Browsing by Issue Date, starting with "2023-02-03"
Now showing 1 - 10 of 18
Results Per Page
Sort Options
- Crime inference using machine learning and geographical dataPublication . Roque, Miguel Francisco Frade; Bertani, NicolòCrimes are not random events in society, and eventually something must influence their occurrence. It is by characterizing the environment that it is possible to create algorithms that predict the criminal activity in a certain place and at some point in time, which allows its anticipation and prevention through decision-making in public policy. This study focusses on finding the best way to predict crimes, that is, which types of features are the most important to consider while predicting crimes, and which methods are the most predictive. An analysis of the city of Philadelphia, in the state of Pennsylvania (USA), is made, taking into account the urban, racial, demographic and socioeconomic characteristics of its different geographical blocks, and the number of criminal occurrences in each of them, over multiple years. The methods used are both linear and non-linear. When non-linear methods are used, via machine learning techniques, it is evident that the prediction of the number of crimes is much more assertive for any type of variable, leading to the conclusion that the relationships studied here are not linear in nature, and therefore tree based models (especially gradient boosting and random forest) represent the most suitable approach for this data. In this perspective, the models that consider only the socio-demographic characteristics of the neighborhoods are significantly more effective in forecasting than the entirely urban ones.
- Do playstation5 product attributes drive console upgrading?Publication . Mostafa, Ibrahim Ahmed Ahmed Mohamed; Romeiro, PauloThe number of times players must wait for a game to update its data before playing can detract from their console gaming experience. This article focuses on the decisionmaking processes that can be inferred from the product qualities that people seek before upgrading their gaming consoles. Apart from that, the benefits of upgrading. As a result, a detailed interview was conducted before the focus groups. To put the theories to the test, an online crosssectional survey was used. The results revealed that physical product attributes such as image quality, processing speed, and product compatibility had a significant impact on the game's update to the PS5. When purchasing new game consoles, gamers place a high value on a variety of product characteristics such as user interface and gameplay experience. Furthermore, gamers prefer a combination of tangible and intangible attributes in their gaming console because it is difficult to determine which attribute is the most desired
- Green brand image in digital retail : a research on green labeling impact on willingness to payPublication . Leggen, Lukas Nils; Romeiro, PauloThe environmental consciousness of customers continues to rise, while digital retail provides them with additional purchase options. This dissertation explores the impact of a sustainable brand image on consumers’ willingness to pay in an online retail context, focusing on the significance of green labels. For the empirical investigation, four distinct stimuli were developed based on the results of an online pre-survey. The stimuli comprise of a T-Shirt with and without a green label and applied to two brands, of which one has been found to be regarded as sustainable, while the other has been found to not be perceived as sustainable. These stimuli were applied to an online questionnaire and presented randomly to 277 respondents. Brand image concerning sustainability and the willingness to pay were assessed, and the determinants of the Theory of planned behavior. This research found that a sustainable brand image positively affects consumers’ willingness to pay. Though the relationship between brand image and willingness was not found to be moderated by the existence of green labeling, green labeling has been found to result in a (significant) higher willingness to pay. The determinants of the Theory of planned behavior participated in the model as partial mediators. This thesis contributes to close the literature gap concerning green labeling and its impact on the relationship between brand image and willingness to pay. From a managerial point of view, this study helps to give implications better to understand the relationships of the different variables under study.
- Relação entre a utilização do instagram e a satisfação corporal nos homensPublication . Figueiredo, Maria Eduarda Correia; