The messaging-based conversational interfaces, commonly called Chatbots, have seen massive growth lately. With the proliferation of Chatbots, there is a growing demand for a better understanding of the design practices behind conversational user experience. This thesis looked into the design process of a Chatbot-based project built on the RCS business messaging platform, and the workflow was investigated through contextual inquiry and critical incident interview techniques. The challenges experienced by practitioners from different disciplines are detailed, with a focus on their respective work tasks and practices. / De meddelandebaserade konversationsgränssnitten, vanligtvis kallade Chatbots, har sett en enorm tillväxt den senaste tiden. Med spridningen av Chatbots finns det en växande efterfrågan på en bättre förståelse för designmetoderna bakom konversationsanvändarupplevelse. Denna avhandling tittade på designprocessen för ett Chatbot-baserat projekt byggt på RCS-affärsmeddelandeplattformen, och arbetsflödet undersöktes genom kontextuell undersökning och tekniker för intervjuad kritisk incident. Utmaningarna som utövarna från olika discipliner upplever är detaljerade med fokus på sina respektive arbetsuppgifter och arbetsmetoder.

Anna Šmilauerová: TV Sitcom Friends: Analysis of character humor strategies based on the violation of Grice's Conversational maxims Abstract The purpose of this diploma thesis is the analysis of the humor strategies employed by the characters of Phoebe and Chandler in the TV Sitcom Friends. The discovered prevailing strategies were then compared with the personalities of the two characters. The data analyzed were the written script of five exemplary episodes from the Season 1, 3, 5, 7 and 9, in which the utterances eliciting laughter from the audience were thoroughly analyzed from the point of Grice's Cooperative Principle: only those utterances were considered in which the characters violated one or more of the conversational maxims (quality, quantity, relation and manner). Phoebe was found to violate most often the maxim of relation, thus it is her being non-factual and non-conventional that constitutes her most entertaining quality. As she develops and grows more mature as a character, the frequency counts of this humor strategy evince a descending tendency. Chandler, on the other hand, is mostly being ironic, violating the maxim of quality. His character also gradually changes but his sense of humor remains the same - ironic throughout the show, as follows from the instances of almost fixed frequency....

[pt] O estudo focaliza os estilos conversacionais emergentes da interação entre terapeutas e clientes no contexto de uma primeira sessão de terapia de família a partir de uma perspectiva teórica de integração entre as ordens institucionais e interacionais do discurso. A partir da análise dos dados, percebemos que a sessão de terapia apresenta uma configuração discursiva híbrida, demonstrando características de discurso institucional e de conversa espontânea, evidenciadas pela natureza do piso conversacional - ora configurando-se como típico da fala do especialista, ora apresentando-se colaborativo, característico de uma fala mais livre - observado no decorrer da interação entre terapeutas e clientes. Argumentamos, então, ao articular os conceitos de ordem institucional e ordem interacional, que essas instâncias de fala-em-interação com características de conversa cotidiana são contextualmente relevantes para a realização do mandato institucional peculiar à terapia de família e não desvios da organização institucional. / [en] This study focuses on the emergent conversational styles in the interaction between therapists and clients in a context of a first session of family therapy from a theoretical point-of-view which aims at integrating the institutional and interactional orders of discourse. It was possible to observe that the therapy session presents a hybrid discursive configuration, displaying the characteristics from both institutional and ordinary talk, which is evidenced by the nature of the conversational floor - at times configuring itself as expert talk, at times structuring itself as spontaneous talk -observed throughout the interaction between therapists and clients. Therefore, by articulating the concepts of institutional order and interactional order, we argue that these instances of talk-in-interaction, characterized as ordinary talk, are contextually relevant for the accomplishment of the institutional mandate peculiar to family therapy and not deviations from the institutional organization.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / A presente pesquisa apresenta uma anÃlise da conversaÃÃo de pessoas com transtornos mentais enfocando trÃs aspectos principais: um que investiga os marcadores conversacionais, outro que procura verificar como os participantes da conversaÃÃo interagem uns com os outros e finalmente um que analisa o fenÃmeno de relevÃncia. Para compreender a conversaÃÃo de pessoas com transtornos mentais, analisamos as conversas de trinta e cinco sujeitos, pacientes do Centro de AtenÃÃo de Quixadà (CAPS) nos anos de 1998, 1999 e 2000. Pela interpretaÃÃo dos resultados, chegou-se à conclusÃo de que tanto em situaÃÃes de surto como de nÃo surto, as pessoas com transtornos mentais dÃo seqÃÃncia aos turnos que exigem a formaÃÃo obrigatÃria e nÃo cancelÃvel de um par adjacente, usam mais sinais conversacionais pÃs-posicionados e utilizam mais os marcadores conversacionais convergentes e indagativos do que os divergentes. Confirmou-se tambÃm a hipÃtese de que em situaÃÃes de surto, essas pessoas apresentam um comprometimento maior no fenÃmeno da relevÃncia do que quando nÃo estÃo em surto. O estudo ressalta, ainda, que as conversas de pessoas com transtornos mentais contÃm elementos coerentes e relevantes, possibilitando reflexÃes sobre as concepÃÃes que defendem o isolamento dessas pessoas por as conceberem totalmente incapazes de um convÃvio social / This research presents an analysis of the conversation of mentally disturbed people focusing on three main aspects: it investigates the conversational markers used in their interactions, it verifies how conversation participants interact with one another and, finally, it analyses the relevance phenomenon. In order to reach our aim, we have analysed the conversation of thirty-five patients of the Centro de AtenÃÃo (CAPS) of QuixadÃ, in Brazilâs northeastern state of Cearà who had their conversations recorded during 1988, 1999 and 2000. The results of the analysis have indicated that both in periods of onset or not, mentally disturbed people give continuation to conversational turns that require the mandatory formation of na adjacency pair, use more post-positioned conversational signals than pre-positioned ones, and utilize more converging and enquiring conversational markers than diverging ones. The hypothesis that, when in crises, the conversation of mentally disturbed people present a greater weakening of the relevance phenomenon has also been confirmed. The study highlights that the conversation of such people is, to some extent, coherent and relevant. This fact calls for a revision as regards the conceptions that defend the isolation of these people by preconceiving them unable of social interaction

Scheduling a time to meet can be time-consuming, especially when coordinating with email. It could be challenging for business people when each participant is required to email back and forth to propose their availability, matching each other's time availability, and finding a suitable location to meet. It is even worse when participants must reschedule the entire meeting.   This thesis aims to design and develop an artificial intelligence (AI) scheduling assistant chatbot mobile app that could assist people in scheduling meetings efficiently in the business environment. The research process involves two rounds of design iterations. In the first design iteration, the goal was to explore and test the possible ways to design the chatbot. In the second design iteration, the goal was to learn from the first iteration and improve the design to fulfil the users' needs. The results implied five options for designers to consider when designing an AI assistant chatbot for the business environment. The considerations include the (1) maturity of natural language processing, (2) instructions to new users, (3) feedback provided by the AI assistant, (4) effort of typing messages, and (5) personality of the AI assistant. / Det kan vara tidskrävande att planera en tid att träffas, särskilt när man samordnar med e-post. Det kan vara utmanande för affärsmän när varje deltagare måste skicka e-post fram och tillbaka för att föreslå deras tillgänglighet, matcha varandras tillgänglighet och hitta en lämplig plats att möta. Det är ännu värre när deltagarna måste planera om hela mötet. Denna avhandling syftar till att utforma och utveckla en artificiell intelligens (AI) schemaläggningsassistent chatbot mobilapp som kan hjälpa människor att schemalägga möten effektivt i affärsmiljön. Forskningsprocessen innefattar två omgångar med design-iterationer. I den första designversionen var målet att utforska och testa möjliga sätt att utforma chatboten. I den andra designiterationen var målet att lära av den första iteration och förbättra designen för att uppfylla användarnas behov. Resultaten innebar fem alternativ för designers att överväga när de designade en AI-assistent-chatbot för affärsmiljön. Övervägandena inkluderar (1) mognad för naturlig språkbehandling, (2) instruktioner till nya användare, (3) feedback från AI-assistenten, (4) ansträngning att skriva meddelanden och (5) AI-assistentens personlighet.

La musique et la parole possèdent toutes deux un degré d’organisation temporelle i.e. de régularité dans le temps. Les stimuli de nature rythmique ont la particularité de pouvoir être anticipés par le cerveau et des études en linguistique et neurosciences ont montré que plus le cerveau est capable d’anticiper les évènements auditifs, meilleure est la qualité du traitement des stimuli. Les enfants sourds, bien que bénéficiant d’un input auditif de plus en plus précis grâce aux implants cochléaires et d’une prise en charge précoce, n’atteignent pas des niveaux de langage homogènes et souffrent de difficultés de perception en milieux bruyants ou lors de conversations. La situation conversationnelle présente un contexte complexe, nécessitant l’activation de la voie audio-motrice pour anticiper et s’adapter aux variations de la parole de son interlocuteur notamment au niveau temporel. Dans ce travail de thèse, nous avons cherché à analyser, grâce à des mesures électrophysiologiques et comportementales, si un entrainement rythmique actif de 30 minutes, pouvait avoir un effet sur les capacités de perception et d’accommodation temporelles de l’enfant sourd dans une tâche de dénomination en alternance avec un partenaire virtuel. Nous avons également testé les capacités rythmiques de ces enfants à différents niveaux de complexités. Les résultats montrent que les enfants sourds souffrent de difficultés à structurer les événements acoustiques selon différent niveaux de hiérarchie mais qu’un entrainement rythmique de 30 minutes versus une stimulation auditive, permet d’améliorer leurs compétences de perception et de production temporelles de la parole dans une situation d’interaction. / Music and speech both possess a certain degree of temporal organization i.e. a certain degree of regularity across time. Studies in linguistics and neuroscience have shown that the brain can extract regularities and use them to anticipate the forthcoming stimuli. It is furthermore established that the better the brain is able to anticipate auditory events, the better the quality of stimulus processing. Deaf children benefit from more and more precise auditory inputs due to advances in cochlear implants development, together with early rehabilitation interventions. However, a great majority of them do not achieve consistent language levels and have strong difficulties in noisy environments or conversations. The conversational situation presents a complex context, requiring the activation of the audio-motor path to anticipate and adapt to the variations of the speech of its interlocutor notably at the temporal level. In this thesis work, we have investigated the temporal perception and accommodation capacities of deaf children in a naming task alternating with a virtual partner, at both behavioral and electrophysiological levels. We have also tested whether an active rhythmic training lasting 30 minutes, could enhance these conversational abilities. Then, we have investigated the rhythmic abilities of these children at different levels complexities. The results show that deaf children suffer from difficulties in structuring acoustic events according to different levels of hierarchy but that a rhythmic training of 30 minutes versus an auditory stimulation, makes it possible to improve their skills of temporal perception and production of speech in a situation of interaction.

Natural language processing and Machine learning are subjects undergoing intense study nowadays. These fields are continually spreading, and are more interrelated than ever before. A case in point is text classification which is an instance of Machine learning(ML) application in Natural Language processing(NLP).Although these subjects have evolved over the recent years, they still have some problems that have to be considered. Some are related to the computing power techniques from these subjects require, whereas the others to how much training data they require.The research problem addressed in this thesis regards lack of knowledge on whether Machine learning techniques such as Word2Vec, Bidirectional encoder representations from transformers (BERT) and Support vector machine(SVM) classifier can be used for text classification, provided only a small training set. Furthermore, it is not known whether these techniques can be run on regular laptops.To solve the research problem, the main purpose of this thesis was to develop two separate conversational interfaces utilizing text classification techniques. These interfaces, provided with user input, can recognise the intent behind it, viz. classify the input sentence within a small set of pre-defined categories. Firstly, a conversational interface utilizing Word2Vec, and SVM classifier was developed. Secondly, an interface utilizing BERT and SVM classifier was developed. The goal of the thesis was to determine whether a small dataset can be used for intent classification and with what accuracy, and if it can be run on regular laptops.The research reported in this thesis followed a standard applied research method. The main purpose was achieved and the two conversational interfaces were developed. Regarding the conversational interface utilizing Word2Vec pre-trained dataset, and SVM classifier, the main results showed that it can be used for intent classification with the accuracy of 60%, and that it can be run on regular computers. Concerning the conversational interface utilizing BERT and SVM Classifier, the results showed that this interface cannot be trained and run on regular laptops. The training ran over 24 hours and then crashed.The results showed that it is possible to make a conversational interface which is able to classify intents provided only a small training set. However, due to the small training set, and consequently low accuracy, this conversational interface is not a suitable option for important tasks, but can be used for some non-critical classification tasks. / Natural language processing och maskininlärning är ämnen som forskas mycket om idag. Dessa områden fortsätter växa och blir allt mer sammanvävda, nu mer än någonsin. Ett område är textklassifikation som är en gren av maskininlärningsapplikationer (ML) inom Natural language processing (NLP).Även om dessa ämnen har utvecklats de senaste åren, finns det fortfarande problem att ha i å tanke. Vissa är relaterade till rå datakraft som krävs för dessa tekniker medans andra problem handlar om mängden data som krävs.Forskningsfrågan i denna avhandling handlar om kunskapsbrist inom maskininlärningtekniker som Word2vec, Bidirectional encoder representations from transformers (BERT) och Support vector machine(SVM) klassificierare kan användas som klassification, givet endast små träningsset. Fortsättningsvis, vet man inte om dessa metoder fungerar på vanliga datorer.För att lösa forskningsproblemet, huvudsyftet för denna avhandling var att utveckla två separata konversationsgränssnitt som använder textklassifikationstekniker. Dessa gränssnitt, give med data, kan känna igen syftet bakom det, med andra ord, klassificera given datamening inom ett litet set av fördefinierade kategorier. Först, utvecklades ett konversationsgränssnitt som använder Word2vec och SVM klassificerare. För det andra, utvecklades ett gränssnitt som använder BERT och SVM klassificerare. Målet med denna avhandling var att avgöra om ett litet dataset kan användas för syftesklassifikation och med vad för träffsäkerhet, och om det kan användas på vanliga datorer.Forskningen i denna avhandling följde en standard tillämpad forskningsmetod. Huvudsyftet uppnåddes och de två konversationsgränssnitten utvecklades. Angående konversationsgränssnittet som använde Word2vec förtränat dataset och SVM klassificerar, visade resultatet att det kan användas för syftesklassifikation till en träffsäkerhet på 60%, och fungerar på vanliga datorer. Angående konversationsgränssnittet som använde BERT och SVM klassificerare, visade resultatet att det inte går att köra det på vanliga datorer. Träningen kördes i över 24 timmar och kraschade efter det.Resultatet visade att det är möjligt att skapa ett konversationsgränssnitt som kan klassificera syften, givet endast ett litet träningsset. Däremot, på grund av det begränsade träningssetet, och konsekvent låg träffsäkerhet, är denna konversationsgränssnitt inte lämplig för viktiga uppgifter, men kan användas för icke kritiska klassifikationsuppdrag.

This paper evaluates and analyzes three conversational AI-platforms; Dialogflow (Google), Watson Assistant (IBM) and Teneo (Artificial Solutions) on how they perform based on a set of criteria; pricing model, ease-of-use, efficiency, experience working in the software and what results to expect from each platform. The main focus was to investigate the platforms in order to acquire an understanding of which platform would best be suited for enterprises. The platforms were compared by performing a variety of tasks aiming to answer these questions. The technical research was combined with an analysis of each company’s pricing model and strategy to get an understanding of how they target their products on the market. This study concludes that different softwares may be suitable for different settings depending on the size of an enterprise and the demand for complex solutions. Overall, Teneo outperformed its competitors in these tests and seems to be the most scalable solution with the ability to create both simple and complicated solutions. It was more demanding to get started in comparison with the other platforms, but became more efficient as time progressed. Some findings include that Dialogflow and Watson Assistant lacked capabilities when faced with  complex and complicated tasks. From a pricing strategy point of view, the companies are similar in their approach but Artificial Solutions and IBM has more flexible methods while Google has a fixed pricing strategy. Combining the pricing strategy and technical analysis this implicates that Teneo would be a better choice for larger enterprises while Watson Assistant and Dialogflow may be more suitable for smaller ones. / Det här arbetet evaluerar och analyserar tre konversationella AI-plattformar; Dialogflow (Google), Watson Assistant (IBM) och Teneo (Artificial Solutions) utifrån hur de presterar baserat på ett antal  kriterier; prismodell, enkel användning, effektivitet, upplevelse att arbeta i programvaran och vilka resultat man förväntar sig från varje plattform. Huvudsakligt fokus var att undersöka plattformarna för att få en uppfattning om vilken plattform som skulle passa bäst för företag. Plattformarna jämfördes genom att utföra en mängd olika uppgifter som syftade till att besvara dessa frågor. Den tekniska forskningen kombinerades med en analys av varje företags prismodell och prisstrategi för att få en uppfattning av hur de riktar sina produkter på marknaden. Denna studie drar slutsatsen att olika programvaror kan vara lämpliga för olika sammanhang beroende på ett företags storlek och dess efterfrågan på komplexa lösningar. Sammantaget överträffade Teneo sina konkurrenter i dessa tester och verkar vara den mest skalbara lösningen med förmågan att skapa både enkla och komplicerade lösningar. Det var mer krävande att komma igång i jämförelse med de andra plattformarna, men det blev mer effektivt med tiden. Vissa fynd inkluderar att Dialogflow och Watson Assistant saknade kapacitet när de mötte komplexa och komplicerade uppgifter. Från en prissättningsstrategisk synvinkel är företagen liknande i sin metod men Artificial Solutions och IBM har mer flexibla metoder medan Google har en fast prissättningstrategi. Genom att kombinera prisstrategi och teknisk analys innebär detta att Teneo skulle vara ett bättre val för större företag medan Watson Assistant och Dialogflow kan vara mer lämpade för mindre.

This thesis explores the use of advanced language models, specifically OpenAI’s Generative Pretrained Transformer (GPT), in the context of interactive tutoring systems built within a Unity-based game environment. The central problem addressed is whether the recent advancements in large language models make them feasible and useful to function as tutors specifically in providing meaningful, engaging, and educationally rich user interactions on a dialogue based learning platform developed by Fictive Reality. There is also a comparison on the effectiveness of GPT versus the model that previously powered the learning platform built in Rasa. The importance of this problem lies in offering people learning opportunities that might not otherwise be available to them, and in seeing if recent advancements in generative AI are sufficient for developing useful interactive AI tutors of soft skills. The Fictive Reality learning platform is powered by a Rasa model that generates appropriate responses to users in the context of roleplay-based learning scenarios while keeping an internal state of the progress of the dialogue. The project entails replacing this model with GPT and a comparison of their performance and respective merits. We also explored the potential for a hybrid model, leveraging the strengths of both systems. Using Rasa for internal state tracking and answering simpler queries, and utilizing GPT to handle those queries whose intent Rasa cannot determine. The first part of this project was integrating GPT with the existing functionality of the platform, this includes changes to the platform that allow people to create and play GPT powered learning scenarios and adopting the existing features and user interface. Additionally, prompt