[pt] A oferta de serviços por meio de interfaces conversacionais, ou chatbots, tem se tornado cada vez mais popular, com aplicações que variam de aplicativos de bancos e reserva de bilheteria a consultas em um banco de dados. No entanto, dado a quantidade massiva de dados disponível em alguns domínios, o usuário pode ter dificuldade em formular as consultas e recuperar as informações desejadas. Esta dissertação tem como objetivo investigar e avaliar o uso de recomendações na busca de informações numa base de dados de filmes através de chatbots. Neste trabalho, implementamos um chatbot por meio do uso de frameworks e técnicas da área de processamento de linguagem natural (NLP - Natural Language Processing). Para o reconhecimento de entidades e intenções, utilizamos o framework RASA NLU. Para a identificação das relações entre essas entidades, utilizamos as redes Transformers. Além disso, propomos diferentes estratégias para recomendações feitas a partir da ontologia de domínio. Para avaliação deste trabalho, conduzimos um estudo com usuários para avaliar o impacto das recomendações no uso do chatbot e aceitação da tecnologia por meio de um questionário baseado no Technology Acceptance Model (TAM). Por fim, discutimos os resultados do estudo, suas limitações e oportunidades de futuras melhorias. / [en] The offer of services provided through conversational interfaces, or chatbots, has become increasingly popular, with applications that range from bank applications and ticket booking to database queries. However, given the massive amount of data available in some domains, the user may find it difficult to formulate queries and retrieve the desired information. This dissertation investigates and evaluates the use of the recommendations in the search for information on a movie database through a chatbot. In this work, we implement a chatbot with the use of frameworks and techniques from the area of natural language processing (NLP). For the recognition of entities and intents, we use the RASA NLU framework. For the identification of relations between those entities, we use the Transformers networks. In addition, we propose different strategies for the recommendation from the domain ontology. To evaluate this work, we have conducted an empirical study with volunteer users to assess the impact of the recommendations on chatbot use and the acceptance of the technology through a survey based on the Technology Acceptance Model (TAM). Lastly, we discuss the results of this study, its limitations, and avenues for future improvements.

Chatbots are very widely used nowadays. However, much of the research on Chatbots have had a technology focus or has been limited to studies of adoption. To take advantage of the potential associated with chatbots, research that addresses the issues online users face when interacting with such programs is needed. The study described in this paper used the task-to technology fit theory to address the question of how individual characteristics and task/technology requirements influence the performance and utilization of chatbots. This paper used the quantitative methodology over two sets of data collected independently from two different populations. The first dataset of 100 respondents was obtained firstly through a structured questionnaire administered at Linnaeus University Campus in Växjö. The respondents are students in the university who use chatbots regularly. A second dataset was also collected from 20 participants through a practical test experiment with three different chatbots (Eliza, Rose, and Watson). The result and the data were then recorded through an online interview via the zoom application. The two datasets were analyzed quantitatively using comparative factor analysis with the aid of Smart PLS software. While few variables provided little support for the claims, the majority of the variables show strong support for the importance of task–technology fit, as a measure of chatbot utilization and performance based on individual characteristics as well as the task/technology requirements.

Chatbots are the new kind of technique for organizations to implement in their operations. Although chatbots have been around for a while they have spiked in the last few years. With this new technique also comes new ways for users to interact and communicate with the organizations. The traditional usability tests do not fully cover the chatbots characteristics and new methods must be created. This paper will create an evaluation model based on chatbots and theories about usability testing and apply the evaluating method to three commercial chatbots and their interaction with a user. The result shows that the evaluation model in some ways was successful and fulfilled its purpose while further research and development is necessary for the model to perform as desired.

