Dissertations / Theses on the topic 'E-commerce Artificial Intelligence'

L'Intelligence Artificielle est présente dans tous les aspects de notre vie à l'ère du Big Data. Elle a entraîné des changements révolutionnaires dans divers secteurs, dont le commerce électronique et la finance. Dans cette thèse, nous présentons quatre applications de l'IA qui améliorent les biens et services existants, permettent l'automatisation et augmentent considérablement l'efficacité de nombreuses tâches dans les deux domaines. Tout d'abord, nous améliorons le service de recherche de produits offert par la plupart des sites de commerce électronique en utilisant un nouveau système de pondération des termes pour mieux évaluer l'importance des termes dans une requête de recherche. Ensuite, nous construisons un modèle prédictif sur les ventes quotidiennes en utilisant une approche de prévision des séries temporelles et tirons parti des résultats prévus pour classer les résultats de recherche de produits afin de maximiser les revenus d'une entreprise. Ensuite, nous proposons la difficulté de la classification des produits en ligne et analysons les solutions gagnantes, consistant en des algorithmes de classification à la pointe de la technologie, sur notre ensemble de données réelles. Enfin, nous combinons les compétences acquises précédemment à partir de la prédiction et de la classification des ventes basées sur les séries temporelles pour prédire l'une des séries temporelles les plus difficiles mais aussi les plus attrayantes : le stock. Nous effectuons une étude approfondie sur chaque titre de l'indice S&amp;P 500 en utilisant quatre algorithmes de classification à la pointe de la technologie et nous publions des résultats très prometteurs<br>Artificial Intelligence has penetrated into every aspect of our lives in this era of Big Data. It has brought revolutionary changes upon various sectors including e-commerce and finance. In this thesis, we present four applications of AI which improve existing goods and services, enables automation and greatly increase the efficiency of many tasks in both domains. Firstly, we improve the product search service offered by most e-commerce sites by using a novel term weighting scheme to better assess term importance within a search query. Then we build a predictive model on daily sales using a time series forecasting approach and leverage the predicted results to rank product search results in order to maximize the revenue of a company. Next, we present the product categorization challenge we hold online and analyze the winning solutions, consisting of the state-of-the-art classification algorithms, on our real dataset. Finally, we combine skills acquired previously from time series based sales prediction and classification to predict one of the most difficult but also the most attractive time series: stock. We perform an extensive study on every single stocks of S&amp;P 500 index using four state-of-the-art classification algorithms and report very promising results

L'Intelligence Artificielle est présente dans tous les aspects de notre vie à l'ère du Big Data. Elle a entraîné des changements révolutionnaires dans divers secteurs, dont le commerce électronique et la finance. Dans cette thèse, nous présentons quatre applications de l'IA qui améliorent les biens et services existants, permettent l'automatisation et augmentent considérablement l'efficacité de nombreuses tâches dans les deux domaines. Tout d'abord, nous améliorons le service de recherche de produits offert par la plupart des sites de commerce électronique en utilisant un nouveau système de pondération des termes pour mieux évaluer l'importance des termes dans une requête de recherche. Ensuite, nous construisons un modèle prédictif sur les ventes quotidiennes en utilisant une approche de prévision des séries temporelles et tirons parti des résultats prévus pour classer les résultats de recherche de produits afin de maximiser les revenus d'une entreprise. Ensuite, nous proposons la difficulté de la classification des produits en ligne et analysons les solutions gagnantes, consistant en des algorithmes de classification à la pointe de la technologie, sur notre ensemble de données réelles. Enfin, nous combinons les compétences acquises précédemment à partir de la prédiction et de la classification des ventes basées sur les séries temporelles pour prédire l'une des séries temporelles les plus difficiles mais aussi les plus attrayantes : le stock. Nous effectuons une étude approfondie sur chaque titre de l'indice S&amp;P 500 en utilisant quatre algorithmes de classification à la pointe de la technologie et nous publions des résultats très prometteurs<br>Artificial Intelligence has penetrated into every aspect of our lives in this era of Big Data. It has brought revolutionary changes upon various sectors including e-commerce and finance. In this thesis, we present four applications of AI which improve existing goods and services, enables automation and greatly increase the efficiency of many tasks in both domains. Firstly, we improve the product search service offered by most e-commerce sites by using a novel term weighting scheme to better assess term importance within a search query. Then we build a predictive model on daily sales using a time series forecasting approach and leverage the predicted results to rank product search results in order to maximize the revenue of a company. Next, we present the product categorization challenge we hold online and analyze the winning solutions, consisting of the state-of-the-art classification algorithms, on our real dataset. Finally, we combine skills acquired previously from time series based sales prediction and classification to predict one of the most difficult but also the most attractive time series: stock. We perform an extensive study on every single stocks of S&amp;P 500 index using four state-of-the-art classification algorithms and report very promising results

