La presente investigación tiene como objetivo analizar cómo los chatbots de Facebook se relacionan con la intención de compra de entradas a exposiciones de arte contemporáneo de Lima Metropolitana. El chatbot es una herramienta automática de atención al cliente que permite conversar con posibles consumidores rápidamente sin la necesidad de una persona de por medio. Las redes sociales han hecho más accesible su implementación en toda clase de negocios por lo que se busca analizar el impacto de la herramienta en la industria del arte. Para conseguir la información cualitativa necesaria se usaron entrevistas a profundidad y focus group a expertos y público objetivo mientras que para la información cuantitativa se usó una encuesta online a personas que usaron la herramienta de las distintas instituciones investigadas. Después de analizar los datos se encontró una relación positiva entre el nivel de satisfacción con el chatbot de Facebook de las instituciones de arte y las distintas variables de intención de compra. Sin embargo, todavía es necesario desarrollar la calidad de las herramientas implementadas actualmente para aprovechar todo su potencial. / This investigation has as an objective to analyze how Facebook chatbots relate with the purchase intention of tickets for contemporary art expositions in Metropolitan Lima. The chatbot is an automatic tool for customer support that allows conversations with potential clients in a fast manner without the need of a person in between. Social media has made accessible its implementation in all type of businesses so there is a need to analyze the impact of this tool in the art industry. To get the qualitative information it was used in-depth interviews and focus groups to experts and the target audience while for the quantitative information it was used an online poll to people that have used the tool of the different researched institutions. After analyzing the data a positive relation was found between the level of satisfaction with the Facebook chatbot from the art institution and the different purchase intention variables. However, there is still a need to develop the quality of the tools that are implemented right now to take advantage of its full potential. / Trabajo de investigación

Chatbots are increasing in popularity, both as a valuable tool for commercial purposes but also as an area of academic research. It has been successfully applied in multiple application areas such as healthcare, education, customer service and work support. Despite the successful use and application of chatbots research has identified two main user experience challenges: technical- and hedonistic limitations. Users also have a hard time understanding what level of competence and functionality they can expect from a chatbot. To combat these challenges, further research is needed to understand the user adoption and perception of chatbots. This study aims to research both the experiences and usage of customer service chatbots but also explore what preferences customers have when it comes to interacting with customer service chatbots or human customer service representatives in e-commerce. The theoretical technology acceptance model called Unified Theory of Acceptance and Use of Technology (UTAUT) was applied for this study. A quantitative study comparing customer service chatbots versus human customer service representatives was performed by conducting a questionnaire. The results from the questionnaire indicated that the participants of the study in general rated human customer service representatives higher in most question categories. The results indicate that the participants would prefer to interact with human service representatives over customer service chatbots which indicates that there is still opposition and challenges surrounding the adoption of customer service chatbots in e-commerce.

This study is rooted in the Service Dominant Logic (SDL) theory and explores the phenomenon of value co-destruction, the context being chatbot-mediated customer service, with a focusing on chatbots affiliation with value-co-destruction and on the impacts outsideof the dyadic relationship between the service provider and receiver, described as the broader consumer ecosystem. It is evident that the relationship between businesses and consumers is being revolutionized through the implementation of chatbots, claiming benefits such as improved operational efficiency, and non-time specific service. However, Plé (2017) argues that SDL research is overly focused on value co-creation and not enough on value codestruction, as well as being too narrowly focused on specific factors, therefore more studies are needed in exploring the effects on the broader ecosystem involved. Therefore, this study aims to explore the value co-destruction and its effects on the broader consumer ecosystem. The research was conducted using a qualitative method, through an interpretivist research philosophy to collect data in the form of the subjective and personal experiences of customers. It also utilized an abductive approach to allow for the generation of new insights through combining established knowledge, empirical evidence, and theory to formulate a proposition and interpretation beyond what is initially apparent. Through semi-structured interviews, 9 participants with prior experiences of customer service orientated interactions with chatbots across various industries were conducted. To analyze the data, a thematic analysis was employed where the findings indicate that in chatbot interactions which result in value co-destruction the effects affect a broader number of stakeholders in the consumer ecosystem such as family and friends.

Research questions: How can businesses improve the trustworthiness and competence of chatbots for consumers? Purpose: The purpose is to investigate how chatbots' perceived competence and trustworthiness impact consumers' willingness to interact with them. The study             aims to identify patterns or businesses to advance  Method: The study relies on qualitative data and semi-structured interviews with consumers about their experience with chatbots. For deeper and more detailed data, it is conducted by interviewing experts with years of experience with chatbots. The data has been then analyzed through thematic analysis.  Conclusion: The study identifies 10 key themes connected to the theories and the two main elements of this research. Companies are advised to give top priority to the identified themes when integrating chatbots into their customer service operations.

