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.

Mejora de la eficacia de Penicillium oxalicum como agente de biocontrol en enfermedades de plantas hortícolas

Actualmente, la reducción de materias activas (UE) y la implantación de la nueva Directiva comunitaria 2009/128/ que establece el marco de actuación para conseguir un uso sostenible de los plaguicidas químicos y la preferencia de uso de métodos biológicos, físicos y otros no químicos, obliga a buscar métodos de control menos perjudiciales para el medio ambiente. El control biológico (CB) de enfermedades vegetales empleando agentes de control biológico (ACB) se percibe como una alternativa más segura y con menor impacto ambiental, bien solos o bien como parte de una estrategia de control integrado. El aislado 212 de Penicillium oxalicum (PO212) (ATCC 201888) fue aislado originalmente de la micoflora del suelo en España y ha demostrado ser un eficaz ACB frente a la marchitez vascular del tomate. Una vez identificado y caracterizado el ACB se inició el periodo de desarrollo del mismo poniendo a punto un método de producción en masa de sus conidias. Tras lo cual se inició el proceso de formulación del ACB deshidratando las conidias para su preservación durante un período de tiempo mayor mediante lecho fluido. Finalmente, se han desarrollado algunos formulados que contienen de forma individual diferentes aditivos que han alargado su viabilidad, estabilidad y facilitado su manejo y aplicación. Sin embargo, es necesario seguir trabajando en la mejora de su eficacia de biocontrol. El primer objetivo de esta Tesis se ha centrado en el estudio de la interacción ACB-patógeno-huésped que permita la actuación de P.oxalicum en diferentes patosistemas. Uno de los primeros puntos que se abordan dentro de este objetivo es el desarrollo de nuevas FORMULACIONES del ACB que incrementen su eficacia frente a la marchitez vascular del tomate. Las conidias formuladas de PO212 se obtuvieron por la adición conjunta de distintos aditivos (mojantes, adherentes o estabilizantes) en dos momentos diferentes del proceso de producción/secado: i) antes del proceso de producción (en la bolsa de fermentación) en el momento de la inoculación de las bolsas de fermentación con conidias de PO212 o ii) antes del secado en el momento de la resuspensión de las conidias tras su centrifugación. De las 22 nuevas formulaciones desarrolladas y evaluadas en plantas de tomate en ensayos en invernadero, seis de ellas (FOR22, FOR25, FOR32, FOR35, FOR36 y FOR37) mejoran significativamente (P=0,05) el control de la marchitez vascular del tomate con respecto al obtenido con las conidias secas de P.oxalicum sin aditivos (CSPO) o con el fungicida Bavistin. Los formulados que mejoran la eficacia de las conidias secas sin aditivos son aquellos que contienen como humectantes alginato sódico en fermentación, seguido de aquellos que contienen glicerol como estabilizante en fermentación, y metil celulosa y leche desnatada como adherentes antes del secado. Además, el control de la marchitez vascular del tomate por parte de los formulados de P. oxalicum está relacionado con la fecha de inicio de la enfermedad. Otra forma de continuar mejorando la eficacia de biocontrol es mejorar la materia activa mediante la SELECCIÓN DE NUEVAS CEPAS de P. oxalicum, las cuales podrían tener diferentes niveles de eficacia. De entre las 28 nuevas cepas de P. oxalicum ensayadas en cámara de cultivo, sólo el aislado PO15 muestra el mismo nivel de eficacia que PO212 (62-67% de control) frente a la marchitez vascular del tomate en casos de alta presión de enfermedad. Mientras que, en casos de baja presión de enfermedad todas las cepas de P. oxalicum y sus mezclas demuestran ser eficaces. Finalmente, se estudia ampliar el rango de actuación de este ACB a OTROS HUÉSPEDES Y OTROS PATÓGENOS Y DIFERENTES GRADOS DE VIRULENCIA. En ensayos de eficacia de P. oxalicum frente a aislados de diferente agresividad de Verticillium spp. y Fusarium oxysporum f. sp. lycopersici en plantas de tomate en cámaras de cultivo, se demuestra que la eficacia de PO212 está negativamente correlacionada con el nivel de enfermedad causada por F. oxysporum f. sp. lycopersici pero que no hay ningún efecto diferencial en la reducción de la incidencia ni de la gravedad según la virulencia de los aislados. Sin embargo, en los ensayos realizados con V. dahliae, PO212 causa una mayor reducción de la enfermedad en las plantas inoculadas con aislados de virulencia media. La eficacia de PO212 también era mayor frente a aislados de virulencia media alta de F. oxysporum f. sp. melonis y F. oxysporum f. sp. niveum, en plantas de melón y sandía, respectivamente. En ambos huéspedes se demuestra que la dosis óptima de aplicación del ACB es de 107 conidias de PO212 g-1 de suelo de semillero, aplicada 7 días antes del trasplante. Además, entre 2 y 4 nuevas aplicaciones de PO212 a la raíces de las plantas mediante un riego al terreno de asiento mejoran la eficacia de biocontrol. La eficacia de PO212 no se limita a hongos patógenos vasculares como los citados anteriormente, sino también a otros patógenos como: Phytophthora cactorum, Globodera pallida y G. rostochiensis. PO212 reduce significativamente los síntomas (50%) causados por P. cactorum en plantas de vivero de fresa, tras la aplicación del ACB por inmersión de las raíces antes de su trasplante al suelo de viveros comerciales. Por otra parte, la exposición de los quistes de Globodera pallida y G. rostochiensis (nematodos del quiste de la patata) a las conidias de P. oxalicum, en ensayos in vitro o en microcosmos de suelo, reduce significativamente la capacidad de eclosión de los huevos. Para G. pallida esta reducción es mayor cuando se emplean exudados de raíz de patata del cv. 'Monalisa', que exudados de raíz del cv. 'Desirée'. No hay una reducción significativa en la tasa de eclosión con exudados de raíz de tomate del cv. 'San Pedro'. Para G. rostochiensis la reducción en la tasa de eclosión de los huevos se obtiene con exudados de la raíz de patata del cv. 'Desirée'. El tratamiento con P. oxalicum reduce también significativamente el número de quistes de G. pallida en macetas. Con el fin de optimizar la aplicación práctica de P. oxalicum cepa 212 como tratamiento biológico del suelo, es esencial entender cómo el entorno físico influye en la capacidad de colonización, crecimiento y supervivencia del mismo, así como el posible riesgo que puede suponer su aplicación sobre el resto de los microorganismos del ecosistema. Por ello en este segundo objetivo de esta tesis se estudia la interacción del ACB con el medio ambiente en el cual se aplica. Dentro de este objetivo se evalúa la INFLUENCIA DE LA TEMPERATURA, DISPONIBILIDAD DE AGUA Y PROPIEDADES FÍSICO-QUÍMICAS DE LOS SUELOS (POROSIDAD, TEXTURA, DENSIDAD...) SOBRE LA SUPERVIVENCIA Y EL CRECIMIENTO DE PO212 en condiciones controladas elaborando modelos que permitan predecir el impacto de cada factor ambiental en la supervivencia y crecimiento de P. oxalicum y conocer su capacidad para crecer y sobrevivir en diferentes ambientes. En las muestras de suelo se cuantifica: i) la supervivencia de Penicillium spp. usando el recuento del número de unidades formadoras de colonias en un medio de cultivo semi-selectivo y ii) el crecimiento (biomasa) de PO212 mediante PCR en tiempo real. En los resultados obtenidos se demuestra que P. oxalicum crece y sobrevive mejor en condiciones de sequía independientemente de la temperatura y del tipo de suelo. Si comparamos tipos de suelo P. oxalicum crece y sobrevive en mayor medida en suelos areno-arcillosos con un bajo contenido en materia orgánica, un mayor pH y una menor disponibilidad de fósforo y nitrógeno. La supervivencia y el crecimiento de P. oxalicum se correlaciona de forma negativa con la disponibilidad de agua y de forma positiva con el contenido de materia orgánica. Sólo la supervivencia se correlaciona también positivamente con el pH. Por otro lado se realizan ensayos en suelos de huertos comerciales con diferentes propiedades físico-químicas y diferentes condiciones ambientales para ESTUDIAR EL ESTABLECIMIENTO, SUPERVIVENCIA Y DISPERSIÓN VERTICAL Y MOVILIDAD HORIZONTAL DE PO212. P. oxalicum 212 puede persistir y sobrevivir en esos suelos al menos un año después de su liberación pero a niveles similares a los de otras especies de Penicillium indígenas presentes en los mismos suelos naturales. Además, P. oxalicum 212 muestra una dispersión vertical y movilidad horizontal muy limitada en los diferentes tipos de suelo evaluados. La introducción de P. oxalicum en un ambiente natural no sólo implica su actuación sobre el microorganismo diana, el patógeno, si no también sobre otros microorganismos indígenas. Para EVALUAR EL EFECTO DE LA APLICACIÓN DE P. oxalicum SOBRE LAS POBLACIONES FÚNGICAS INDIGENAS PRESENTES EN EL SUELO de dos huertos comerciales, se analizan mediante electroforesis en gradiente desnaturalizante de poliacrilamida (DGGE) muestras de dichos suelos a dos profundidades (5 y 10 cm) y a cuatro fechas desde la aplicación de P. oxalicum 212 (0, 75, 180 y 365 días). El análisis de la DGGE muestra que las diferencias entre las poblaciones fúngicas se deben significativamente a la fecha de muestreo y son independientes del tratamiento aplicado y de la profundidad a la que se tomen las muestras. Luego, la aplicación del ACB no afecta a la población fúngica de los dos suelos