Mestrado em Gestão de Sistemas de Informação<br>Nos últimos anos, tem havido um grande avanço tecnológico e as novas tecnologias têm estado cada vez mais presentes no nosso dia-a-dia, mesmo sem nos apercebermos, um destes casos é a inteligência artificial (IA). Um dos grandes impactos que a IA tem no nosso dia-a-dia é no e-commerce. O e-commerce é conhecido como o modelo de negócio que utiliza as diversas plataformas eletrónicas para fazer comércio online de bens. Este estudo centra-se na análise da perceção dos consumidores relativamente à influência da inteligência artificial no e-commerce, com particular enfoque nos chatbots. O presente trabalho considera a apresentação de um modelo conceptual, baseado, principalmente, na Teoria Unificada de Aceitação e Uso da Tecnologia 2 (UTAUT2), juntamente com a análise das variáveis latentes que foram identificadas como necessárias. A abordagem escolhida foi a quantitativa, baseando-se na realização de um inquérito, que decorreu durante o mês de Junho, onde foi obtido uma amostra de 569 inquiridos. A análise dos dados foi realizada através do Partial Least Squares (PLS), sendo utilizado o software Smart PLS 3.0. Por fim, o resultado da análise dos dados recolhidos comprovou a validade do modelo apresentado, sendo que os fatores que mais influenciaram a intenção dos consumidores de utilizarem chatbots no e-commerce são o Hábito, Utilidade, Facilidade de Uso, Atitude e Diversão Percebida.<br>Over the years, there has been a great technological advance and new technologies are increasingly present in our daily lives, even when we do not realize it, one of these cases is artificial intelligence (AI). One of the major impacts AI has on our daily lives is on e-commerce. E-commerce is known as the business model that uses the various electronic platforms to trade goods online. Thus, this study focuses on the analysis of consumers' perception of the influence of AI on e-commerce, specifically on chatbots. Additionally, the present work considers the presentation of a conceptual model, based mainly on the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), together with the analysis of the latent variables that were identified as necessary. The quantitative approach chosen was based on a survey that was online in June and that obtained a sample of 569 answers. The data analysis was performed using Partial Least Squares (PLS), using the Smart PLS 3.0 software. Finally, the result of this analysis proved the validity of the presented model and the factors that had the greatest influence on consumers' intention to use chatbots in e-commerce are Habit, Utility, Ease of Use, Attitude and Perceived Fun.<br>info:eu-repo/semantics/publishedVersion

Mestrado em Gestão de Sistemas de Informação<br>Numa altura em que as tecnologias web estão cada vez mais presentes nas nossas vidas, a forma como lidamos com elas evolui. O E-business tem-se desenvolvido permitindo que determinados produtos e serviços cheguem a um maior número de pessoas num maior espaço geográfico com um custo mais reduzido. As empresas adaptam-se a esta nova realidade e dispõem de novas, interativas e mais versáteis opções de compra. A presente dissertação tem como principal objetivo estudar o E-business e a ajudar a desmistificar alguns conceitos da Web 3.0, as vantagens e desvantagens encontradas no comércio eletrónico e o efeito nos consumidores. Para a concretização deste estudo foi efetuada uma abordagem quantitativa com base na análise de um inquérito, realizado através de software SPSS, procedendo-se à análise dos resultados obtidos nos questionários. Com este estudo pode afirmar-se que foram atingidos os objetivos propostos, foi explorada a perceção que o utilizador tem do comércio eletrónico e observou-se que o consumidor gosta de fazer uma pré-análise quando pretende adquirir online, comparando preços e optando pelo mais preço mais favorável e que lhe trará mais benefício, para além disso trata-se de uma alternativa mais cómoda.<br>As web technologies are increasingly present in our lives, the way we deal with them evolves. E-business has developed in a way that products and services reach more people in a larger geographic space at the lowest cost. Companies adapt to new reality by being more interactive and have more versatile buying options. The main objective of this dissertation is to study e-business and to help demystify some concepts of Web 3.0, as advantages and disadvantages found in electronic retail and the effect that has on consumers. To achieve the objective of this study we made a quantitative methodology based on the data analysis, through SPSS software, then proceeded to the analysis of the results obtained in the questionnaires. We can say that we have achieved the proposed objectives, we have explored the perception that the user have on electronic retail and we observed that the consumer likes to make some pre-order analyse by comparing the price and by choosing the less expensive and more profit product, becoming a more convenient alternative.<br>info:eu-repo/semantics/publishedVersion

Background: E-commerce is widespread in today’s shopping routines and conversational commerce (CC) as an expansion, aims at integrating customers and businesses on a whole new level. Through the application of chatbots fueled by artificial intelligence, a more personal and individual way of remote shopping is offered. Purpose: Our research question What potential attributes of AI-fueled CC applications and their possible inherent characteristics are determining the willingness to use them and to what extent, in the context of digital natives living in Sweden and Germany? aims at identifying the demanded attributes of conversational commerce from a consumer perspective. Method: We facilitate a quantitative questionnaire with 118 valid answers to administer a traditional full-profile conjoint analysis. Conclusion: Our results indicate that German digital natives deem a CC application’s behavior as the most important attribute, followed by payment method, personality and communication form (voice or text). The Swedish digital natives however, attach the most importance to the payment method, followed by behavior, communication form and personality. Both have in common that they prefer a rather passive behavior over being actively approached, a personality that is balanced between humor and seriousness and text-based communication over voice. A difference is the Swedish preference for direct in-app payment while German digital natives would select a redirection to a secondary payment provider (e.g. PayPal).

