeal of artificially-mediated communication, there has been an increasing need for SCAs to be socially and emotionally intelligent, that is, to infer and adapt to their human interlocutors’ emotions on the fly, in order to ascertain an affective, empathetic and naturalistic interaction. An enhanced quality of interaction would reduce users’ frustrations and consequently increase their satisfactions. These reasons have motivated the development of SCAs towards including socio-emotional elements, turning them into affective and socially-sensitive interfaces. One barrier to the creation of such interfaces has been the lack of methods for modelling emotions in a task-independent environment. Most emotion models for spoken dialog systems are task-dependent and thus cannot be used “as-is” in different applications. This Thesis focuses on improving this, in which it concerns computational modeling of emotion, personality and their interrelationship for task-independent autonomous SCAs. The generation of emotion is driven by needs, inspired by human’s motivational systems. The work in this Thesis is organised in three stages, each one with its own contribution. The first stage involved defining, integrating and quantifying the psychological-based motivational and emotional models sourced from. Later these were transformed into a computational model by implementing them into software entities. The computational model was then incorporated and put to test with an existing SCA host, a HiFi-control agent. The second stage concerned automatic prediction of affect, which has been the main challenge towards the greater aim of infusing social intelligence into the HiFi 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. In this stage, we attempted to address part of this challenge by considering the roles of user satisfaction ratings and conversational/dialog features as the respective target and predictors in discriminating contentment and frustration, two types of emotions that are known to be prevalent within spoken human-computer interaction. The final stage concerned the evaluation of the emotional model through the HiFi agent. A series of user studies with 70 subjects were conducted in a real-time environment, each in a different phase and with its own conditions. All the studies involved the comparisons between the baseline non-modified and the modified agent. The findings have gone some way towards enhancing our understanding of the utility of emotion in spoken dialog systems in several ways; first, an SCA should not express its emotions blindly, albeit positive. Rather, it should adapt its emotions to user states. Second, low performance in an SCA may be compensated by the exploitation of emotion. Third, the expression of emotion through the exploitation of prosody could better improve users’ perceptions of an SCA compared to exploiting emotions through just lexical contents. Taken together, these findings not only support the success of the emotional model, but also provide substantial evidences with respect to the benefits of adding emotion in an SCA, especially in mitigating users’ frustrations and ultimately improving their satisfactions. Resumen Es relativamente fácil experimentar cierta frustración al interaccionar con agentes conversacionales (Spoken Conversational Agents, SCA), a menudo porque parecen ser un poco insensibles. En general, la calidad de la interacción persona-agente se ve en cierto modo afectada por la incapacidad de los SCAs para identificar y adaptarse al estado emocional de sus usuarios. Actualmente, y debido al creciente atractivo e interés de dichos agentes, surge la necesidad de hacer de los SCAs unos seres cada vez más sociales y emocionalmente inteligentes, es decir, con capacidad para inferir y adaptarse a las emociones de sus interlocutores humanos sobre la marcha, de modo que la interacción resulte más afectiva, empática y, en definitiva, natural. Una interacción mejorada en este sentido permitiría reducir la posible frustración de los usuarios y, en consecuencia, mejorar el nivel de satisfacción alcanzado por los mismos. Estos argumentos justifican y motivan el desarrollo de nuevos SCAs con capacidades socio-emocionales, dotados de interfaces afectivas y socialmente sensibles. Una de las barreras para la creación de tales interfaces ha sido la falta de métodos de modelado de emociones en entornos independientes de tarea. La mayoría de los modelos emocionales empleados por los sistemas de diálogo hablado actuales son dependientes de tarea y, por tanto, no pueden utilizarse "tal cual" en diferentes dominios o aplicaciones. Esta tesis se centra precisamente en la mejora de este aspecto, la definición de modelos computacionales de las emociones, la personalidad y su interrelación para SCAs autónomos e independientes de tarea. Inspirada en los sistemas motivacionales humanos en el ámbito de la psicología, la tesis propone un modelo de generación/producción de la emoción basado en necesidades. El trabajo realizado en la presente tesis está organizado en tres etapas diferenciadas, cada una con su propia contribución. La primera etapa incluyó la definición, integración y cuantificación de los modelos motivacionales de partida y de los modelos emocionales derivados a partir de éstos. Posteriormente, dichos modelos emocionales fueron plasmados en un modelo computacional mediante su implementación software. Este modelo computacional fue incorporado y probado en un SCA anfitrión ya existente, un agente con capacidad para controlar un equipo HiFi, de alta fidelidad. La segunda etapa se orientó hacia el reconocimiento automático de la emoción, aspecto que ha constituido el principal desafío en relación al objetivo mayor de infundir inteligencia social en el agente HiFi. En los últimos años, los estudios sobre reconocimiento de emociones a partir de la voz han pasado de emplear datos actuados a usar datos reales en los que la presencia u observación de emociones se produce de una manera mucho más sutil. El reconocimiento de emociones bajo estas condiciones resulta mucho más complicado y esta dificultad se pone de manifiesto en tareas tales como el etiquetado y el aprendizaje automático. En esta etapa, se abordó el problema del reconocimiento de las emociones del usuario a partir de características o métricas derivadas del propio diálogo usuario-agente. Gracias a dichas métricas, empleadas como predictores o indicadores del grado o nivel de satisfacción alcanzado por el usuario, fue posible discriminar entre satisfacción y frustración, las dos emociones prevalentes durante la interacción usuario-agente. La etapa final corresponde fundamentalmente a la evaluación del modelo emocional por medio del agente Hifi. Con ese propósito se llevó a cabo una serie de estudios con usuarios reales, 70 sujetos, interaccionando con diferentes versiones del agente Hifi en tiempo real, cada uno en una fase diferente y con sus propias características o capacidades emocionales. En particular, todos los estudios realizados han profundizado en la comparación entre una versión de referencia del agente no dotada de ningún comportamiento o característica emocional, y una versión del agente modificada convenientemente con el modelo emocional propuesto. Los resultados obtenidos nos han permitido comprender y valorar mejor la utilidad de las emociones en los sistemas de diálogo hablado. Dicha utilidad depende de varios aspectos. En primer lugar, un SCA no debe expresar sus emociones a ciegas o arbitrariamente, incluso aunque éstas sean positivas. Más bien, debe adaptar sus emociones a los diferentes estados de los usuarios. En segundo lugar, un funcionamiento relativamente pobre por parte de un SCA podría compensarse, en cierto modo, dotando al SCA de comportamiento y capacidades emocionales. En tercer lugar, aprovechar la prosodia como vehículo para expresar las emociones, de manera complementaria al empleo de mensajes con un contenido emocional específico tanto desde el punto de vista léxico como semántico, ayuda a mejorar la percepción por parte de los usuarios de un SCA. Tomados en conjunto, los resultados alcanzados no sólo confirman el éxito del modelo emocional, sino xv que constituyen además una evidencia decisiva con respecto a los beneficios de incorporar emociones en un SCA, especialmente en cuanto a reducir el nivel de frustración de los usuarios y, en última instancia, mejorar su satisfacción.