analizados. El análisis de las secuencias de la DGGE confirma los resultados anteriores y permiten identificar la presencia del ACB en los suelos. La presencia de P. oxalicum en el suelo se encuentra especialmente relacionada con factores ambientales como la humedad. Por tanto, podemos concluir que Penicillium oxalicum cepa 212 puede considerarse un óptimo Agente de Control Biológico (ACB), puesto que es ecológicamente competitivo, eficaz para combatir un amplio espectro de enfermedades y no supone un riesgo para el resto de microorganismos fúngicos no diana presentes en el lugar de aplicación. ABSTRACT Currently, reduction of active (EU) and the implementation of the new EU Directive 2009/128 which establishing the framework for action to achieve the sustainable use of chemical pesticides and preference of use of biological, physical and other non-chemical methods, forces to look for control methods less harmful to the environment. Biological control (CB) of plant diseases using biological control agents (BCA) is perceived as a safer alternative and with less environmental impact, either alone or as part of an integrated control strategy. The isolate 212 of Penicillium oxalicum (PO212) (ATCC 201888) was originally isolated from the soil mycoflora in Spain. P. oxalicum is a promising biological control agent for Fusarium wilt and other tomato diseases. Once identified and characterized the BCA, was developed a mass production method of conidia by solid-state fermentation. After determined the process of obtaining a formulated product of the BCA by drying of product by fluid-bed drying, it enables the preservation of the inoculum over a long period of time. Finally, some formulations of dried P. oxalicum conidia have been developed which contain one different additive that have improved their viability, stability and facilitated its handling and application. However, further work is needed to improve biocontrol efficacy. The first objective of this thesis has focused on the study of the interaction BCA- pathogen-host, to allow P.oxalicum to work in different pathosystems. The first point to be addressed in this objective is the development of new FORMULATIONS of BCA which increase their effectiveness against vascular wilt of tomato. PO212 conidial formulations were obtained by the joint addition of various additives (wetting agents, adhesives or stabilizers) at two different points of the production-drying process: i) to substrate in the fermentation bags before the production process, and (ii) to conidial paste obtained after production but before drying. Of the 22 new formulations developed and evaluated in tomato plants in greenhouse tests, six of them (FOR22 , FOR25 , FOR32 , FOR35 , FOR36 and FOR3) improved significantly (P = 0.05) the biocontrol efficacy against tomato wilt with respect to that obtained with dried P.oxalicum conidia without additives (CSPO) or the fungicide Bavistin. The formulations that improve the efficiency of dried conidia without additives are those containing as humectants sodium alginate in the fermentation bags, followed by those containing glycerol as a stabilizer in the fermentation bags, and methylcellulose and skimmed milk as adherents before drying. Moreover, control of vascular wilt of tomatoes by PO212 conidial formulations is related to the date of disease onset. Another way to further improve the effectiveness of biocontrol is to improve the active substance by SELECTION OF NEW STRAINS of P. oxalicum, which may have different levels of effectiveness. Of the 28 new strains of P. oxalicum tested in a culture chamber, only PO15 isolate shows the same effectiveness that PO212 (62-67 % of control) against tomato vascular wilt in cases of high disease pressure. Whereas in cases of low disease pressure all strains of P. oxalicum and its mixtures effective. Finally, we study extend the range of action of this BCA TO OTHER GUESTS AND OTHER PATHOGENS AND DIFFERENT DEGREES OF VIRULENCE. In efficacy trials of P. oxalicum against isolates of different aggressiveness of Verticillium spp. and Fusarium oxysporum f. sp. lycopersici in tomato plants in growth chambers, shows that the efficiency of PO212 is negatively correlated with the level of disease caused by F. oxysporum f. sp. lycopersici. There is not differential effect in reducing the incidence or severity depending on the virulence of isolates. However, PO212 cause a greater reduction of disease in plants inoculated with virulent isolates media of V. dahlia. PO212 efficacy was also higher against isolates of high and average virulence of F. oxysporum f. sp. melonis and F. oxysporum f. sp. niveum in melon and watermelon plants, respectively. In