Previous studies highlight the major technological advances in artificial intelligence (AI) and the benefits of introducing artificial intelligence systems in the form of chatbots in, among other things, customer support. They also describe the impact of chatbots on organizations. Artificial intelligence technology provides computers with the ability to display human-like traits, such as reasoning, creativity, learning and planning. The advantages of artificial intelligence-based chatbots can be, for example, improved processes in customer support, reduced costs and less burden on employees regarding monotonous tasks. It is important beforehand, and also during the course of the project, to have a well-functioning operation from an organizational and technical perspective. If there are problems within the business that are not addressed, this can have consequences. It is therefore not only the implementation of the chatbot that should be focused on in order to succeed with the change, but also organizational and technical parts of the business that will very likely be affected by the implementation. About 87% of artificial intelligence system implementation projects fail. This is because several factors relating to the project or to the product implemented have not been taken into account. Therefore, this study investigates the research question: What critical success factors need to be considered when implementing chatbots in customer support? The purpose of this research question is to identify which critical success factors (CSFs) are most necessary for decision makers to be aware of. This is to have better conditions to succeed with the implementation and use of artificial intelligence-based chatbots in customer support. To answer this question, a qualitative study has been carried out in the form of a case study with interviews as a data collection method. Respondents relevant to the research area have been interviewed. In conducting this study, thematic analysis has been used as a data analysis method to identify, analyze and report patterns in the collected data. These have then been used to answer our question. The result generated 3 themes, 15 categories and 45 codes. The categories are divided according to critical success factors from previous studies identified using the codes extracted from the interviews. These critical success factors are in turn divided into three different themes that represent the different characteristic features among the factors, these are organizational, technical and resource factors. The critical success factors that are included in the theme organizational factors are: Lack of benefits visibility, Change management, Organizational culture, Organizational structure, Resistance and Ambiguous strategic vision. In the technical factors: Ethics issues, Insuf icient quantity of data, Integration complexity, Low data quality, Data governance issues, Security and confidentiality and Scalable and flexible system. The theme resource factors include: Selection of vendors and High cost of AI. The conclusion of this study suggests that there are a variety of critical success factors of organizational, technical, and resource-related nature that are important to consider when implementing chatbots in customer support. Those considered to be of the highest priority are the technical factors. Next in line are the organizational factors Organizational culture, Organizational structure and Resistance, followed by the rest of the factors in the same theme. The factors considered to be of the lowest priority are the ones included in the theme resource factors. There are also critical success factors from previous studies, that this study builds upon, that are considered not important enough to take into account. The critical success factors generated in this investigation are discussed regarding how the data extracted has a connection to a certain critical success factor. In the discussion, critical success factors generated are compared with the critical success factors from previous studies. Since this is a case study containing a limited number of respondents, there is of course a likelihood that there are more critical success factors affecting customer support chatbot implementation projects. However, the critical success factors that this study resulted in are considered to be of higher priority and can therefore have the greatest impact on such a project.

In the realm of Artificial Intelligence (AI), Chatbots have become a widely used technology for streamlining the services between organizations and consumers. The present study advances the understanding of human-AI interactions by producing new knowledge on chatbot affordances and customer value creation. While many organizations jumped on the trend and implemented chatbots as a new communication channel, less is known about chatbot affordances and how value creation take place from a consumer perspective. Using affordance concept as a lens this study uncoversnine affordances of banking chatbots and showcase how consumer value is created from these affordances. By undertaking a qualitative research approach, the study conducted semi structured interviews with six banking consumers from three different banks in Sweden to inductively gain rich insights. The results show how consumer value yield and fluctuate with different banking chatbot affordances in the three phases of Affordance Concept; Affordance Perception, Affordance Actualization and Affordance Effects. The study’s contribution is twofold. First, it proposes a conceptual model which includes useful insights to guide researchers towards understanding the consumer value creation from exploring the affordances of banking chatbots. Secondly, the study supports financial institutes to improve chatbot service’s usability and usefulness by incorporating the value creating affordances. Furthermore, the study identified areas for future research which provides a groundwork for researchers seeking to engage with the research area.

