This diploma thesis deals with the influence of implementation of a chatbot technology for sales promotion and customer services. The aim of the paper is to evaluate effects, that influence the consumer shopping behaviour after the implementation of chatbot and to identify factors that have an impact on the communication with this technology. Then the goal of this thesis is to create recommendations for companies considering the implementation of chatbot technology, which would also reflect the specifics of a Czech consumer. Data for primary research are obtained through questionnaire surveys and in-depth interviews. Quantitative research is used to process factor analysis that identifies factors influencing communication with chatbots.

Chatbotar används idag i olika område, som utbildning, kundtjänst och vård, och harmånga fördelar. Inom kundtjänst kan chatbotar bidra till att spara tid och kostnader förföretag. För att chatbotar ska användas brett av användare måste de ge en njutbar ochnyttig upplevelse. Användarupplevelsen hos chatbotar kan undersökas ur olikaperspektiv. I det här arbetet syftar vi till att undersöka användarupplevelse hos tvåchatbotar som vi har implementerat på olika sätt och med olika tekniker. En avchatbotarna är regelbaserad och den andra är AI-baserad. Vi använde en enkät för attgenomföra den här undersökningen. 24 personer testade båda chatbotarna och sedanutvärderade dem ur ett pragmatiskt och hedoniskt perspektiv. Resultatet av den härundersökningen visar att AI-chatboten tillhandhåller en bra upplevelse och denbedömdes av deltagarna som uppfinningsrik och intressant. Å andra sidan fickdeltagarna inte en bra upplevelse med användning av den regelbaserade chatboten meninte en dålig upplevelse heller. När det gäller effektivitet är chatboten som använderAI-tekniker mycket effektivare än den andra. Att båda chatbotarna är identiska på allaegenskaper förutom tillvägagångsättet att implementera, kan man dra slutsatsen attimplementation påverkar användarupplevelsen.

The investigation aims to provide an answer to if a chatbot is a potential complement to an internal service desk of a company. The work has centered around developing a chatbot able to handle simple Q&A-interaction of the internal service desk of Bisnode, the company in question. The chatbot acted as an proof of concept, which then was tested by 15 individuals. The testing was done with pre- defined user scenarios, where the test person ultimately had to fill in a questionnaire with statements related to the overall experience. By summarizing the user evaluations from the questionnaires, combined with an SWOT analysis, the work concluded that a chatbot is indeed a potential complement to an internal service desk of a company, if it handles Q&A-interaction.

This study was conducted to investigate the empathy between human chatbot interactions amongcomputer science students at Uppsala University, Sweden. This was done by exploring howparticipants perceive anthropomorphic chatbots as machines or humans, the existence of verbalabuse during human chatbot interactions, and the expectation of chatbot helpfulness dependingon gender dynamics. A semi-structured interview methodology with five students was conductedfor qualitative data collection. The collected data was manually analyzed using thematic analysis.The results of this study found that there is empathy in human chatbot interaction, regardless ofwhether participants perceive anthropomorphic chatbots as humans or machines. However, thelevel of empathy is generally low as participants frustrate when they are dissatisfied with theresponse of chatbots and exit the chatbots without expressing their frustration, and they usuallyforget their frustration and come again with other questions another time. The study also showsthat participants might expect more help and politeness if chatbots are more likely to be female.

We study human-chatbot collaborative conversation systems that enable humans to leverage AI chatbot outputs during an online conversation with others. We evaluate response quality in two collaborative systems and compare them with human-only and chatbot-only settings. Both collaborative systems present AI chatbot results as suggestions but encourage the synthesis of human and chatbot responses to different extents. We also examine the influence of chatbot choices, including both retrieval-based and generation-based methods, and the number of suggestions on collaborative systems. Experimental results show that our collaborative systems can significantly improve the efficiency to formulate a response and improve its quality compared with a human-only system while sacrificing the fluency and humanness of the messages. Compared with a chatbot, collaborative systems can provide answers that are more fluent, human-like, and informative. We also found that the retrieval-based chatbots perform better than the generation-based one from all aspects. The optimal number of chatbot suggestions is one, and showing more suggestions has reduced user efficiency. / Master of Science / Artificial Intelligence (AI) systems have become remarkably interactive and accurate with them becoming an integral part of our life. The increasing use of personal assistants like Siri and the application of AI in important real-world tasks such as medical imaging and diagnosis show that AI can perform as good as trained human experts. Organizations today are expanding at a rapid rate and need to service millions of customers concurrently to remain competitive in the market. With the recent success of AI chatbots, the collaboration of Human and AI to augment customer service management is one of the most sought out solutions to this requirement. A service flow where virtual agents and people work together can be a boon to the industry by making the human agents smarter with a bot "whispering" in their ears. We present the design of various collaborative systems we have developed and discuss the improvements in response efficiency and quality due to them in multiple online user experiments. The results of this study can be used to improve conversational chat systems that assist human agents to improve their response time and quality and identify features of the AI agent that are most beneficial for improving the conversation.