Advancements in Artificial Intelligence, especially in the field of natural language processing have opened new possibilities for educational chatbots. One of these is a chatbot that can simulate a conversation between the teacher and the student for continuous learner support. In an up-scaled learning environment, teachers have less time to interact with each student individually. A resource to practice interactions with students could be a boon to alleviate this issue. In this thesis, we present a machine-learning model combined with a heuristic approach used in the creation of a chatbot. The machine learning model learns language understanding using prebuilt language representations which are fine-tuned with teacher-student conversations. The heuristic compares responses and picks the highest score for response retrieval. A data quality analysis is also performed on the teacher-student conversation dataset. For results, the best-base-cased language model performed best for text classification with a weighted F1-score of 0.70. The dataset used for the machine learning model showed consistency and completeness issues regarding labelling. The Technology Acceptance Model has been used to evaluate the model. The results of this evaluation show a high perceived ease of use, but a low perceived usefulness of the present solution. The thesis contributes with the innovative TUM (topic understanding model), an educational chatbot and an evaluation of the teacher-student chatroom corpus regarding the usage for text classification. / Teknologiska framsteg i artificiell intelligens, speciellt inom språkteknologi, har öppnat för nya möjligheter för chatbottar inom utbildningssektorn. Chatbots har sett en ökande användning i olika lärandeändamål. En av dessa är en chatbot som kan simulera en konversation mellan en lärare och en student för lärandestöd. När inlärning sker på en allt större skala, har lärare allt mindre tid att lägga individuellt på varje student. En resurs för att öva på interaktioner med studenter skulle därför kunna vara ett bra hjälpmedel. I denna masteruppsats presenteras en maskininlärnings modell kombinerad med ett heuristiskt tillvägagångsätt i skapandet av en chatbot. Maskininlärningsmodellen använder sig av färdigbyggda språkrepresentationer som är finjusterade med lärare-studentkonversationer. Heuristiken jämför svar och väljer den högsta poängen för svarshämtning. En datakvalité analys är också gjord på lärare-studentkonversations datasetet. För resultat, den BERT-baserade språkmodellen gav bäst resultat för textklassificering med en weigthed-F1- score på 0.70. Datasetet som användes för maskininlärningsmodellen visade konsistens och fullständighet problem rörande etiketter. Teknologi acceptans modellen har använts för att evaluera modellen. Resultatet av evalueringen visade hög upplevd användarvänlighet, men låg upplevd användbarhet. Detta arbete bidrar med TUM (topic understanding model), en utbildningschatbot och en evaluering av datasetet teacherstudent chatroom corpus för användning till textklassificering.

As advancements in chatbots and Large Language Models (LLMs) such as GPT-3.5 and GPT-4 continue, their applications in diverse fields, including psychology, expand. This study investigates the effectiveness of LLMs optimized through prompt engineering, aiming to enhance their performance in psychological applications. To this end, two distinct versions of a GPT-3.5-based chatbot were developed: a version similar to the base model, and a version equipped with a more extensive system prompt detailing expected behavior. A panel of professional psychologists evaluated these models based on a predetermined set of questions, providing insight into their potential future use as psychological tools. Our results indicate that an overly prescriptive system prompt can unintentionally limit the versatility of the chatbot, making a careful balance in instruction specificity essential. Furthermore, while our study suggests that current LLMs such as GPT-3.5 are not capable of fully replacing human psychologists, they can provide valuable assistance in tasks such as basic question answering, consolation and validation, and triage. These findings provide a foundation for future research into the effective integration of LLMs in psychology and contribute valuable insights into the promising field of AI-assisted psychological services. / I takt med att framstegen inom chatbots och stora språkmodeller (LLMs) som GPT-3.5 och GPT-4 fortsätter utvidgas deras potentiella tillämpningar inom olika områden, inklusive psykologi. Denna studie undersöker effektiviteten av LLMs optimerade genom prompt engineering, med målet att förbättra deras prestanda inom psykologiska tillämpningar. I detta syfte utvecklades två distinkta versioner av en chatbot baserad på GPT-3.5: en version som liknar bas-modellen, och en version utrustad med en mer omfattande systemprompt som detaljerar förväntat beteende. En panel av professionella psykologer utvärderade dessa modeller baserat på en förbestämd uppsättning frågor, vilket ger inblick i deras potentiella framtida användning som psykologiska verktyg. Våra resultat tyder på att en överdrivet beskrivande systemprompt kan ofrivilligt begränsa chatbotens mångsidighet, vilket kräver en noggrann balans i specificiteten av prompten. Vidare antyder vår studie att nuvarande LLMs som GPT-3.5 inte kan ersätta mänskliga psykologer helt och hållet, men att de kan ge värdefull hjälp i uppgifter som grundläggande frågebesvaring, tröst och bekräftelse, samt triage. Dessa resultat ger en grund för framtida forskning om effektiv integration av LLMs inom psykologi och bidrar med värdefulla insikter till det lovande fältet av AI-assisterade psykologtjänster.

This paper presents Cyberbot, a conversational user interface (CUI) designed to be used for information retrieval and interaction with smart Internet of Things (IoT) devices (smart objects), in an office environment. Built through the Research Through Design method, tested together with a selection of smart objects, the artifact was used to explore how CUIs can help office workers in their everyday working activities and the impact they can have on the usability in interaction with IoT, as opposed to more commonplace types of interfaces. Cyberbot was evaluated through user tests, surveys and semi-structured interviews. Results show that Cyberbot can be a welcome addition to an office and is useful for getting quick overviews of information from company intranet websites or smart objects that gather data from its environment, at the expense of more in-depth and detailed information. In general, Cyberbot can be said to be a usable and effective interface for IoT interaction. / Denna studie presenterar Cyberbot, ett konversationsgränssnitt (CUI) desginat för att användas för informationsinhämtning samt interaction med smarta Internet of Things (IoT) föremål (smart objects) i en kontorsmiljö. Byggd medelst forskningsmetoden Research Through Design, testad tillsammans med en utvald selektion av smarta föremål, så användes denna artefakt för att undersöka hur CUIs kan hjälpa kontorsanställda i deras dagliga liv och den påverkan de kan ha på användarvänligheten i interaktion med IoT, till skillnad från mer vanliga typer av gränssnitt. Cyberbot utvärderades genom användartester, enkäter och semi-strukturerade intervjuer. Resultaten visar att Cyberbot kan vara ett välkommet tillskott till kontor och är användbar för att få snabba överblickar över information ifrån företags intranätsidor eller smarta föremål som samlar in data från sin omgivning, på bekostnad av mer djuplodande och detaljerad information. Generellt så kan Cyberbot ses som ett användbart och effektivt gränssnitt för IoT-interaktion.

Fenomenet konversationshandel beskriver en växande handelsplattform världen över som främjar interaktivitet för att skapa personlig köpupplevelse och relation mellan konsumenter och företag. Detta fenomen växer i takt med utvecklingen av artificiell intelligens (AI) och robotar för att underlätta och eventuellt ersätta den mänskliga faktorn inom e-handel. Detta syns då termen “konversationshandel” börjat användas mer och mer sedan 2010 för varje år som går. Konventionell e-handel har fått en negativ syn på sig från många konsumenter på grund av dess ytliga och kalla gränssnitt vilket skapar risk för bortval från konsumenter till att handla på vissa webbplatser. På grund av detta har företag och utvecklare funnit behovet av interaktivitet och mer personliga köpupplevelser som flera konsumenter kräver i deras handel. Detta har lett till att e-handeln behöver efterlikna den traditionell butikshandeln mer men med hjälp av ny teknik. Konversationshandel har dock en begränsad användning på grund av sin nya tillkomst och därmed är syftet i denna studie att undersöka vad som skiljer konversationshandel från konventionell e-handel och undersöka den svenska befolkningens uppfattning om aspekter inom konversationshandel. Studien har använt sig av en systematisk litteraturstudie och en digital enkätstudie för att besvara syftets frågeställning. Resultatet visar på att interaktivitet, AI och konversationell assistans urskiljer konversationshandel från konventionell e-handel och att svenska konsumenter främst använder chatt över röstassistenter men att handel via båda har en liten användning. Slutsatserna av studien förklarar att konversationshandel har en förmåga att komplettera konventionell e-handel med hjälp av interaktivitet. Digitala assistenters förmåga att efterlikna människor är en svår uppgift att genomföra men kan bidra till en större personlig upplevelse för kunder. Sverige är ett land som inte utnyttjat konversationhandel till en stor del, vilket beror på en uppfattad komfortabilitet för konventionell e-handel och begränsade resurser för implementation. / The phenomenon of conversational commerce describes a growing trading platform around the world that is enhancing interactivity for the sake of creating a personal purchase experience and relation between consumers and businesses. This phenomenon is growing with the development of artificial intelligence (AI) and robots in order to favour and eventually replace the human factor in e-commerce. This shows because the term “conversational commerce” has started being used more and more since 2010 and is growing every year that passes. Conventional e-commerce has received a negative view from many consumers due to its superficial and cold interface, which creates a risk of consumers opting out of shopping on certain websites. Because of this, companies and developers have found the need for interactivity and more personalized shopping experiences that several consumers demand in their shopping. This has led to e-commerce needing to emulate the traditional retail trade further but with the help of new technology. Conversational commerce is however, used to a limited extent due to its new origins and thus the purpose of this study is to investigate what distinguishes conversational commerce from conventional e-commerce and to examine the Swedish population’s perception of aspects of conversational commerce. The study used a systematic literature study and a digital survey to answer the scientific questions. The results show that interactivity, AI and conversational assistance distinguish conversational commerce from e-commerce and that the theoretical Swedish population mainly uses chat over voice assistants, but that commerce via both has little use. The conclusions of the study explain that conversational commerce has an ability to complement conventional e-commerce through interactivity. The ability of digital assistants to imitate people is a difficult task to accomplish but can contribute to a greater personal experience for customers. Sweden is a country that has not used conversational commerce much, which is due to the perceived convenience of conventional e-commerce and limited resources for implementation.