The purpose of this thesis is to explore the drivers and barriers of adoption of technology such as AI and how it could facilitate an internationalization. The thesis also hopes to explore how Born Global e-commerce companies are working towards an adoption of AI into the companies CRM-system. The model of multiple case study was chosen together with a qualitative method and an abductive approach. This was done by collecting primary data from three different Born Global firms within the e-commerce sector, as well as one CRM expert company, all located in Sweden and active on the international market. Furthermore, the theoretical framework presented Born Global, Technology Adoption, Customer Relationship Management and Artificial Intelligence. With the theoretical framework in mind, the conceptual framework was developed to show how the theories are connected to each other. Thereafter, the empirical findings were first presented and analysed together with theory and the voice of the authors of the thesis. The analysis visualize both similarities and dissimilarities between the empirical findings and the theory presented in the thesis. The final chapter concludes the barriers and drivers that Born Global e-commerce companies faces when adopting AI into their CRM-system. It also explains how AI within CRM could be beneficial as a tool on the Global market, rather than in the actual process of internationalization. To conclude the authors present the fact that companies are not working towards an adoption of AI into their CRM-systems. However, further development within the field is presented as well as theoretical and practical implications.

A remarkable gap exists between the financial impact of fraud in the B2C e-commerce mail order business and the amount of research conducted in this area — whether it be qualitative or quantitative research about fraud prevention. Projecting published fraud rates of only approx. one percent to e-commerce sales data, the affected sales volume amounts to $651 million in the German market, and in the North American market, the volume amounts to $5.22 billion; empirical data, however, indicate even higher fraud rates. Low profit margins amplify the financial damage caused by fraudulent activities. Hence, companies show increasing concern for raising numbers of internet fraud. The problem motivates companies to invest into data analytics and, as a more sophisticated approach, into automated machine learning systems in order to inspect and evaluate the high volume of transactions in which potential fraud cases can be buried. In other areas that face fraud (e.g. automobile insurance), machine learning has been applied successfully. However, there is little evidence yet about which variables may act as fraud risk indicators and how to design such systems in the e-commerce mail order business. In this research, mixed methods are applied in order to investigate the question how computer-aided systems can help detect and prevent fraudulent transactions. In the qualitative part, experts from fraud prevention companies are interviewed in order to understand how fraud prevention has been conventionally conducted in the e-commerce mail order business. The quantitative part, for which a dataset containing transactions from one of the largest e-commerce firms in Europe has been analyzed, consists of three analytical components: First, feature importance is evaluated by computing information gain and training a decision tree in order to find out which features are relevant fraud indicators. Second, a prediction model is built using logistic regression and gradient boosted trees. The prediction model allows to estimate the fraud risk of future transactions. Third, because risk estimation alone does not equal profit maximization, utility theory is woven into prioritization of transactions such that the model optimizes the financial value of fraud prevention activities. Results indicate that the interviewed companies want to use intelligent computer-aided systems that support manual inspection activities through the use of data mining techniques. Feature analysis reveals that some features, such as whether a shipment has been sent to a parcel shop, can help separate fraudulent from legitimate orders better than others. The predictive model yields promising results as it is able to correctly identify approximately 86% of the 2% most suspicious transactions as fraud. When the model is used to optimize the financial outcome instead of pure classification quality, results suggest that the company providing the dataset could achieve substantial additional savings of up to 87% through introduction of expected utility as a ranking measure when being constrained by limited inspection resources.

