Chatbots are increasing in popularity and interacting with humans via written language. Previous research has looked at chatbots within several domains, but not towards women’s general sexual and reproductive health. This offers a need to extend the small body of current research.  This report aimed to do so by describing women’s experiences of a sexual and reproductive health chatbot used as a decision support tool. The chatbot was designed based on a user-centered approach, allowing women to express desired personality traits in a person when discussing the topic. This resulted in the design creation of two chatbots with two different personalities. Exploratory Wizard of Oz studies were conducted with 6 users by simulating interaction with both chatbots operated by a human. Users were followed up with a survey and interview creating insights to their experiences with each chatbot. Findings resulted in contributing to research with proposed guidelines for how to design a sexual and reproductive health chatbot. / Chatbots blir allt mer populära och interagerar med människor genom skriftligt språk. Tidigare forskning har utforskat olika användningsområden för chatbots, men kvinnors sexuella och reproduktiva hälsa har inte varit en av dessa områden. Detta skapar ett behov att expandera den nuvarande smala forskningen. Denna studie syftar till att göra det genom att beskriva kvinnors erfarenheter av en chatbot för sexuell och reproduktiv hälsa som används som ett beslutsstödsverktyg. Chatboten utformades utifrån ett användarcentrerat tillvägagångssätt, vilket tillät kvinnor att uttrycka önskade personlighetsdrag hos en person som man diskuterar ämnet med. Detta resulterade i ett design skapade av två chatbots med två olika personligheter. Wizard of Oz studier genomfördes med 6 användare genom att simulera interaktionen med båda chatbots drivna av en människa. Deltagarna följdes upp med en enkät, följt av en intervju för bättre insikt till deras erfarenhet med varje chatbot. Resultaten resulterade i att bidra till forskning med förslag på riktlinjer för hur man utformar en chatbot för sexuell och reproduktiv hälsa.

This study explores the ways a Swedish state agency works with equality during the making oftheir AI-chatbots. The study explores an AI-team’s work in dealing with questions of equality,with equality through own experiences and values, the study has also placed its central themesaround equality work and language. The study does this with help of what Sara Ahmed callsinstitutionalized work. The study concludes that what Edmund Husserl calls the natural attitudeto working with equality constitutes a risk of leaving out perspectives and experiences. Workingwith Swedish as the state language, the study discusses how the use of language might representwhiteness and what effects this use of language may have on the interaction with their chatbot.The study has based its material around interviews made with six members of the state agenciesAI-team as well as programmed words and sentences in the chatbot. / <p>Ingen anmärkning</p> / Nej

In the rapidly evolving landscape of artificial intelligence, the potential of large language models (LLMs) remains a focal point of exploration, especially in the domain of education. This research delves into the capabilities of AI-enhanced chatbots, with a spotlight on the "Teacher Assistant" &amp; "Study Buddy" approaches. The study highlights the role of AI in offering adaptive learning experiences and personalized recommendations. As educational institutions and platforms increasingly turn to AI-driven solutions, understanding the intricacies of how LLMs can be harnessed to create meaningful and accurate educational content becomes paramount.The research adopts a systematic and multi-faceted methodology. At its core, the study investigates the interplay between prompt construction, engineering techniques, and the resulting outputs of the LLM. Two primary methodologies are employed: the application of prompt structuring techniques and the introduction of advanced prompt engineering methods. The former involves a progressive application of techniques like persona and template, aiming to discern their individual and collective impacts on the LLM's outputs. The latter delves into more advanced techniques, such as the few-shot prompt and chain-of-thought prompt, to gauge their influence on the quality and characteristics of the LLM's responses. Complementing these is the "Study Buddy" approach, where curricula from domains like biology, mathematics, and physics are utilized as foundational materials for the experiments.The findings from this research are poised to have significant implications for the future of AI in education. By offering a comprehensive understanding of the variables that influence an LLM's performance, the study paves the way for the development of more refined and effective AI-driven educational tools. As educators and institutions grapple with the challenges of modern education, tools that can generate accurate, relevant, and diverse educational content can be invaluable. This thesis not only contributes to the academic understanding of LLMs and provides practical insights that can shape the future of AI-enhanced education, but as education continues to evolve, the findings underscore the need for ongoing exploration and refinement to fully leverage AI's benefits in the educational sector