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.

Application of user-centered design in the development of an atomated/assisted testing system for self-service devices

Automated Teller Machines (ATMs) are sensitive self-service systems that require important investments in security and testing. ATM certifications are testing processes for machines that integrate software components from different vendors and are performed before their deployment for public use. This project was originated from the need of optimization of the certification process in an ATM manufacturing company. The process identifies compatibility problems between software components through testing. It is composed by a huge number of manual user tasks that makes the process very expensive and error-prone. Moreover, it is not possible to fully automate the process as it requires human intervention for manipulating ATM peripherals. This project presented important challenges for the development team. First, this is a critical process, as all the ATM operations rely on the software under test. Second, the context of use of ATMs applications is vastly different from ordinary software. Third, ATMs’ useful lifetime is beyond 15 years and both new and old models need to be supported. Fourth, the know-how for efficient testing depends on each specialist and it is not explicitly documented. Fifth, the huge number of tests and their importance implies the need for user efficiency and accuracy. All these factors led us conclude that besides the technical challenges, the usability of the intended software solution was critical for the project success. This business context is the motivation of this Master Thesis project. Our proposal focused in the development process applied. By combining user-centered design (UCD) with agile development we ensured both the high priority of usability and the early mitigation of software development risks caused by all the technology constraints. We performed 23 development iterations and finally we were able to provide a working solution on time according to users’ expectations. The evaluation of the project was carried out through usability tests, where 4 real users participated in different tests in the real context of use. The results were positive, according to different metrics: error rate, efficiency, effectiveness, and user satisfaction. We discuss the problems found, the benefits and the lessons learned in the process. Finally, we measured the expected project benefits by comparing the effort required by the current and the new process (once the new software tool is adopted). The savings corresponded to 40% less effort (man-hours) per certification. Future work includes additional evaluation of product usability in a real scenario (with customers) and the measuring of benefits in terms of quality improvement.

Factors associated with successful answering of clinical questions using an information retrieval system*

Objectives: Despite the growing use of online databases by clinicians, there has been very little research documenting how effectively they are used. This study assessed the ability of medical and nurse-practitioner students to answer clinical questions using an information retrieval system. It also attempted to identify the demographic, experience, cognitive, personality, search mechanics, and user-satisfaction factors associated with successful use of a retrieval system.

Event ordering through temporal expression resolution

In this paper a multilingual method for event ordering based on temporal expression resolution is presented. This method has been implemented through the TERSEO system which consists of three main units: temporal expression recognizing, resolution of the coreference introduced by these expressions, and event ordering. By means of this system, chronological information related to events can be extracted from documental databases. This information is automatically added to the documental database in order to allow its use by question answering systems in those cases referring to temporality. The system has been evaluated obtaining results of 91 % precision and 71 % recall. For this, a blind evaluation process has been developed guaranteing a reliable annotation process that was measured through the kappa factor.

An authoring tool for decision support systems in context questions of ecological knowledge

Decision support systems (DSS) support business or organizational decision-making activities, which require the access to information that is internally stored in databases or data warehouses, and externally in the Web accessed by Information Retrieval (IR) or Question Answering (QA) systems. Graphical interfaces to query these sources of information ease to constrain dynamically query formulation based on user selections, but they present a lack of flexibility in query formulation, since the expressivity power is reduced to the user interface design. Natural language interfaces (NLI) are expected as the optimal solution. However, especially for non-expert users, a real natural communication is the most difficult to realize effectively. In this paper, we propose an NLI that improves the interaction between the user and the DSS by means of referencing previous questions or their answers (i.e. anaphora such as the pronoun reference in “What traits are affected by them?”), or by eliding parts of the question (i.e. ellipsis such as “And to glume colour?” after the question “Tell me the QTLs related to awn colour in wheat”). Moreover, in order to overcome one of the main problems of NLIs about the difficulty to adapt an NLI to a new domain, our proposal is based on ontologies that are obtained semi-automatically from a framework that allows the integration of internal and external, structured and unstructured information. Therefore, our proposal can interface with databases, data warehouses, QA and IR systems. Because of the high NL ambiguity of the resolution process, our proposal is presented as an authoring tool that helps the user to query efficiently in natural language. Finally, our proposal is tested on a DSS case scenario about Biotechnology and Agriculture, whose knowledge base is the CEREALAB database as internal structured data, and the Web (e.g. PubMed) as external unstructured information.