engineering GPT to act as a tutor and to stay within the context of a learning environment. Changes had to be made to the platform so that the already existing features of Rasa scenarios could be replicated in GPT scenarios. Finally there is a systematic comparison of the user experience and performance metrics when interacting with either a GPT or a Rasa chatbot in a learning scenario. Specifically these metrics are determined from the conversational flow between bot and user, the context and continuity, finish rate, chit-chat handling and length of average session. The results suggest a distinct user preference for the GPT model due to its superior conversational capabilities, despite Rasa’s faster response times and state-tracking feature. The study suggest that GPT is sufficient for creating useful learning scenarios in restricted contexts. Therefore we suggest that large language models can be leveraged in interactive learning systems, with potential impacts on edtech, AI in education, and conversational AI. / Detta examensarbete utforskar användningen av avancerade språkmodeller, särskilt OpenAI’s Generative Pretrained Transformer (GPT), tillsammans med interaktiva handledningssystem byggda i en Unity-baserad spelmiljö. Det centrala problemet som tas upp är om det är genomförbart och användbart att använda GPT som handledare. Vidare genomfördes också en jämförelse av effektiviteten hos GPT jämfört med en mer traditionell modell, Rasa, när det gäller att tillhandahålla meningsfulla, engagerande och lärorika interaktioner. Detta problem har betydelse för att erbjuda människor lärandemöjligheter som annars kanske inte skulle vara tillgängliga för dem och för att se om de senaste framstegen inom generativ AI är tillräckliga för användbar interaktiv handledning av mjuka färdigheter, så kallade soft skills". Lärplattformen Fictive Reality drivs av en Rasa-modell som genererar lämpliga svar till användare i samband med vissa inlärningsscenarier samtidigt som man behåller ett internt tillstånd av dialogens framsteg. Projektet syftar till att ersätta denna modell med GPT och göra en jämförelse av prestandan och hos respektive modell. Vi undersökte också potentialen för en hybridmodell som utnyttjar båda systemens styrkor genom att använda Rasa för intern tillståndsspårning och svara på enklare frågor, och använda GPT för att hantera de frågor vars avsikt Rasa inte kan avgöra. Den första delen av projektet var att integrera GPT med plattformens befintliga funktionalitet, detta inkluderar förändringar av plattformen som gör det möjligt för människor att skapa och spela GPT-drivna inlärningsscenarier med det befintliga användargränssnittet och funktioner för Rasa-drivna scenarier. Förändringar var tvungna att göras på plattformen så att de redan befintliga funktionerna i Rasa-scenarier kunde replikeras i GPT-scenarier. Slutligen gjordes en systematisk jämförelse av prestandan och användarupplevelsen när man interagerar med antingen en GPT- eller en Rasa-chatbot i ett inlärningsscenario. Resultaten tyder på en distinkt användarpreferens för GPT-modellen på grund av dess överlägsna konversationsförmåga, trots Rasa:s snabbare svarstider och tillståndsspårningsfunktion. Studien tyder på att GPT är tillräckligt för att skapa användbara lärande scenarier i begränsade sammanhang. Denna studie tyder på att stora språkmodeller kan utnyttjas i interaktiva inlärningssystem, med potentiella effekter på edtech, AI inom utbildning och konversations-AI-områden.

The study is based on qualitative research which aimed to show how four teachers interact in book talks with the students i connection with Swedish lessons in first grade. The study has focused on teachers´ questions to analyze the types of conversational genres that arise in connection with book talks in first grade.  Data was collected through interviews and observations. The interviews and observations were made at a school in the county of Stockholm. Four teachers´ Swedish lessons were observed and they were interviewed about how they describe and explain their use of book talks in Swedish lessons. Hultin´s categorization for speech genres and rhetorical and non-rhetorical questions´ (2006) served as main theoretical frameworks for analyzing the empirical material. From the interviews, I have come to the conclusion that the teachers are aware of different speech genres, and they use rhetorical and non-rhetorical questions for different purposes. The results of the study show that, depending on their purpose with book talk, teachers use different conversational genres.  The reults of the study show that a conversational genre dominated in the book talks that were examined but they also showed elements of other genres to a certain extent. It appears that two conversational genres were dominant in the teaching of teachers, that is, the teaching examination and the text oriented talk. On the other hand the culturally oriented talk and the informal book talk only appeared as compliments to the dominant genres. The study shows that teachers´ awareness of these genres is of great importance for developing the pupils´ reflective minds and skills.