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.

Leadership and project management competences in executive women: different styles from the agents' perceptions

The exercise of management and leadership are key aspects to achieve objectives, goals and relationships in the current knowledge society marked by increasingly competitive environments in which cognitive factors, creativity, knowledge and information determine the success of organizations. Both concepts have been historically associated with the male domain because of the underrepresentation of women in managerial positions. However, the increasing participation of women in the workplace has led to the development of an extensive literature on the possible existence of differences between the styles of male and female leadership, although it has not been addressed from the analysis of competences associated with each sex. Through a participatory process the abilities and skills related to women managers are analyzed and the differences in leadership styles. The results indicate that women particularly value the skills associated with human relationships, and that female leadership style tends to be transformational.

Analysis of behavior directed towards changing the innovation agents. Case study of the Canary Islands

In 2005 the Directorate General for Industrial Development and Technological Innovation of the Canary Islands proceeded to carry out a project to measure the behavioral skills of various government agencies and companies in the Canary Islands in order to prepare a White Paper to assess the most effective measures for the stimulation of innovation in this autonomous community and to facilitate the objectives of public subsidies. This paper shows a portion of the work performed comparing the activity oriented towards innovation and the one aimed at sustaining the status quo of the organizations in the sample.

I feel you: Towards affect-sensitive domotic Spoken Conversational Agents

We describe the work on infusion of emotion into limitedtask autonomous spoken conversational agents (SCAs) situated in the domestic environment, using a Need-inspired task-independentEmotion model (NEMO). In order to demonstrate the generation of a?ect through the use of the model, we describe the work of integrating it with a naturallanguage mixed-initiative HiFi-control SCA. NEMO and the host system communicates externally, removing the need for the Dialog Manager to be modi?ed as done in most existing dialog systems, in order to be adaptive. We also summarize the work on automatic a?ect prediction, namely frustration and contentment from dialog features, a non-conventional source, in the attempt of moving towards a more user-centric approach.

Los agentes de la edificación desde el punto de vista de la Gestión Integral de Proyectos: diferencias y semejanzas entre la Legislación Española y los Estándares Internacionales = Expert agents in the Global Development of a building project: differences and similarities between Spanish and International contexts.

Desde los años 60, crece en Europa y Estados Unidos la preocupación y la necesidad de mejorar los procesos de gerencia de los proyectos de construcción al volverse estos más complejos. Esto ha llevado a la continua aparición de nuevos profesionales desde la fecha citada hasta nuestros días. De ahí la complejidad de conocer las cualidades de cada uno de ellos, así como las funciones a realizar o la formación que deben tener para poder desarrollar el puesto de trabajo según el papel que desempeñan para cada actividad. Muchos agentes son los que pueden intervenir en la edificación, muchas son las funciones que llevan a cabo estos agentes, muchas son las habilidades que se necesitan para realizar estas misiones, y una buena gestión de la edificación es la que hay que desarrollar para lograr el gran éxito. El presente trabajo fin de máster, dirigido a arquitectos, arquitectos técnicos, ingenieros, abogados, economistas y todos los profesionales del sector inmobiliario y de la construcción, trata de resolver todas aquellas dudas sobre los diferentes sujetos que estarán presentes desde la definición del proyecto en la fase inicial hasta el final de la obra, pasando por las fases de pre-construcción, construcción y post-construcción. (ENGLISH VERSION) Since the 1960s, most construction projects have become more and more complex, and new concerns and necessities related to the management of a project have been on the rise in Europe and in the United States. Thence, the need for more specialized professionals in the field has become a common fact, as well as the inclusion of new curricular subjects in most building engineering studies. There are different agents that play a relevant role in a building project; some of them are expected to perform a highly specialized set of functions that require specific management skills for the work to be successful. This research work—aimed mainly at engineers, quantity surveyors, lawyers, economists, real estate and construction professionals—shows the major implications of the building construction process including both pre-tender/construction and post-tender/construction stages as far as the main expert agents are involved.

MAIA: an event-based modular architecture for intelligent agents

Online services are no longer isolated. The release of public APIs and technologies such as web hooks are allowing users and developers to access their information easily. Intelligent agents could use this information to provide a better user experience across services, connecting services with smart automatic. behaviours or actions. However, agent platforms are not prepared to easily add external sources such as web services, which hinders the usage of agents in the so-called Evented or Live Web. As a solution, this paper introduces an event-based architecture for agent systems, in accordance with the new tendencies in web programming. In particular, it is focused on personal agents that interact with several web services. With this architecture, called MAIA, connecting to new web services does not involve any modification in the platform.

Toxicity and sublethal effects of six insecticides to last instar larvae and adults of the biocontrol agents Chrysoperla carnea (Stephens) (Neuroptera: Chrysopidae) and Adalia bipunctata (L.) (Coleoptera: Coccinellidae)

To further develop Integrated Pest Management (IPM) strategies against crop pests, it is important to evaluate the effects of insecticides on biological control agents. Therefore, we tested the toxicity and sublethal effects (fecundity and fertility) of flonicamid, flubendiamide, metaflumizone, spirotetramat, sulfoxaflor and deltamethrin on the natural enemies Chrysoperla carnea and Adalia bipunctata. The side effects of the active ingredients of the insecticides were evaluated with residual contact tests for the larvae and adults of these predators in the laboratory. Flonicamid, flubendiamide, metaflumizone and spirotetramat were innocuous to last instar larvae and adults of C. carnea and A. bipunctata. Sulfoxaflor was slightly toxic to adults of C. carnea and was highly toxic to the L4 larvae of A. bipunctata. For A. bipunctata, sulfoxaflor and deltamethrin were the most damaging compounds with a cumulative larval mortality of 100%. Deltamethrin was also the most toxic compound to larvae and adults of C. carnea. In accordance with the results obtained, the compounds flonicamid, flubendiamide, metaflumizone and spirotetramat might be incorporated into IPM programs in combination with these natural enemies for the control of particular greenhouse pests. Nevertheless, the use of sulfoxaflor and deltamethrin in IPM strategies should be taken into consideration when releasing either of these biological control agents, due to the toxic behavior observed under laboratory conditions. The need for developing sustainable approaches to combine the use of these insecticides and natural enemies within an IPM framework is discussed.

Natural and experimental oral infection of nonhuman primates by bovine spongiform encephalopathy agents

Experimental lemurs either were infected orally with the agent of bovine spongiform encephalopathy (BSE) or were maintained as uninfected control animals. Immunohistochemical examination for proteinase-resistant protein (prion protein or PrP) was performed on tissues from two infected but still asymptomatic lemurs, killed 5 months after infection, and from three uninfected control lemurs. Control tissues showed no staining, whereas PrP was detected in the infected animals in tonsil, gastrointestinal tract and associated lymphatic tissues, and spleen. In addition, PrP was detected in ventral and dorsal roots of the cervical spinal cord, and within the spinal cord PrP could be traced in nerve tracts as far as the cerebral cortex. Similar patterns of PrP immunoreactivity were seen in two symptomatic and 18 apparently healthy lemurs in three different French primate centers, all of which had been fed diets supplemented with a beef protein product manufactured by a British company that has since ceased to include beef in its veterinary nutritional products. This study of BSE-infected lemurs early in their incubation period extends previous pathogenesis studies of the distribution of infectivity and PrP in natural and experimental scrapie. The similarity of neuropathology and PrP immunostaining patterns in experimentally infected animals to those observed in both symptomatic and asymptomatic animals in primate centers suggests that BSE contamination of zoo animals may have been more widespread than is generally appreciated.

Carboxyfullerenes as neuroprotective agents

Two regioisomers with C3 or D3 symmetry of water-soluble carboxylic acid C60 derivatives, containing three malonic acid groups per molecule, were synthesized and found to be equipotent free radical scavengers in solution as assessed by EPR analysis. Both compounds also inhibited the excitotoxic death of cultured cortical neurons induced by exposure to N-methyl-d-aspartate (NMDA), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), or oxygen-glucose deprivation, but the C3 regioisomer was more effective than the D3 regioisomer, possibly reflecting its polar nature and attendant greater ability to enter lipid membranes. At 100 μM, the C3 derivative fully blocked even rapidly triggered, NMDA receptor-mediated toxicity, a form of toxicity with limited sensitivity to all other classes of free radical scavengers we have tested. The C3 derivative also reduced apoptotic neuronal death induced by either serum deprivation or exposure to Aβ1–42 protein. Furthermore, continuous infusion of the C3 derivative in a transgenic mouse carrying the human mutant (G93A) superoxide dismutase gene responsible for a form of familial amyotrophic lateral sclerosis, delayed both death and functional deterioration. These data suggest that polar carboxylic acid C60 derivatives may have attractive therapeutic properties in several acute or chronic neurodegenerative diseases.

An intercalation inhibitor altering the target selectivity of DNA damaging agents: Synthesis of site-specific aflatoxin B1 adducts in a p53 mutational hotspot

Aflatoxin B1 (AFB1) is a potent human carcinogen implicated in the etiology of hepatocellular carcinoma. Upon metabolic activation to the reactive epoxide, AFB1 forms DNA adducts primarily at the N7 position of guanines. To elucidate more fully the molecular mechanism of AFB1-induced mutagenesis, an intercalation inhibitor was designed to probe the effects of intercalation by AFB1 epoxide on its reaction with DNA. DNA duplexes were prepared consisting of a target strand containing multiple potentially reactive guanines and a nontarget strand containing a cis-syn thymidine-benzofuran photoproduct. Because the covalently linked benzofuran moiety physically occupies an intercalation site, we reasoned that such a site would be rendered inaccessible to AFB1 epoxide. By strategic positioning of this intercalation inhibitor in the intercalation site 5′ to a specific guanine, the adduct yield at that site was greatly diminished, indicating that intercalation by AFB1 epoxide contributes favorably to adduct formation. Using this approach it has been possible to simplify the production of site-specifically modified oligonucleotides containing AFB1 adducts in the sequence context of a p53 mutational hotspot. Moreover, we report herein isolation of site-specifically AFB1-modified oligonucleotides in sequences containing multiple guanines. Use of intercalation inhibitors will facilitate both investigation of the ability of other carcinogens to intercalate into DNA and the synthesis of specific carcinogen-DNA adducts.

p53-dependent regulation of MDR1 gene expression causes selective resistance to chemotherapeutic agents

Loss of functional p53 paradoxically results in either increased or decreased resistance to chemotherapeutic drugs. The inconsistent relationship between p53 status and drug sensitivity may reflect p53’s selective regulation of genes important to cytotoxic response of chemotherapeutic agents. We reasoned that the discrepant effects of p53 on chemotherapeutic cytotoxicity is due to p53-dependent regulation of the multidrug resistance gene (MDR1) expression in tumors that normally express MDR1. To test the hypothesis that wild-type p53 regulates the endogenous mdr1 gene we stably introduced a trans-dominant negative (TDN) p53 into rodent H35 hepatoma cells that express P-glycoprotein (Pgp) and have wild-type p53. Levels of Pgp and mdr1a mRNA were markedly elevated in cells expressing TDN p53 and were linked to impaired p53 function (both transactivation and transrepression) in these cells. Enhanced mdr1a gene expression in the TDN p53 cells was not secondary to mdr1 gene amplification and Pgp was functional as demonstrated by the decreased uptake of vinblastine. Cytotoxicity assays revealed that the TDN p53 cell lines were selectively insensitive to Pgp substrates. Sensitivity was restored by the Pgp inhibitor reserpine, demonstrating that only drug retention was the basis for loss of drug sensitivity. Similar findings were evident in human LS180 colon carcinoma cells engineered to overexpress TDN p53. Therefore, the p53 inactivation seen in cancers likely leads to selective resistance to chemotherapeutic agents because of up-regulation of MDR1 expression.

Aminopeptidase A inhibitors as potential central antihypertensive agents

Overactivity of the brain renin-angiotensin system (RAS) has been implicated in the development and maintenance of hypertension in several experimental models, such as spontaneously hypertensive rats and transgenic mice expressing both human renin and human angiotensinogen transgenes. We recently reported that, in the murine brain, angiotensin II (AngII) is converted to angiotensin III (AngIII) by aminopeptidase A (APA), whereas AngIII is inactivated by aminopeptidase N (APN). If injected into cerebral ventricles (ICV), AngII and AngIII cause similar pressor responses. Because AngII is metabolized in vivo into AngIII, the exact nature of the active peptide is not precisely determined. Here we report that, in rats, ICV injection of the selective APA inhibitor EC33 [(S)-3-amino-4-mercaptobutyl sulfonic acid] blocked the pressor response of exogenous AngII, suggesting that the conversion of AngII to AngIII is required to increase blood pressure (BP). Furthermore, ICV injection, but not i.v. injection, of EC33 alone caused a dose-dependent decrease in BP by blocking the formation of brain but not systemic AngIII. This is corroborated by the fact that the selective APN inhibitor, PC18 (2-amino-4-methylsulfonyl butane thiol), administered alone via the ICV route, increases BP. This pressor response was blocked by prior treatment with the angiotensin type 1 (AT1) receptor antagonist, losartan, showing that blocking the action of APN on AngIII metabolism leads to an increase in endogenous AngIII levels, resulting in BP increase, through interaction with AT1 receptors. These data demonstrate that AngIII is a major effector peptide of the brain RAS, exerting tonic stimulatory control over BP. Thus, APA, the enzyme responsible for the formation of brain AngIII, represents a potential central therapeutic target that justifies the development of APA inhibitors as central antihypertensive agents.

Synthesis and biological activity of DNA damaging agents that form decoy binding sites for the estrogen receptor

It is a goal of cancer chemotherapy to achieve the selective killing of tumor cells while minimizing toxicity to normal tissues. We describe the design of selective toxins forming DNA adducts that attract the estrogen receptor (ER), a transcription factor that is overexpressed in many human breast and ovarian tumors. The compounds consist of 4-(3-aminopropyl)-N,N-(2-chloroethyl)-aniline linked to 2-(4′-hydroxyphenyl)-3-methyl-5-hydroxy-indole. The former moiety is a DNA damaging nitrogen mustard and the latter is a ligand for the ER. The connection between these groups was refined to permit DNA adducts formed by the mustard portion of the molecule to present the ligand domain so that it was able to interact efficiently with the ER. By using 16-mers containing specific DNA adducts, it was determined that monoadducts and putative intrastrand crosslinks were preferred targets for the ER over interstrand crosslinks. A series of structurally related 2-phenylindole mustards was prepared, some of which were selectively toxic to the ER-positive breast cancer cell line MCF-7, as compared with the ER(−) negative line MDA-MB231. The ability both to bind to DNA and to interact significantly with the ER were essential to achieve selective lethality toward ER(+) cells. Compounds forming DNA adducts without the ability to bind receptor showed similar toxicities in the two cell lines. Several models could explain the selective toxicity of the mustard–phenylindole compounds toward ER(+) cells. The favored model suggests that a mustard–DNA adduct is shielded by the ER from DNA repair enzymes and hence cells possessing an abundance of the ER selectively retain the adduct and are killed.

tRNAVal-heterodimeric maxizymes with high potential as geneinactivating agents: Simultaneous cleavage at two sites in HIV-1 tat mRNA in cultured cells

It has been demonstrated that shortened forms of (stem II-deleted) hammerhead ribozymes with low intrinsic activity form very active dimers with a common stem II (very active short ribozymes capable of forming dimers were designated maxizymes). Intracellular activities of heterodimeric maxizymes and conventional ribozymes, under the control of a human tRNAVal-promoter, were compared against the cleavage of HIV-1 tat mRNA. The pol III-driven maxizymes formed very active heterodimers, and they successfully cleaved HIV-1 tat mRNA in mammalian cells at two sites simultaneously. The cleaved fragments were identified directly by Northern blotting analysis. Despite the initial concerns that a complicated dimerization process and formation of inactive homodimers were involved in addition to the process of association with the target, the overall intracellular activities of tRNAVal-driven maxizymes were significantly higher in mammalian cells than those of two sets of independent, conventional hammerhead ribozymes that were targeted at the same two sites within HIV-1 tat mRNA. Because the tRNAVal-driven maxizymes tested to date have been more effective than tRNAVal-driven “standard” hammerhead ribozymes, the tRNAVal-driven heterodimeric maxizymes appear to have potential utility as gene-inactivating agents.