Assessing a conversational agent’s understanding capabilities is critical, as poor user interactions could seal the agent’s fate at the very beginning of its lifecycle with users abandoning the system. In this thesis we explore the use of paraphrases as a testing tool for conversational agents. Paraphrases, which are different ways of expressing the same intent, are generated based on known working input by performing lexical substitutions and by introducing multiple spelling divergences. As the expected outcome for this newly generated data is known, we can use it to assess the agent’s robustness to language variation and detect potential understanding weaknesses. As demonstrated by a case study, we obtain encouraging results as it appears that this approach can help anticipate potential understanding shortcomings, and that these shortcomings can be addressed by the generated paraphrases. / Att bedöma en konversationsagents språkförståelse är kritiskt, eftersom dåliga användarinteraktioner kan avgöra om agenten blir en framgång eller ett misslyckande redan i början av livscykeln. I denna rapport undersöker vi användningen av parafraser som ett testverktyg för dessa konversationsagenter. Parafraser, vilka är olika sätt att uttrycka samma avsikt, skapas baserat på känd indata genom att utföra lexiska substitutioner och genom att introducera flera stavningsavvikelser. Eftersom det förväntade resultatet för denna indata är känd kan vi använda resultaten för att bedöma agentens robusthet mot språkvariation och upptäcka potentiella förståelssvagheter. Som framgår av en fallstudie får vi uppmuntrande resultat, eftersom detta tillvägagångssätt verkar kunna bidra till att förutse eventuella brister i förståelsen, och dessa brister kan hanteras av de genererade parafraserna.

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.

Today's digital progress highlights disruptive digital technology as a numerous impact on marketing and societies. Artificial intelligence in conjunction with chatbots creates a seamless and powerful tool for customer service. Companies are estimated to save 30% in customer service related costs while chatbots can assist with up to 80% of routine questions. Such numbers are considered appealing for companies to implement chatbots with the aim of streamlining and maximizing profits. An important aspect that can be neglected by companies is quality, which is an essential factor for customers' user experience. In this context, the user experience creates satisfaction and repeated use. An understanding of chatbot quality increases the chances of successfully implementing and developing the right tool with the aim of satisfying customers both efficiently and qualitatively in customer service. From such a perspective, the survey has concentrated around the user experience for a chatbot interaction. How is the quality of the user experience and what factors have a direct impact on chatbots in customer service? What characteristics should a chatbot possess to enhance the user experience? The survey has been carried out with the choice of a multimethod research with a qualitative approach. Some quantitative measures have also been used to concretize certain values, but the main focus area has been on the survey of the user experience from a qualitative perspective. Two theoretical frameworks have been used to analyze the user experience of chatbots. These focus on qualitative factors that have an effect on the user experience. The result shows that the quality of the user experience is considered high on the selected chatbot for simpler questions and less complex matters. The factors response-relevance, response-comprehensibility, dialogue-result and dialogue-efficiency showed a relatively high user experience. The characteristics that should be implemented in a chatbot are understandability, reliability, responsiveness and assurance. This creates a confirmation, satisfaction and repeat use for the user experience. Influential factors that have been analyzed in the survey are the user's question formulation. A qualitative advantage is created for users who formulate formal questions, while more personal questions increase the margin of error of the chatbot's interpretive ability to answer the question. The perspective can be seen as a lack of quality from the chatbot or as a missing feature. A final factor impacted in the user experience is technology anxiety, which has been shown to be lowered for users after interacting with a chatbot. Users do not have a high sense of difficulty to use a chatbot but may hold a sense of uncertainty and discouragement. With greater experience and a perceived positive user experience of a chatbot, these factors are drastically reduced. The 5 conclusion leads to the user experience being of high quality in terms of interaction with a chatbot, while several influencing factors are highlighted. Future research is recommended to replicate the study on a larger scale and with more personal and complicated cases for analysis of the user experience of chatbot quality.

As research in Artificial Intelligence (AI) advances, it is crucial to focus on having seamless communication between humans and machines in order to effectively accomplish tasks. Smooth human-machine communication requires the machine to be sensible and human-like while interacting with humans, while simultaneously being capable of extracting the maximum information it needs to accomplish the desired task. Since a lot of the tasks required to be solved by machines today involve the understanding of images, training machines to have human-like and effective image-grounded conversations with humans is one important step towards achieving this goal. Although we now have agents that can answer questions asked for images, they are prone to failure from confusing input, and cannot ask clarification questions, in turn, to extract the desired information from humans. Hence, as a first step, we direct our efforts towards making Visual Question Answering agents human-like by making them resilient to confusing inputs that otherwise do not confuse humans. Not only is it crucial for a machine to answer questions reasonably, it should also know how to ask questions sequentially to extract the desired information it needs from a human. Hence, we introduce a novel game called the Visual 20 Questions Game, where a machine tries to figure out a secret image a human has picked by having a natural language conversation with the human. Using deep learning techniques like recurrent neural networks and sequence-to-sequence learning, we demonstrate scalable and reasonable performances on both the tasks. / Master of Science