With the proliferation of digital services and technologies, industries, including thebanking sector, are looking for innovative ways to improve customer experience. The criticalaspect of improving customer experience is increased customer satisfaction, which helps toattract new customers and to retain existing ones. Chatbots have emerged as a promisingsolution, offering a convenient and personalized way for customers to interact with banks andother financial institutions. This thesis analyses the impact of chatbots on customer satisfactionthrough a comparative study of customer satisfaction between Sweden and India. By providinginsights into the similarities and differences in customer perceptions between the two countries,this thesis aims to contribute to the existing literature on the effectiveness of chatbots. This studytakes a socio-technical perspective to explore how customers perceive satisfaction when usingchatbots in their financial services. The findings shed light on the impact of chatbots on customersatisfaction in different markets and contribute to the understanding of the adoption of chatbots inroutine banking assistance, especially in a cross-cultural context. In designing effective strategiesfor using chatbot technology to improve customer experience in the digital age, the findings arevaluable for banking institutions and policymakers.

Despite being explored and constantly improved through the years, onboarding of new hires in corporate organizations has remained a challenge. Many of the issues can be linked to a lack of communication between the organization and the new employee, as well as the common nature of these environments where information is spread across job titles and sources. This thesis discusses the feasibility of implementing a basic chatbot that will allow new hires to ask questions and request varied information at all times, using an interface such as a messaging app. This research explores the way chatbots should be designed in order to be effective, reliable and enjoyable from a user experience perspective. The chatbot was developed using the Chatfuel platform and tested by new employees at a corporate environment. The users were requested to explore the chatbot freely and then asked to answer a survey. The interactions were also recorded and analyzed from in both qualitative and quantitative ways (chat logs and analytics). The study proves that an onboarding chatbot is a useful tool for new hires and can be used as a communication facilitator between the organization and the new hires during the first weeks of employment, and also after that, serving as an information source and a broadcasting method. The chatbot gives the new hires an accessible source of information that helps on the process of getting to speed, and enables a positive experience that increases familiarity in the new workplace.

The digital age has brought many new technologies that are disrupting the way that companies interact with customers. Automated technologies are one aspect of technological development in recent days, and it has changed how regular business operations are performed. Chatbots, for instance, has changed the way that customers interact with companies. Facilitating interaction and being available around the clock to serve customers who have questions. This study investigated how automated technologies are used in CRM-activities of companies and how the implementation of such technologies can lead to customer engagement. CRM, Customer Relationship Management, is a core business strategy that aims to strengthen customer relationships through co-creating value with the use of supporting technologies. Customer engagement is a measurement of a customer's involvement with a firm over time, with the end goal of customers becoming “fans” that advocate and generate word-of-mouth for the firm. This study investigates how companies utilize automated technologies in CRM-activities and if these technologies can help lead to customer engagement. The study aims to answer the following research question: How are automated technologies, more specifically chatbots, used in existing CRM strategies to further create customer engagement? The study conducted investigates these technologies within five companies that utilize chatbot-technologies from chatbot-providers. The chatbot-providers are also part of the study to get a more comprehensive view of the functionalities and purpose of these technologies. The contrasting view of companies and chatbot-providers is used to answer the sub-purposes of the thesis regarding how companies implement automated technologies. A qualitative study with interviews was conducted, along with secondary data from written documentary sources. The data were analyzed in a theoretical thematic analysis in which themes were developed from the theoretical framework. The findings of the study show that despite the potential of automated technologies, the most prominent use is for customer service operations. The potential for automated technologies in CRM is greater than what was found in the interviewed companies. The findings also demonstrate that automated technologies facilitate connection, interaction, and increases customer satisfaction, which can lead to customer engagement. The findings also show that there is a discrepancy in the views of companies who use chatbots in their business processes and the chatbot-providers that help implement them.