both hosts the optimum dose of the BCA application is 107 conidia PO212 g-1 soil, applied on seedlings 7 days before transplantation into the field. Moreover, the reapplication of PO212 (2-4 times) to the roots by irrigation into the field improve efficiency of biocontrol. The efficacy of PO212 is not limited to vascular pathogens as those mentioned above, but also other pathogens such as Oomycetes (Phytophthora cactorum) and nematodes (Globodera pallida and G. rostochiensis). PO212 significantly reduces symptoms (50 %) caused by P. cactorum in strawberry nursery plants after application of BCA by dipping the roots before transplanting to soil in commercial nurseries. Moreover, the exposure of G. pallida and G. rostochiensis cysts to the conidia of P. oxalicum, in in vitro assays or in soil microcosms significantly reduces hatchability of eggs. The reduction in the rate of G. pallida juveniles hatching was greatest when root diffusates from the `Monalisa´ potato cultivar were used, followed by root diffusates from the `Désirée´ potato cultivar. However, no significant reduction in the rate of G. pallida juveniles hatching was found when root diffusates from the ‘San Pedro” tomato cultivar were used. For G. rostochiensis reduction in the juveniles hatching is obtained from the root diffusates 'Desirée' potato cultivar. Treatment with P. oxalicum also significantly reduces the number of cysts of G. pallida in pots. In order to optimize the practical application of P. oxalicum strain 212 as a biological soil treatment, it is essential to understand how the physical environment influences the BCA colonization, survival and growth, and the possible risk that can cause its application on other microorganisms in the ecosystem of performance. Therefore, the second objective of this thesis is the interaction of the BCA with the environment in which it is applied. Within this objective is evaluated the INFLUENCE OF TEMPERATURE, WATER AVAILABILITY AND PHYSICAL-CHEMICAL PROPERTIES OF SOILS (POROSITY, TEXTURE, DENSITY...) ON SURVIVAL AND GROWTH OF PO212 under controlled conditions to develop models for predicting the environmental impact of each factor on survival and growth of P. oxalicum and to know their ability to grow and survive in different environments. Two parameters are evaluated in the soil samples: i) the survival of Penicillium spp. by counting the number of colony forming units in semi-selective medium and ii) growth (biomass) of PO212 by real-time PCR. P. oxalicum grows and survives better in drought conditions regardless of temperature and soil type. P. oxalicum grows and survives more in sandy loam soils with low organic matter content, higher pH and lower availability of phosphorus and nitrogen. Survival and growth of P. oxalicum negatively correlates with the availability of water and positively with the organic content. Only survival also correlated positively with pH. Moreover, trials are carried out into commercial orchards soils with different physic-chemical properties and different environmental conditions TO STUDY THE ESTABLISHMENT, SURVIVAL, VERTICAL DISPERSION AND HORIZONTAL SPREAD OF PO212. P. oxalicum 212 can persist and survive at very low levels in soil one year after its release. The size of the PO212 population after its release into the tested natural soils is similar to that of indigenous Penicillium spp. Furthermore, the vertical dispersion and horizontal spread of PO212 is limited in different soil types. The introduction of P. oxalicum in a natural environment not only involves their action on the target organism, the pathogen, but also on other indigenous microorganisms. TO ASSESS THE EFFECT OF P. oxalicum APPLICATION ON SOIL INDIGENOUS FUNGAL COMMUNITIES in two commercial orchards, soil samples are analyzed by Denaturing Gradient Gel Electrophoresis polyacrylamide (DGGE). Samples are taken from soil at two depths (5 and 10 cm) and four dates from the application of P. oxalicum 212 (0, 75, 180 and 365 days). DGGE analysis shows that differences are observed between sampling dates and are independent of the treatment of P. oxalicum applied and the depth. BCA application does not affect the fungal population of the two soil analyzed. Sequence analysis of the DGGE bands confirms previous findings and to identify the presence of BCA on soils. The presence of P. oxalicum in soil is especially related to environmental factors such as humidity. Therefore, we conclude that the 212 of strain Penicillium oxalicum can be considered an optimum BCA, since it is environmentally competitive and effective against a broad spectrum of diseases and does not have any negative effect on soil non-target fungi communities.