Conversational agents are becoming increasingly common in our day to day lives. We can speak to our phones, our cars and our smart home devices. Despite these advances, the current conversational agents are still far from perfect. The complexities of language as well as the technologies that are used to enable conversational agents pose many challenges to the people designing and developing them. This study aims to bring light to how practitioners design and develop conversational agents that exist out there today. Interviews were conducted with 11 practitioners that have been creating conversational agents for various industries and use cases. The results show that practitioners face a range of challenges when creating conversational agents, including collecting data about the target users during the design process as well as integrating the conversational agent with other systems. The study suggests that practitioners may benefit from involving users early on in the design process. It also advocates for HCI educators to prepare future graduates for designing conversational agents, through training them in human conversation and communication. This study gives insights into how conversational agents are built today, the processes that are followed and the challenges that are faced by the people creating them. / Konverserande system blir allt vanligare i våra dagliga liv. Vi kan prata med våra telefoner, våra bilar och våra smarta hem enheter. Trots dessa framsteg är de nuvarande konverserande systemen fortfarande långt ifrån perfekta. Språkets komplexitet och de teknologier som används, ställer många utmaningar för de människor som designar och utvecklar dem. Denna studie syftar till att belysa hur yrkesarbetare designar och utvecklar konverserande system. Intervjuer genomfördes med 11 yrkesarbetare som har skapat konverserande system för olika branscher och användningsområden. Resultaten visar att yrkesarbetarna står inför en rad utmaningar när de skapar dessa system, bland annat att samla in data om användarna under designprocessen och att integrera konverserande system med andra system. Studien tyder på att utövare kan ha nytta av att involvera användare tidigt i designprocessen. Studien förespråkar också att utbildare inom människa-datorinteraktion bör förbereda studenter för att utforma konverserande system, genom att utbilda dem i mänsklig konversation och kommunikation. Denna studie ger insikter i hur konverserande system är uppbyggda idag, de processer som följs och de utmaningar som människorna som skapar dem står inför.

京都大学 / 新制・論文博士 / 博士(医学) / 乙第13571号 / 論医博第2297号 / 新制||医||1069(附属図書館) / (主査)教授 川上 浩司, 教授 渡邉 大, 教授 村井 俊哉 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

Abstract This paper discussed the increasing use of artificial intelligence (AI) in various industries, particularly in the form of chatbots. Chatbots are AI-powered systems that interact with humans in order to support, collect, and deliver information. The technology is used to streamline internal workflow, improve customer experience (CX) and reduce business costs. The essay noted that chatbots promised to provide streamlined service and hence ameliorate interactions between customers and companies. Because of this, chatbots as a technology is projected to have a valuation of 102$ billion by the year 2026 and is in general seen as vital in the development of accessible technology. In spite of this, consumers’ acceptance of chatbots is in relative terms low, and users’ wishes are shown to be ignored or rejected by firms implementing the technology. In part because of this low rate of acceptance, a majority of chatbot projects are expected to fail. However, the present literature demonstrated a fragmented explanation as to why this is the case. The authors hence used a qualitative research strategy to describe the important factors to account for in the implementation of chatbots. The main data collection was done through semi-structured interviews with respondents involved in these implementations. Furthermore, to be able to use the knowledge already present, a semi-systematic literature review was conducted. Through the primary collection of data, the authors presented several factors that affect chatbot implementation; including the workload before launching a chatbot, the role of chatbot suppliers, meeting user expectations, and the need for building sufficient competence in the chatbot. The literature review then enabled the authors to conduct a detailed analysis of the presented results. The analysis presents the study’s compiled data from both the conducted interviews and the literature review to demonstrate the four main influencers in chatbot implementation; chatbot supplier input, company input, customer input, and total output. The authors hope that the finding will provide a starting point for further research and assist managers in better navigating the complex stages of chatbot implementation.

The release of ChatGPT has really set new standards for what an artificial intelligence chatbot should be. It has even shown its potential in answering university-level exam questions from different subjects. This research is focused on evaluating its capabilities in programming subjects. To achieve this, coding questions taken from software engineering exams were posed to the AI (N = 23) through an experiment. Then, statistical analysis was done to find out how good of a student ChatGPT is by analyzing its answer’s correctness, degree of completion, diversity of response, speed of response, extraneity, number of errors, length of response and confidence levels. GPT-3.5 is the version analyzed. The experiment was done using questions from three different programming subjects. Afterwards, results showed a 93% rate of correct answer generation, demonstrating its competence. However, it was found that the AI occasionally produces unnecessary lines of code that were not asked for and thus treated as extraneity. The confidence levels given by ChatGPT, which were always high, also didn't always align with response quality which showed the subjectiveness of the AI’s self-assessment. Answer diversity was also a concern, where most answers were repeatedly written nearly the same way. Moreover, when there was diversity in the answers, it also caused much more extraneous code. If ChatGPT was to be blind tested for a software engineering exam containing a good number of coding questions, unnecessary lines of code and comments could be what gives it away as being an AI. Nonetheless, ChatGPT was found to have great potential as a learning tool. It can offer explanations, debugging help, and coding guidance just as any other tool or person could. It is not perfect though, so it should be used with caution.