Es planteja en aquesta tesi l'efecte produït sobre la responsabilitat derivada de les accions realitzades autònomament per sistemes dotats d'intel·ligència artificial, sense la participació directa de cap ésser humà, en els temes més directament relacionats amb el comerç electrònic. Per a això s'analitzen les activitats realitzades per algunes de les principals empreses internacionals de comerç electrònic, com el grup nord-americà eBay o el grup xinès Alibaba. Després de desenvolupar els principals conceptes implicats en el tema en estudi, com els d’intel·ligència artificial, la responsabilitat, el comerç electrònic, Internet i la Web 3.0 o la influència de les noves tecnologies en les branques tradicionals del dret, es presenten els resultats de l'estudi. La metodologia utilitzada es basa en l'anàlisi de la diferent normativa i la jurisprudència existent sobre l'atribució de responsabilitat en les matèries relacionades amb les activitats del comerç electrònic. Des d'aquesta perspectiva se centra l'estudi en matèries com: la gestió de dades personals; la generalització de tècniques de Big Data; la robòtica aplicada a tasques de logística; la mobilitat mitjançant vehicles de funcionament plenament autònom; el pagament mitjançant monedes virtuals generades per algorismes; l'ús automatitzat de cookies, hash, links i altres dispositius similars a Internet; el màrqueting i la publicitat invasiva; la computació en núvol o la web profunda. De tot això es desprèn que la responsabilitat només és atribuïble legal i jurídicament als éssers humans. D'altra banda, s'estén l'estudi als drets que més es poden veure afectats per la generalització de metodologies d'intel·ligència artificial en el comerç electrònic, com poden ser els de privacitat, l'anonimat, la propietat intel·lectual, els drets d'autor i el copyright i altres similars. En tots els casos estudiats la presència de sistemes amb funcionament autònom no altera de cap manera, ni en la legislació ni en la jurisprudència, el fet que a les màquines no se'ls pot atribuir cap responsabilitat, ja que aquesta només és atribuïble de manera exclusiva als éssers humans. Es conclou que probablement les màquines arribin aviat a tenir més memòria, capacitat de càlcul, eficiència en la presa de decisions, precisió en la manipulació d'objectes, o capacitat d'aprendre que les persones. Però la responsabilitat continuarà sent de les persones sense que la puguin delegar en sistemes intel·ligents. Això serà així mentre les màquines no tinguin consciència del jo, autopercepció i capacitat de valorar èticament les conseqüències de les seves accions, doncs això és el que diferencia un ésser humà d'una màquina.<br>Se plantea en esta tesis el efecto producido sobre la responsabilidad derivada de las acciones realizadas autónomamente por sistemas dotados de inteligencia artificial, sin la participación directa de ningún ser humano, en los temas más directamente relacionados con el comercio electrónico. Para ello se analizan las actividades realizadas por algunas de las principales empresas internacionales de comercio electrónico, como el grupo norteamericano eBay o el grupo chino Alibaba. Después de desarrollar los principales conceptos implicados en el tema en estudio, como los de inteligencia artificial, la responsabilidad, el comercio electrónico, Internet y la Web 3.0 o la influencia de las nuevas tecnologías en las ramas tradicionales del derecho, se presentan los resultados del estudio. La metodología utilizada se basa en el análisis de la diferente normativa y la jurisprudencia existente sobre la atribución de responsabilidad en las materias relacionadas con las actividades del comercio electrónico. Desde esta perspectiva se centra el estudio en materias como: la gestión de datos personales; la generalización de técnicas de Big Data; la robótica aplicada a tareas de logística; la movilidad mediante vehículos de funcionamiento plenamente autónomo; el pago mediante monedas virtuales generadas por algoritmos; el uso automatizado de cookies, hash, links y otros dispositivos similares en Internet; el marketing y la publicidad invasiva; la computación en nube o la web profunda. De todo ello se desprende que la responsabilidad solo es atribuible legal y jurídicamente a los seres humanos. Por otro lado, se extiende el estudio a los derechos que más pueden verse afectados por la generalización de metodologías de inteligencia artificial en el comercio electrónico, como pueden ser los de privacidad, el anonimato, la propiedad intelectual, los derechos de autor y el copyright y otros similares. En todos los casos estudiados la presencia de sistemas con funcionamiento autónomo no altera de ninguna manera, ni en la legislación ni en la jurisprudencia, el hecho de que a las máquinas no se les puede atribuir responsabilidad alguna, pues esta solo es atribuible de manera exclusiva a los seres humanos. Se concluye que probablemente las máquinas lleguen pronto a tener mayor memoria, capacidad de cálculo, eficiencia en la toma de decisiones, precisión en la manipulación de objetos o capacidad de aprender que las personas. Pero la responsabilidad seguirá siendo de estas sin que la puedan delegar en sistemas inteligentes. Esto será así mientras las máquinas no posean conciencia del yo, autopercepción y capacidad de valorar éticamente las consecuencias de sus acciones, pues esto es lo que diferencia a un ser humano de una máquina.<br>This thesis considers the effect on responsibility produced by actions taken autonomously by systems equipped with artificial intelligence, without the direct involvement of human beings, for issues directly related to e-commerce. To this end the activities of some of the major international e-commerce companies such as the American group eBay and the Chinese group Alibaba are analyzed. After developing the main concepts involved in the subject under study, such as artificial intelligence, responsibility, e-commerce, Internet and Web 3.0 and the influence of new technologies on traditional areas of law, the results of the study are presented. The methodology is based on the analysis of different regulations and existing case law on the allocation of responsibility for matters relating to e-commerce activities. From this perspective, the study focuses on subjects such as: management of personal data; generalization of Big Data techniques; robotics applied to logistics tasks; mobility via fully operationally autonomous vehicles; payment with virtual currencies generated by algorithms; the automated use of cookies, hash, links and any other similar devices on the Internet; invasive marketing and advertising; cloud computing or Deep Web. From this investigation it follows that responsibility is only legally attributable to human beings. The study also extends to examine the rights that may be more affected by the spread of artificial intelligence methodologies in e-commerce, such as rights of privacy, anonymity, intellectual property, copyright and other similar rights. In all cases studied the presence of systems with autonomous operation does not alter in any way, either in legislation or in case law, the fact that responsibility cannot be attributed to machines, as it is exclusively attributable to humans. We conclude that while machines will probably soon have more memory, computing capability, efficiency in decision-making, precision in object manipulation and ability to learn than people, responsibility cannot be delegated to intelligent systems. This will remain the case until machines have self-awareness, self-perception and the ability to ethically assess the consequences of their actions, as this is what differentiates a human being from a machine.