Conflict and cognitive control during sentence comprehension: recruitment of a frontal network during the processing of Spanish object-first sentences.

During sentence processing there is a preference to treat the first noun phrase found as the subject and agent, unless marked the other way. This preference would lead to a conflict in thematic role assignment when the syntactic structure conforms to a non-canonical object-before-subject pattern. Left perisylvian and fronto-parietal brain networks have been found to be engaged by increased computational demands during sentence comprehension, while event-reated brain potentials have been used to study the on-line manifestation of these demands. However, evidence regarding the spatiotemporal organization of brain networks in this domain is scarce. In the current study we used Magnetoencephalography to track spatio-temporally brain activity while Spanish speakers were reading subject- and object-first cleft sentences. Both kinds of sentences remained ambiguous between a subject-first or an object-first interpretation up to the appearance of the second argument. Results show the time-modulation of a frontal network at the disambiguation point of object-first sentences. Moreover, the time windows where these effects took place have been previously related to thematic role integration (300–500 ms) and to sentence reanalysis and resolution of conflicts during processing (beyond 500 ms post-stimulus). These results point to frontal cognitive control as a putative key mechanism which may operate when a revision of the sentence structure and meaning is necessary

A Multi-agent System for Emergency Decision Support

This paper describes the multi-agent organization of a computer system that was designed to assist operators in decision making in the presence of emergencies. The application was developed for the case of emergencies caused by river floods. It operates on real-time receiving data recorded by sensors (rainfall, water levels, flows, etc.) and applies multi-agent techniques to interpret the data, predict the future behavior and recommend control actions. The system includes an advanced knowledge based architecture with multiple symbolic representation with uncertainty models (bayesian networks). This system has been applied and validated at two particular sites in Spain (the Jucar basin and the South basin).

Rhizoctonia solani as causative agent of damping off of Swiss chard in Europe

During September 2011, post-emergence damping off of Swiss chard (Beta vulgaris subsp. cicla L.) was observed in a greenhouse in Villa del Prado (Spain). About 20% of the seedlings showed damping off symptoms. Lesions were initially water soaked, dark brown necrosis of crown tissue, irregular in shape and sunken in appearance on large plants, causing the infected seedlings to collapse and eventually die. Rhizoctonia solani was isolated consistently from symptomatic plants. After morphological and molecular identification of the isolates, pathogenicity was tested by placing agar plugs of four isolates adjacent to the stem at the three or four true leaf stage. In inoculated plants, brown crown and stem necrosis occurred while control plants did not show disease symptoms. Pathogenicity using non-germinated seeds was also tested. All four isolates produced extensive damping off when inoculated on non-germinated seeds. To our knowledge, this is the first report of damping off of Swiss chard caused by R. solani in Europe.

