With an increase in alternative forms of work, people are no longer limited to traditional office spaces. The aim for a healthier integration of private and work comes with the advantages of experimenting with new technologies. As a result, coworking spaces are spreading through the urban centers. But our way of dealing with work is still marked by our corporate-focused past. This project aims to explore how can co-working spaces occupy a more meaningful role by connecting people with their interests. My interest is to unveil the social rules of the space and turn interactions between coworkers more pleasant and easy. The result is a reflection about the future of collaborative workplaces. The success of the experiments reflect the openness of most co-workers and hosts. On a higher level the project gave me a better understanding of how AI could help to improve the social aspect of our workplaces.

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.

From a system developer's perspective, designing a spoken dialogue system can be a time-consuming and difficult process. A developer may spend a lot of time anticipating how a potential user might interact with the system and then deciding on the most appropriate system response. These decisions are encoded in a dialogue strategy, essentially a mapping between anticipated user inputs and appropriate system outputs. To reduce the time and effort associated with developing a dialogue strategy, recent work has concentrated on modelling the development of a dialogue strategy as a sequential decision problem. Using this model, reinforcement learning algorithms have been employed to generate dialogue strategies automatically. These algorithms learn strategies by interacting with simulated users. Some progress has been made with this method but a number of important challenges remain. For instance, relatively little success has been achieved with the large state representations that are typical of real-life systems. Another crucial issue is the time and effort associated with the creation of simulated users. In this thesis, I propose an alternative to existing reinforcement learning methods of dialogue strategy development. More specifically, I explore how XCS, an evolutionary reinforcement learning algorithm, can be used to find dialogue strategies that cover large state spaces. Furthermore, I suggest that hand-coded simulated users are sufficient for the learning of useful dialogue strategies. I argue that the use of evolutionary reinforcement learning and hand-coded simulated users is an effective approach to the rapid development of spoken dialogue strategies. Finally, I substantiate this claim by evaluating a learned strategy with real users. Both the learned strategy and a state-of-the-art hand-coded strategy were integrated into an end-to-end spoken dialogue system. The dialogue system allowed real users to make flight enquiries using a live database for an Edinburgh-based airline. The performance of the learned and hand-coded strategies were compared. The evaluation results show that the learned strategy performs as well as the hand-coded one (81% and 77% task completion respectively) but takes much less time to design (two days instead of two weeks). Moreover, the learned strategy compares favourably with previous user evaluations of learned strategies.

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.

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.

Conversational interfaces (CIs) have been a trending topic in recent years. As of the last decade, CIs have emerged with the aim of simplifying human-machine interactions and found a wide use case in the market. For example, Siri and Google Assistant are some of the most well-known CIs developed by the tech giants Apple and Google. The digital landscape has evolved from web, to mobile apps, to recently CIs. Nowadays, CIs, in particular chatbots and voicebots, are becoming increasingly common. Whether navigating the web or messaging on a phone, it is likely that CIs have been confronted offering the user help.However, CIs have not managed to reach a large-scale use. Furthermore, the reasons regarding the challenges faced by CIs as well as their usability are not greatly explored. In this thesis, we explore the most relevant uses of CIs and the reasons hindering a widespread use of CIs. Our goal is to provide an insight into CIs’ uses and list the reasons regarding the challenges faced by CIs. The research study followed a mixed method approach connecting an explorative qualitative literature study, a survey and an interview. The data was collected by using a systematic mapping approach for it being more suitable for conducting an effective literature review. The survey and the interview were conducted in order to confirm the findings.According to our research, it was found that the most common use cases of CIs were in customer service, sales, travel and bookings, education, healthcare and as voice assistants. The most prominent challenges faced by CIs were poor usability, language processing and understanding, speech recognition and natural language generation and security and privacy. As a conclusion, the future looks promising for CIs, however, they need to be furher researched and developed in order to help them reach a widespread use in the future. / Konversation Gränssnitt (CIs) har varit ett trendande ämne de senaste åren. Sedan det senaste decenniet har CIs kommit fram i syfte att förenkla interaktioner mellan människor och maskiner och har hittat ett brett användningsfall på marknaden. Det digitala landskapet har utvecklats från webb, till mobila appar till nyligen CI. Numera blir CIs, i synnerhet chatbots och voicebots, allt vanligare. Vare sig du navigerar på webben eller meddelanden i en telefon, är det troligt att CIs har konfronterats med att erbjuda användaren hjälp.CIs har dock inte lyckats uppnå storskalig användning. Dessutom är orsakerna till de utmaningar som CIs står inför och deras användbarhet inte utforskas i hög grad. I den här avhandlingen undersöker vi de mest relevanta användningarna av CIs och orsakerna till en utbredd användning av CIs. Vårt mål är att ge en inblick i CI: s användningar och lista orsakerna till de utmaningar som CIs står inför. Forskningsstudien följde en blandad metodstrategi som ansluter en utforskande kvalitativ litteraturstudie, en undersökning och en intervju. Uppgifterna samlades in med hjälp av en systematisk kartläggningsätt för att göra dem mer lämpliga för att genomföra en effektiv litteraturgranskning. Undersökningen och intervjun genomfördes för att bekräfta resultaten.Enligt vår forskning konstaterades att de vanligaste användningsfallen för CIs var kundservice, försäljning, resor och bokningar, utbildning, sjukvård och som röstassistenter. De mest framstående utmaningarna för CIs var dålig användbarhet, språkhantering och förståelse, taligenkänning och naturlig språkgenerering och säkerhet och integritet. Sammanfattningsvis ser framtiden lovande ut för CIs, men de måste undersökas och utvecklas ytterligare för att hjälpa dem att uppnå utbredd användning i framtiden.

Interaction with digital products and interfaces concern more and more of human decision-making and the problems regarding environmental, financial and social sustainability are consequences much due to our behavior. The issues and goals of sustainable development therefore implies how we have to think differently about digital design. In this paper, we examine the adequacy of influencing sustainable behavior with a data-driven design approach, applying a conversational user interface. A case study regarding the United Nation’s goals of technological development and economic distribution was conducted, to see if a hypothetical business with a proof-of-concept digital product could be effective in influencing where companies base their operations. The test results showed a lack of usability and influence, but still suggested a potential with language-based interfaces. Even though the results could not prove anything, we argue that leveraging data analysis to design for sustainable behavior could be a very valuable strategy. A data-driven approach could enable ambitions of profit and user experience to coincide with those of sustainability, within a business organization.