Background: One of the most growing areas of the Internet is E-commerce, which gave consumers a flexibility that was not previously available. The advantages of digital commerce are many, where both companies and consumers can use the technology's possibilities. However, a problem area in the E-commerce area is the number of products and the consumers' difficulties finding what is demanded. One of the techniques used to solve the problem is Artificial Intelligence. Within the artificial intelligence there is the sub-area Machine learning. The basis of the technology is to teach a computer to perform specific tasks better and better over time. To implement machine learning, artificial neural network used, which this study is based on. Problem statements: In this research, machine learning will be tested to see if the technology can be used to increase the conversion rate in E-commerce. The study will also be supplemented with knowledge about the opportunities and challenges that a future implementation could bring. The questions to be answered are: - How can product searches based on machine learning be used to increase the conversion rate in E-commerce? - What are the most common opportunities and challenges associated with machine learning in E-commerce? Purpose: The study's purpose is to make it easier for consumers when it comes to finding the right products and to increase the knowledge base of machine learning in E- commerce. Results: The study concluded that search engines based on Machine Learning can be used to increase the conversion rate. The opportunities that were found were improved business strategies, more personalized approaches and increased revenue. The challenges that were discovered were resource shortages, data quality and lack of project management.<br>Bakgrund: Ett av de mest växande områdena på internet är E-handel, som gett konsumenter en flexibilitet som inte tidigare fanns. Fördelarna med den digitala handeln är många, där både företag och konsumenter kan nyttja teknikens möjligheter. Ett problemområde inom E-handeln är dock mängden produkter och konsumenters svårigheter att finna det som efterfrågas. En av de tekniker som används för att lösa problematiken är Artificiell Intelligens. Ett delområde inom den artificiella intelligensen är maskininlärning. Grunden bakom tekniken handlar om att lära upp en dator för att utföra specifika uppgifter bättre och bättre över tid. En metod som används för att implementera maskininlärning är artificiella neurala nätverk som används i denna studie. Frågeställning: I denna forskning kommer maskininlärning att testas för att se om tekniken kan användas för att öka konverteringsgraden inom E-handel. Studien skall även kompletteras med kunskap kring de möjligheter och utmaningar som medföljer en framtida implementation. De frågor som därmed skall besvaras är: - Hur kan produktsök som baseras på maskininlärning användas för att öka konverterings- graden inom E-handel? - Vilka är de vanligast förekommande möjligheterna och utmaningarna med maskininlärning inom E-handel?’ Syfte: Studiens syfte är att underlätta för konsumenter när det kommer till att hitta rätt produkter samt för att öka kunskapsbanken gällande maskininlärning inom E-handeln. Resultat: Studien resulterade i att sökmotorer som baseras på maskininlärning kan användas för att öka konverteringsgraden. Studien pekade även på fördelar gällande förbättrade affärsstrategier, personligare bemötande och ökade intäkter. De utmaningar som upptäcktes var resursbrist, datakvalité samt bristande projektledning.

The master thesis deals about proposal, implementation and creating chatbot logic for e-commerce. Chatbot will serve for communication with customers mostly, but also as an online marketing channel. Thesis includes theoretical part, analysis of present conditions and requirments and also practical part which is separated in more specific subparts. In theoretical part we can find necessary informations to understand the chatbot problematic. Analysis of present conditions and requirments is focused to already working e-commerce system, which is also used as source of informations in practical part.

A inovação tecnológica e a democratização da inteligência artificial (IA) têm vindo a alavancar o potencial de sucesso em todas as áreas que conhecemos hoje, com expectativas do que ainda está para vir. A presente dissertação propõe uma análise das aplicações da IA na indústria da moda, particularmente nas plataformas de marcas de moda do comércio eletrónico, e de que forma está a ter impacto na esfera pessoal do consumidor, particularmente no processo de tomada de decisão dos consumidores da Geração Z. O âmbito da IA tem vindo a evoluir de tal forma que permitiu às empresas não só melhorar a sua oferta e a procura dos clientes, como também proporcionar uma experiência de compra que vai para além da “seleção e compra” mecânica: os pontos de contacto impulsionados pela IA influenciam e enriquecem cada fase do processo de tomada de decisão, seja de forma mais positiva ou negativa. Em última análise, esta dissertação pretende proporcionar ao leitor um melhor conhecimento sobre a IA e o comércio eletrónico de moda, bem como delinear o seu impacto no comportamento online do consumidor.<br>Technological innovation and democratization of artificial intelligence (AI) have been leveraging the potential success in every field we know today, while more is yet to come. The following dissertation proposes an analysis of AI achievements within the fashion industry, particularly in e-commerce fashion brand platforms, and how it is impacting the consumer personal sphere, particularly the decision-making process of Gen-Z consumers. The field of AI has been evolving in such a way that allows companies to not only improve their supply and customer demand, but also provide a shopping experience that goes beyond the mechanical “select and buy“: AI-driven touchpoints influence and enrich each stage of the decision-making process, whether more positively or negatively. Ultimately, this dissertation intends to provide the reader a better knowledge of AI and fashion e-commerce joining applications, and to delineate its impact on the online customer journey.

Nowadays, every company struggles to success, however there are few of them that actually do. Artificial intelligence is something that is revolutionize the market, as well as, the way that companies structure themselves. This work pretends to illuminate who is searching about the effects that the artificial intelligence can have on customer loyalty and on entrepreneurship. To do that, I created an online survey, answered by 103 people and I made eleven interviews to people that worked or are working in one of these three areas. With these, I pretend to have a better notion about the impact that AI is having or can have in a near future in these areas, the benefits and disadvantages that can create for the companies, or even the utility that can be for a worker launch his own business. After conducting the questionnaire and the interviews and comparing it with the literature review, I came to some conclusions right now that in a very near future the use of AI will be fundamental to the success of any company. You will inevitably have to use AI tools to optimize processes, save costs, but never mischaracterize the company. On the other hand, customer satisfaction is extremely important because it leads to a buyback from the clientes and the good publicity they will do with third parties, however this is only achieved when there is a good mix between the quality of the salesman and the product.<br>Hoje em dia, todas as empresas lutam para serem bem-sucedidas, no entanto, poucas delas realmente o são. A inteligência artificial é algo que está a revolucionar o mercado, bem como a forma como as empresas se estruturam e por isso pode ter um papel fundamental. Com este trabalho pretendo iluminar quem está à procura de respostas sobre os efeitos que a inteligência artificial pode ter na fidelidade do cliente e no empreendedorismo. Para isso, realizei um inquérito "online", respondido por 103 pessoas e fiz 11 entrevistas a pessoas que trabalharam ou estão a trabalhar numa destas três áreas. Com isso, pretendo ter uma melhor noção sobre o impacto que a IA está a ter ou pode vir a ter num futuro próximo nestas áreas, os benefícios e desvantagens que podem criar para as empresas, ou mesmo a utilidade que pode ter para o lançamento de um negócio próprio. Após realizar o questionário e as entrevistas e comparando com a revisão de literatura cheguei a algumas conclusões desde já o facto de num futuro bastante próximo a IA ser fundamental para o sucesso de qualquer empresa. Terá inevitavelmente de usar ferramentas de AI para otimizar processos, poupar nos custos, mas nunca descaracterizando a empresa. Por outro lado, a satisfação do cliente é extremamente importante pois leva a uma recompra por parte dos mesmos e à boa publicidade que irão fazer junto de terceiros, porém isto só é conseguido quando existe um bom mix entre a qualidade do vendedor e do produto.