Advanced predictive quality control strategy involving different facilities

There are many industries that use highly technological solutions to improve quality in all of their products. The steel industry is one example. Several automatic surface-inspection systems are used in the steel industry to identify various types of defects and to help operators decide whether to accept, reroute, or downgrade the material, subject to the assessment process. This paper focuses on promoting a strategy that considers all defects in an integrated fashion. It does this by managing the uncertainty about the exact position of a defect due to different process conditions by means of Gaussian additive influence functions. The relevance of the approach is in making possible consistency and reliability between surface inspection systems. The results obtained are an increase in confidence in the automatic inspection system and an ability to introduce improved prediction and advanced routing models. The prediction is provided to technical operators to help them in their decision-making process. It shows the increase in improvement gained by reducing the 40 % of coils that are downgraded at the hot strip mill because of specific defects. In addition, this technology facilitates an increase of 50 % in the accuracy of the estimate of defect survival after the cleaning facility in comparison to the former approach. The proposed technology is implemented by means of software-based, multi-agent solutions. It makes possible the independent treatment of information, presentation, quality analysis, and other relevant functions.

Design and modeling of the multi-agent robotic system: SMART

This article presents the design, kinematic model and communication architecture for the multi-agent robotic system called SMART. The philosophy behind this kind of system requires the communication architecture to contemplate the concurrence of the whole system. The proposed architecture combines different communication technologies (TCP/IP and Bluetooth) under one protocol designed for the cooperation among agents and other elements of the system such as IP-Cameras, image processing library, path planner, user Interface, control block and data block. The high level control is modeled by Work-Flow Petri nets and implemented in C++ and C♯♯. Experimental results show the performance of the designed architecture.

An Agent Architecture to fulfill Real-Time Requirements

In this paper we present AMSIA, an agent architecture that combines the possibility of using di erent reasoning methods with a mechanism to control the resources needed by the agent to ful ll its high level objectives. The architecture is based on the blackboard paradigm which o ers the possibility of combining di erent reasoning techniques and opportunistic behavior. The AMSIA architecture adds a representation of plans of objectives allowing di erent reasoning activities to create plans to guide the future behavior of the agent. The opportunism is in the acquisition of high-level objectives and in the modi cation of the predicted activity when something doesn't happen as expected. A control mechanism is responsible for the translation of plans of objectives to concrete activities, considering resource-boundedness. To do so, all the activity in the agent (including control) is explicitly scheduled, but allowing the necessary exibility to make changes in the face of contingencies that are expected in dynamic environments. Experimental work is also presented.

B2DI a bayesian BDI agent model with causal belief updating based on MSBN

In this paper, we introduce B2DI model that extends BDI model to perform Bayesian inference under uncertainty. For scalability and flexibility purposes, Multiply Sectioned Bayesian Network (MSBN) technology has been selected and adapted to BDI agent reasoning. A belief update mechanism has been defined for agents, whose belief models are connected by public shared beliefs, and the certainty of these beliefs is updated based on MSBN. The classical BDI agent architecture has been extended in order to manage uncertainty using Bayesian reasoning. The resulting extended model, so-called B2DI, proposes a new control loop. The proposed B2DI model has been evaluated in a network fault diagnosis scenario. The evaluation has compared this model with two previously developed agent models. The evaluation has been carried out with a real testbed diagnosis scenario using JADEX. As a result, the proposed model exhibits significant improvements in the cost and time required to carry out a reliable diagnosis.

I feel you: the design and evaluation of a domotic affect-sensitive spoken conversational agent

We describe the work on infusion of emotion into a limited-task autonomous spoken conversational agent situated in the domestic environment, using a need-inspired task-independent emotion model (NEMO). In order to demonstrate the generation of affect through the use of the model, we describe the work of integrating it with a natural-language mixed-initiative HiFi-control spoken conversational agent (SCA). NEMO and the host system communicate externally, removing the need for the Dialog Manager to be modified, as is done in most existing dialog systems, in order to be adaptive. The first part of the paper concerns the integration between NEMO and the host agent. The second part summarizes the work on automatic affect prediction, namely, frustration and contentment, from dialog features, a non-conventional source, in the attempt of moving towards a more user-centric approach. The final part reports the evaluation results obtained from a user study, in which both versions of the agent (non-adaptive and emotionally-adaptive) were compared. The results provide substantial evidences with respect to the benefits of adding emotion in a spoken conversational agent, especially in mitigating users' frustrations and, ultimately, improving their satisfaction.

NEMOHIFI: An Affective HiFi Agent

This demo concerns a recently developed prototype of an emotionally-sensitive autonomous HiFi Spoken Conversa- tional Agent, called NEMOHIFI. The baseline agent was developed by the Speech Technology Group (GTH) and has recently been integrated with an emotional engine called NEMO (Need-inspired Emotional Model) to enable it to adapt to users emotion and respond to the users using ap- propriate expressive speech. NEMOHIFI controls and man- ages the HiFi audio system, and for end users, its functions equate a remote control, except that instead of clicking, the user interacts with the agent using voice. A pairwise com- parison between the baseline (non-adaptive) and NEMO- HIFI is also presented.

A unified framework for linear function approximation of value functions in stochastic control

This paper contributes with a unified formulation that merges previ- ous analysis on the prediction of the performance ( value function ) of certain sequence of actions ( policy ) when an agent operates a Markov decision process with large state-space. When the states are represented by features and the value function is linearly approxi- mated, our analysis reveals a new relationship between two common cost functions used to obtain the optimal approximation. In addition, this analysis allows us to propose an efficient adaptive algorithm that provides an unbiased linear estimate. The performance of the pro- posed algorithm is illustrated by simulation, showing competitive results when compared with the state-of-the-art solutions.

A cooperative multi-agent robotics system: design and modelling

This paper presents the development of the robotic multi-agent system SMART. In this system, the agent concept is applied to both hardware and software entities. Hardware agents are robots, with three and four legs, and an IP-camera that takes images of the scene where the cooperative task is carried out. Hardware agents strongly cooperate with software agents. These latter agents can be classified into image processing, communications, task management and decision making, planning and trajectory generation agents. To model, control and evaluate the performance of cooperative tasks among agents, a kind of PetriNet, called Work-Flow Petri Net, is used. Experimental results shows the good performance of the system.

Using Knowledge Modelling Tools for Agent-Based Systems: The Experience of KSM

The aim of this chapter is to discuss the applicability of recently proposed knowledge modelling tools to the development of agent-based systems. The discussion is derived from the real world experience of a particular software tool called KSM (Knowledge Structure Manager). The chapter provides details about this tool and then proceeds to show in which forms the software may be used to support the development of agent-based systems. Two multiagent systems, one in the field of telecommunications management and the other one in the field of flood control, are described. Conclusions about these studies are presented, summarizing the main contributions that knowledge modelling tools can bring to the development of agent-based systems.

Development of the infrastructure for a multi-agent traffic management system simulation

This document contains detailed description of the design and the implementation of a multi-agent application controlling traffic lights in a city together with a system for simulating traffic and testing. The goal of this thesis is to design and build a simplified intelligent and distributed solution to the problem with the traffic in the big cities following different good practices in order to allow future refining of the model of the real world. The problem of the traffic in the big cities is still a problem that cannot be solved. Not only is the increasing number of cars a reason for the traffic jams, but also the way the traffic is organized. Usually, the intersections with traffic lights are replaced by roundabouts or interchanges to increase the number of cars that can cross the intersection in certain time. But still there are places where the infrastructure cannot be changed and the traffic light semaphores are the only way to control the car flows. In real life, the traffic lights have a predefined plan for change or they receive information from a centralized system when and how they have to change. But what if the traffic lights can cooperate and decide on their own when and how to change? Using this problem, the purpose of the thesis is to explore different agent-based software engineering approaches to design and build a non-conventional distributed system. From the software engineering point of view, the goal of the thesis is to apply the knowledge and use the skills, acquired during the various courses of the master program in Software Engineering, while solving a practical and complex problem such as the traffic in the cities.