Many colleges face a lack of academic and emotional support for their computer science students. Previous research into this problem produced Maria, a virtual teaching assistant (TA). This initial version of Maria was able to answer student questions, provide error explanations, and praise students for effort on programming assignments. This research continues work on the Maria project with three design goals: (1) reducing obstacles to use of Maria, (2) allowing Maria to provide better academic support, and (3) allowing Maria to provide better emotional support (with less focus on this goal). Improvements were made to the initial version of Maria, including increasing the number of questions that Maria could answer, allowing Maria to suggest questions for students to ask, and adding longer back-and-forth dialogs between Maria and students. Following this, Maria was deployed to students for an evaluation. The evaluation revealed that certain features were popular (including the longer dialogs and easier access to error explanation) and that Maria was unable to provide relevant answers to many questions asked by students. Using data from the evaluation, more improvements were made to Maria to address some of her shortcomings and build on her popular features. Answers to more questions were added for questions about testing, general knowledge questions, questions about many other topics. Many of these new answers used the popular back-and-forth dialog feature. Additionally, this research discusses a system that could be used to automate the creation of new answers for Maria or any virtual teaching assistant using crowdsourcing. / Master of Science / Many colleges face a lack of academic and emotional support for their computer science students. Previous research into this problem produced Maria, a virtual teaching assistant (TA). This initial version of Maria was able to answer student questions, provide error explanations, and praise students for effort on programming assignments. This research continues work on the Maria project with three design goals: (1) reducing obstacles to use of Maria, (2) allowing Maria to provide better academic support, and (3) allowing Maria to provide better emotional support (with less focus on this goal). Improvements were made to the initial version of Maria, including increasing the number of questions that Maria could answer, allowing Maria to suggest questions for students to ask, and adding longer back-and-forth dialogs between Maria and students. Following this, Maria was deployed to students for an evaluation. The evaluation revealed that certain features were popular (including the longer dialogs and easier access to error explanation) and that Maria was unable to provide relevant answers to many questions asked by students. Using data from the evaluation, more improvements were made to Maria to address some of her shortcomings and build on her popular features. Answers to more questions were added for questions about testing, general knowledge questions, questions about many other topics. Many of these new answers used the popular back-and-forth dialog feature. Additionally, this research discusses a system that could be used to automate the creation of new answers for Maria or any virtual teaching assistant using crowdsourcing.

Chatbotar används idag på en mängd olika sätt. Det är vanligt att chatbotar implementeras som kundservice i olika företag, men att använda sig av en chatbot i en kommun är inte lika förekommande. Chatboten Sally är den första i sitt slag som en servicebot i Helsingborgs kommun och svarar på frågor om Vuxenutbildningen. Denna uppsats är ett resultat av en kvalitativ studie vars syfte har varit att ta reda på vilka förväntningar som fanns på chatboten innan införandet och vad resultat blev. Intervjuerna har gjorts med utvecklarna av chatboten på Helsingborgs kommun, chef samt kommunvägledare på Kontaktcenter. Dessa har haft någon interaktion med chatboten. Studien har identifierat vad som motiverade kommunen till att skapa en chatbot och vad resultat blev, och kunskapsprodukten av den här studien är en genomgång av vad man kan tänka på när man skapar en bot i en kommun.

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-10-11T12:09:21Z No. of bitstreams: 1 nataliaduartemarcao.pdf: 3039217 bytes, checksum: 3d02a3fd20a157117bfde56d24c4edcc (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-10-16T13:53:22Z (GMT) No. of bitstreams: 1 nataliaduartemarcao.pdf: 3039217 bytes, checksum: 3d02a3fd20a157117bfde56d24c4edcc (MD5) / Made available in DSpace on 2018-10-16T13:53:22Z (GMT). No. of bitstreams: 1 nataliaduartemarcao.pdf: 3039217 bytes, checksum: 3d02a3fd20a157117bfde56d24c4edcc (MD5) Previous issue date: 2018-09-10 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O trabalho aqui apresentado insere-se no projeto Multilingual Knowledge Base ou simplesmente m.knob (www.mknob.com), o qual tem por objetivo criar um aplicativo web e vem sendo desenvolvido no Laboratório FrameNet Brasil de Linguística Computacional – FN-Br – (SALOMÃO, 2009). De forma mais geral, a FN-Br vem explorando a implementação das teorias da Semântica de Frames e da Gramática de Construções através da criação de recursos linguísticos computacionais, como o Lexicon e o Constructicon do Português Brasileiro (PB) – um repertório de construções disponível online. Dessa maneira, esta dissertação busca embasamento em pressupostos teóricos relacionados à Linguística Cognitiva, tais como a Semântica de Frames (FILLMORE, 1982) e a Gramática das Construções de Berkeley (KAY & FILLMORE, 1999) e segue a metodologia de análise da Berkeley FrameNet (FILLMORE ET AL. 2003). Nesse contexto, o objetivo dessa dissertação é apresentar a descrição e a modelagem linguístico-computacional (cf. DIAS-DA-SILVA, 1996) das construções Interrogativas QU- em PB na base construcional da FrameNet Brasil, a qual sustenta o aplicativo web m.knob. A pesquisa se justifica pelo fato de o chatbot – uma interface de recomendação de atrações turísticas baseada em compreensão de língua natural – empregado no aplicativo não suportar que o usuário interaja através de sentenças interrogativas. Nesse sentido, a descrição e modelagem das construções QU- se faz necessária de modo a contribuir para que o usuário, durante a interação com o aplicativo, busque informações adicionais acerca das atrações recomendadas através de perguntas. Para fomentar essa funcionalidade nova, esta dissertação traz uma proposta de modelagem de onze construções QU- em PB. Ao final, aplica-se um teste de prova de conceito para avaliação do modelo proposto. / This work is part of the Multilingual Knowledge Base project or simply m.knob (www.mknob.com). It aims to create a web application and has been developed at the FrameNet Brasil Laboratory of Computational Linguistics - FN-Br - (SALOMÃO, 2009). More generally, FN-Br has been exploring the implementation of Framing Semantics and Construction Grammar theories through the creation of computational linguistic resources, such as the Lexicon and Constructicon of the Brazilian Portuguese (PB) - a repertoire of constructions available online. In this way, this Master’s thesis is based on the theoretical assumptions of Cognitive Linguistics, such as Frame Semantics (FILLMORE, 1982) and the Berkeley Constructions Grammar (KAY & FILLMORE, 1999). Our analytical methodology is the one by Berkeley FrameNet (FILLMORE et al., 2003). In this context, this thesis aims to present the description and the linguistic-computational modeling (cf. DIAS-DA-SILVA, 1996) of the Interrogative Wh-constructions of the Brazilian Portuguese in the constructional base of FrameNet Brasil, which supports the web application m.knob. The research is justified by the fact that the chatbot - a tourist attraction recommendation interface based on natural-language understanding - employed in the application does not support the user interacting through interrogative sentences. In that sense, the description and modeling of Wh-constructions are necessary in order to help the user, during interaction with the application, look for additional information about the recommended attractions through questions. To foster this new function, this thesis presents a proposal for modeling eleven Wh-Constructions in Brazilian Portuguese. At the end, a proof of concept test is applied to evaluate the proposed model.

Combitech is aiming to develop their own intraweb and they are investigating if it ispossible to implement a bot that handles menial task to make daily tasks easier on their employees. This thesis investigates whether a chatbot which handles the booking of me-etings will be useful to the company. After implementing the chatbot and testing it with 5 employees the conclusion was made that a chatbot for meetings was not very useful.However, the testers were very positive about having a chatbot on the intraweb if it isimplemented for tasks which are timely or complicated to solve. / Combitech håller på att utveckla ett eget intranät och undersöker om det är möjligtatt implementera en bot som hanterar enkla uppgifter för att förenkla vardagen för sinamedarbetare. Denna rapport undersöker om en chatbot som hanterar bokningen av mötenkan vara användbar för företaget. Efter implementering av chatboten och testning medhjälp av 5 medarbetare blev slutsatsen att en chatbot för mötesbokning inte är särskiltanvändbar för företaget. Däremot var alla testare väldigt positiva till att ha en chatboti intranätet om den implementeras för att hantera mer tidskrävande eller kompliceradeuppgifter.

Medan teknologin ständigt utvecklas strävar mer och mer företag efter att automatisera arbetsuppgifter. Användningen av chatbotar sociala media blir allt mer vanligt som en ersättning av kundtjänster då de både är effektivt och pålitligt, samt att företagen sparar pengar. I denna projektrapport har ett IT-företag frågat efter en implementation och konceptbevis på en integration mellan deras applikation och meddelande tjänsten Microsoft Teams. De vill att det ska vara möjligt att använda en chatbot i Teams för att kunna fylla i information i ett formulär som sedan skickas som ticket till deras system. De som arbetar i systemet ska då kunna se ticketen och direkt kunna börja arbeta på uppgiften. I denna rapport beskrivs en lösning på detta problem samt, hur det fungerar, hur det implementeras samt en diskussion på lösningen. Diskussionen fokuserar på den slutliga lösningen, användartesterna, relaterat arbete, etiska problem samt framtida arbete. / As technology evolves, more and more companies strive to automate as much work as possible. The use of chatbots in messaging apps is becoming more and more common as replacements in the operation of customer service because of the efficiency and reliability it poses. Replacing customer service with chatbots also saves companies money. In this paper an IT company have asked for the implementation and proof of concept of integration between their server-side application, to Microsoft Teams. They want it to be possible to use a chatbot in Teams to send tickets containing data to their system in order to be handled by users of their application. In this paper a solution to this problem is described as well as discussions about the chosen implementation. The discussion is focused on the final solution, user tests, related work, ethical issues as well as future work. User tests of the implemented solution are included in the results. The conclusion of the work is that this implementation is useful and saves time and effort both for the customer and the employees. There is much potential for future work to be expanded upon, including specialized Ticket Forms and implemented AI.