Artificial Intelligence (AI) is intensifying its importance, being increasingly present in people's everyday interactions and activities, while shaping a variety of industries, among which is the field of online shopping, that is currently experiencing significant growth in terms of volume and acceptance among customers. A range of factors have contributed to the change in how shoppers behave, including the ease of access and a greater variety of products and services available online. Nonetheless, retail sectors still face some difficulties in adopting e-commerce, particularly related to the issue of converting the in-store experience to the online environment. To answer these constraints, several AI-based tools, namely Chatbots, Recommendation Agents, and Virtual Try On Systems (VTOs) have been introduced in the e-commerce field in order to enhance the online experience. Ultimately these technologies promise to provide new possibilities for the field of online shopping, to increase the digital interaction between customers and brands, and to influence customers purchase behaviour. As so, this dissertation aims to assess why AI should become increasingly relevant for companies in order to better understand and meet consumers' needs through personalised experiences, offering a possibility to gain competitive advantages. Hence, the impact of these AI-based tools on the online experience of the customer and subsequent purchase intention was examined using a mixed approach of quantitative methods, considering both a corporate and a consumer perspective. The results highlight that AI is seen as a promising tool among both Portuguese brands and customers using online retail platforms. In particular, it was possible to conclude that the use of AI tools enables a better customer experience when shopping online, ultimately leading customers to complete their purchases.<br>A Inteligência Artificial tem cada vez maior importância na sociedade atual, não só estando progressivamente mais presente nas interações e atividades diárias das pessoas, como também contribuindo crescentemente para a transformação de diversas indústrias, entre as quais o e-commerce. Simultaneamente, o campo do comércio online está a experienciar um crescimento significativo em termos de volume e aceitação entre os clientes. A facilidade de acesso e maior variedade de produtos e serviços disponíveis online são fatores que têm contribuído para a mudança no comportamento dos compradores. No entanto, os sectores retalhistas ainda enfrentam algumas dificuldades na adoção do comércio eletrónico, particularmente relacionadas com a questão da conversão da experiência na loja para o ambiente online. Para responder a estas questões, várias ferramentas baseadas na IA, nomeadamente Chatbots, Agentes de Recomendação e Sistemas de Virtual Try-On (VTOs), foram introduzidas no âmbito do comércio eletrónico a fim de melhorar a experiência online, proporcionar novas possibilidades de interação entre consumidores e marcas, e, em última análise, influenciar a intenção de compra dos clientes. Esta dissertação visa avaliar por que razão a IA deve tornar-se cada vez mais relevante para que as empresas compreendam e satisfaçam melhor as necessidades dos consumidores, através de experiências personalizadas, oferecendo uma possibilidade de obter vantagens competitivas. Assim, o impacto destes Sistemas Inteligentes sobre a experiência online do cliente e subsequente intenção de compra foi examinado, utilizando uma abordagem mista de métodos quantitativos, considerando tanto uma perspetiva empresarial como uma perspetiva do consumidor. Os resultados salientam que a IA é vista como uma ferramenta promissora tanto entre as marcas portuguesas como entre os clientes que utilizam plataformas de retalho online. Em particular, foi possível concluir que o uso de ferramentas de IA permite uma melhor experiência do cliente quando faz compras online, levando em última análise os clientes a completar as suas compras.

This thesis relates to the usage of trust modelling in multi-agent systems - environments in which there are interacting software agents representing various users (for example, buyers and sellers exchanging products and services in an electronic marketplace). In such applications, trust modelling may be crucial to allow one group of agents (in the e-commerce scenario, buyers) to make effective decisions about which other agents (i.e., sellers) are the most appropriate partners. A number of existing multi-agent trust models have been proposed in the literature to help buyers accurately select the most trustworthy sellers. Our contribution is to propose several modifications that can be applied to existing probabilistic multi-agent trust models. First, we examine how the accuracy of the model can be improved by limiting the network to a portion of the population consisting of the most trustworthy agents, such that the less trustworthy contributions of the remaining agents can be ignored. In particular, we explore how this can be accomplished by either setting a maximum size for a buyer's advisor network or setting a minimum trustworthiness threshold for agents to be accepted into that advisor network, and develop methods for appropriately selecting the values to limit the network size. We demonstrate that for two models, both the Personalized Trust Model (PTM) developed by Zhang as well as TRAVOS, these approaches will yield significant improvements to the accuracy of the trust model, as opposed to using an unrestricted advisor network. Our final proposed modification is to use an advisor referral system in combination with one of the network-limiting approaches. This would ensure that if a particular agent within the advisor network had not met a specified level of experience with the seller under consideration, it could be replaced by another agent that had greater experience with that seller, which should in turn allow for a more accurate modelling of the seller's trustworthiness. We present a particular approach for replacing advisors, and show that this will yield additional improvements in trust-modelling accuracy with both PTM and TRAVOS, especially if the limiting step were such that it would yield a very small advisor network. We believe that these techniques will be very useful for trust researchers seeking to improve the accuracy of their own trust models, and to that end we explain how other researchers could apply these modifications themselves, in order to identify the optimal parameters for their usage. We discuss as well the value of our proposals for identifying an "optimal" size for a social network, and the use of referral systems, for researchers in other areas of artificial intelligence.