The following thesis investigates how Transportstyrelsen, a Swedish authority, can use chatbots in their customer service to increase effectivity, save resources and be more available to their customers. Therefore, this report aims to map the effects a chatbot-implementation would have on Transportstyrelsen’s business and what technical and organisational factors that contribute to a successful implementation of the chatbot. To answer these questions, a qualitative approach has been used where participants from other authorities have been interviewed about their chatbot-implementations. The experiences from Skatteverket, Pensionsmyndigheten and Lantmäteriet are compared to the situation on Transportstyrselsen with the help from socio-technical theory, diffusion theory and change management theory. The discussion concludes that the difficulties with the implementation does not lie with the technology itself and that Transportstyrelsen should allocate a lot of time for AI-trainers, include people from the legal, communication, and IT divisions in the implementation team, have an interagency strategy, vision and goal, prepare for measurements, share the progress with the organisation, include the sceptics in the development and consider the benefits of using the same chatbot algorithm as the other authorities.

The Rich Communication Service (RCS) aims to be the default messaging protocol in mobile devices. The native integration of RCS opens up the possibility of RCS chatbots replacing single-purpose apps. This report analyzes this possibility through in-depth experimentation of the chatbot functionality, followed by user testing of the chatbot features. The report found that the RCS chatbot could replace mobile applications since many solve a specific task and follow a closed-loop system. The report also identified RCS as a solution to the accumulated unused apps users have on their mobile devices. This report also attempts to fill the research gap on the current situation of RCS and the reasons behind the different rollout rates of RCS globally. The examination of interviews with RCS stakeholders was the basis of a stakeholder analysis. This analysis found that the Mobile Network Operators have different standings regarding RCS. Some see potential, and some see issues with RCS hindering the global rollout. Furthermore, the reluctance from Apple and the high involvement of Google are reviewed and contrasted with the answers from the interviews. / Rich Communication Service (RCS) siktar på att bli det nya standardprotokollet för att skicka meddelanden mellan mobiler. RCS potentiella integration i mobilens egna meddelande applikation öppnar upp för möjligheten att RCS- chatbotar även kan ersätta enklare appar med begränsade användningsområden. Denna rapport undersöker den här möjligheten genom experimenterande med chatbotens funktionalitet följt av användartester på chatbotens olika funktioner. Rapporten fann att chatboten i RCS har goda möjligheter att ersätta enklare mobila applikationer då många löser specifika uppgifter som följer ett system med en sluten slinga. Rapporten identifierade även RCS som en lösning till den mängd oanvända appar som användare har i sina mobiler. Denna rapport försöker även fylla det forskningsgap som finns angående RCS nuvarande situation och anledningarna till de olika adoptionshastigheterna i världen. Granskning av de intervjuer som hölls med olika intressenter på marknaden blev grunden för den intressentanalys som presenteras. Denna analys visade att teleoperatörer har olika ställningar gentemot RCS. Vissa ser potential medan andra främst ser problem vilket hindrar spridningen av RCS. Dessutom analyserades Apples ovilja att anamma RCS samt Googles

What do you think of when you hear the word chatbot? A helpful assistant when booking flight tickets? Maybe a frustrating encounter with a company’s customer support, or smart technologies that will eventually take over your job? The field of chatbots is under constant development and bots are more and more taking a place in our everyday life, but how well do they really understand us humans?  The objective of this thesis is to investigate how capable two open source chatbots are in understanding human language when given input containing spelling errors, synonyms or faulty syntax. The study will further investigate if the bots get better at identifying what the user’s intention is when supplied with more training data to base their analysis on.  Two different chatbot frameworks, Botpress and Rasa, were consulted to execute this experiment. The two bots were created with basic configurations and trained using the same data. The chatbots underwent three rounds of training and testing, where they were given additional training and asked control questions to see if they managed to interpret the correct intent. All tests were documented and scores were calculated to create comparable data. The results from these tests showed that both chatbots performed well when it came to simpler spelling errors and syntax variations. Their understanding of more complex spelling errors were lower in the first testing phase but increased with more training data. Synonyms followed a similar pattern, but showed a minor tendency towards becoming overconfident and producing incorrect results with a high confidence in the last phase. The scores pointed to both chatbots getting better at understanding the input when receiving additional training. In conclusion, both chatbots showed signs of understanding language variations when given minimal training, but got significantly better results when provided with more data. The potential to create a bot with a substantial understanding of human language is evident with these results, even for developers who are previously not experienced with creating chatbots, also taking into consideration the vast possibilities to customise your chatbot.