Limitadas por recursos escassos, as pequenas e médias empresas (PMEs) enfrentam uma maior pressão para garantir uma posição competitiva no mercado global, marcado pelo galopante avanço tecnológico e pelo contexto pandémico. Como tal, o desenvolvimento de plataformas de comércio eletrónico com recurso à artificial intelligence (AI) em contextos ágeis tem sido cada vez maior. Por um lado, a AI permite um maior tratamento de dados e a criação de modelos de projeção. Por outro lado, a adoção de métodos ágeis assegura uma maior gestão da imprevisibilidade e adaptação às novas exigências de mercado. Verificando-se um reduzido número de estudos que abrangem as três áreas acima mencionadas, propõe-se a criação de um novo modelo que identifique os determinantes que favorecem ou inibem o desenvolvimento do comércio eletrónico, da AI e dos métodos ágeis em PMEs. Este será suportado pela aplicação de técnicas de mapeamento cognitivo, através da abordagem Strategic Options Development and Analysis (SODA), juntamente com a técnica DEcision MAking Trial and Evaluation Laboratory (DEMATEL), procurando extrair um conjunto de recomendações que permita às PMEs adquirir uma posição competitiva no mercado global.<br>Constrained by scarce resources, SMEs face a great deal of pressure to maintain a competitive position in the global market, especially during the Covid-19 pandemic. As such, e-commerce platforms using Artificial Intelligence (AI) with agile methodology have been thriving. On the one hand, AI allows for better data processing and the creation of forecasting models. On the other hand, agile adoption ensures a management improvement over uncertainty and a better adaptation to new market requirements. Considering the limited number of studies that cover these three knowledge fields, we propose the creation of a new model to identify determinants that favor or inhibit the development of e-commerce, AI and agile methods in SMEs. This model will be developed using cognitive mapping techniques, based on the Strategic Options Development and Analysis (SODA) approach, and the DEcision MAking Trial and Evaluation Laboratory (DEMATEL) technique. The aim is to retrieve a set of recommendations that will allow SMEs to acquire competitive positions in the global market.

O início da pandemia COVID-19 teve um impacto significativo na economia mundial e pode vir a transformar o paradigma atual. As regras impostas aos indivíduos como as restrições de movimentos, distanciamento social e o confinamento obrigatório imposto, abalaram significativamente o tecido empresarial, uma vez que as empresas foram forçadas a parar por tempo indeterminado. O tecido empresarial depara-se com novos desafios sobretudo na maneira como os negócios são realizados. Este terá de reagir por forma a sobreviver e permanecer competitivo no ambiente económico atual. O e-commerce disparou, sendo que este se tornou a única maneira de adquirir produtos e a IA começa a ser ponderada, já que o distanciamento social e a redução do contacto se tornaram imperativos. O Objetivo desta investigação passa precisamente por averiguar, se a pandemia COVID-19 correspondeu a uma oportunidade para a utilização de IA pelas empresas, visto que possibilita a automatização dos processos, permitindo a redução do contacto. Para tal, foi feita uma pesquisa incorporando a esfera dos consumidores e das empresas, no sentido de apurar se existiu um novo posicionamento dos consumidores em relação ao uso de IA, bem como a perceção das empresas relativamente a estes temas, assim como a sua possibilidade de adoção de sistemas inteligentes. Foi possível concluir que houve uma alteração na recetividade dos consumidores e na perceção das empresas relativamente ao uso de IA com a pandemia COVID-19, e que a possibilidade de utilização de IA pelas empresas é impactada pela sua perceção e pelos seus benefícios.<br>The onset of the COVID-19 pandemic has had a significant impact on the world economy and may well transform the current paradigm. The rules imposed on individuals such as movement restrictions, social distancing, and the mandatory confinement imposed, have significantly shaken the business fabric as companies have been forced to shut down indefinitely. Business is facing new challenges, especially in the way business is conducted. It will have to react to survive and remain competitive in the current economic environment. E-commerce has skyrocketed to become the only way to purchase products and AI is starting to be considered, as social distance and contact reduction have become imperative. The objective of this research is precisely to find out if the COVID-19 pandemic was an opportunity for companies to use AI, since it enables the automation of processes, allowing contact reduction. To this end, a survey was conducted incorporating the sphere of consumers and companies, to ascertain whether there was a new positioning of consumers concerning the use of AI, as well as the perception of companies regarding these issues, and their possibility of adopting intelligent systems. It was possible to conclude that there was a change in consumer receptivity and the perception of companies regarding the use of AI with the COVID-19 pandemic and that the possibility of AI use by companies is impacted by their perception and benefits.

In recent years, online shopping has gained immense popularity due to its feedback mechanism. By composing online comments, previous buyers share opinions and expe-riences regarding the items that they have purchased. These user-generated reviews, in turn, provide valuable information to potential customers in regards to deciding which products to purchase. The reviews also help vendors understand customer needs and improve product quality. Yet despite these benefits, the unprecedentedly rapid growth of user-generated content has overwhelmed human ability in online review scrutiny. On-line reviews that possess varying content further impedes useful knowledge distillation. The large volume of online reviews that are uneven in quality puts growing pressure on automatic approaches for effective review utilization and informative content prioritiza-tion. Review helpfulness prediction leverages machine learning methods to identify and recommend helpful reviews to customers. In particular, review characteristics form the backbone of helpfulness information acquisition. Prior literature has observed and as-sociated a large body of determinants with review helpfulness. However, these deter-minants heavily rely on the domain knowledge of experts. The selection of and the interaction between the determinants also remain understudied, leaving ample room for exploration. The general lack of systematic experiment protocols among the existing methods further harms the task’s reproducibility, comparability, and generalizability. This thesis aims to automatically model helpfulness information from online user- generated reviews. The thesis proposes effective modeling techniques and novel so-lutions to tackle the aforementioned challenges, with more emphasis on sophisticated feature learning and interaction. The thesis has made the following contributions to standardize the research field and advance the accuracy in helpfulness prediction. 1. A comprehensive survey is conducted to identify frequently used content-based determinants for automatic helpfulness prediction. A computational framework is developed to empirically evaluate the identified features across domains. Three selection scenarios are considered for feature behavior analysis. The domain-specific and domain-independent feature selection guidelines are summarized to facilitate future research prototyping. The implementation details of the study are discussed to standardize the task of automatic helpfulness prediction. 2. A deep neural framework is designed to enrich the interaction between review texts and star ratings during automatic helpfulness prediction. A gated convolu-tional component is introduced to learns content representations. A gated em- bedding method is proposed for encoding sophisticated yet adaptive rating infor- mation. An element alignment mechanism is proposed to explicitly capture the text-rating interaction. Ablation studies and qualitative analysis are conducted to discover insights into the interactive behavior of star ratings. 3. An end-to-end neural architecture is proposed to contextualize automatic helpful- ness prediction using review neighbors. Four weighting schemes are designed to encode a review’s surrounding neighbors as its context information into content representation learning. Three types of reviews neighbors of varied length are considered during context construction. Finally, discussions on the experimental results and the trade-o between model complexity and performance are given, along with case studies, to understand the proposed architecture.

Throughout the times it has been witnessed a continuous evolution in the way people make business transactions, across sectors. This has been highly influenced by technological developments and the constant need for companies to adjust to their clients’ needs and expectations. The retail sector has been no exception, evolving alongside innovation, and adapting to new trends. One of its results is the emergence of conversational commerce, a new form of commerce that combines the trend of communicating via instant messages and the use of artificial intelligence, introducing virtual assistants to the retail context. The aim of this study is to better understand the potentialities virtual assistants have in the Portuguese retail context, amongst young adults. By identifying which of the dimensions of the interaction with a retailer virtual assistant - "Cognitive Perception, Affective Engagement, and Communication Quality" - have a significant impact on the expectation users create towards their customer experience and how this is determinant to their patronage intentions towards the retailer. A quantitative methodology was used to perform this investigation, with the development of an introductory chatbot and an online survey, completed by 385 individuals (Portuguese young adults, with ages ranging from 18 to 35, and that had access to the Facebook Messenger app). The Customer Experience Expectation was proven to have a significant impact on the respondent' patronage intentions towards the virtual assistant and the retailer. However, only the cognitive dimension of the virtual assistant was confirmed to significantly impact the expectations on the customer experience<br>Ao longo dos tempos, tem-se vindo a testemunhar uma evolução na forma como o comércio é feito nos vários setores de atividade. As empresas têm de se reinventar constantemente para satisfazer as necessidades e expectativas dos consumidores, que resultam dos avanços tecnológicos. O mesmo acontece no setor do Retalho, que tem vindo a inovar, acompanhando a tecnologia e as tendências dos consumidores. Um exemplo disto é o aparecimento da uma nova forma de comércio, o comércio conversacional. Este combina a tendência de comunicação via mensagens instantâneas com o desenvolvimento da inteligência artificial, introduzindo assistentes virtuais neste setor. O principal objetivo deste estudo prende-se com a investigação do impacto que a inclusão de um assistente virtual teria no setor do retalho interagindo com os jovens adultos portugueses. Para tal, procurou identificar-se quais as dimensões da interação com um assistente virtual - cognitiva, afetiva e comunicativa - que influenciariam as expectativas relativas à experiência de compra e consequentemente as intenções de uso e compra dos consumidores. Para a investigação foi utilizada uma metodologia quantitativa, com a criação de um "chatbot" informativo e de um questionário "online", ao qual responderam 385 portugueses com idades desde os 18 até aos 35 anos, com acesso ao Facebook Messenger. Neste estudo foi provado que as expectativas dos consumidores em relação à experiência de compra influenciam as suas intenções de uso (assistente virtual) e compra (retalhista). No entanto, apenas a dimensão cognitiva mostrou ter um impacto significativo na criação de expectativas relativas à experiência de compra.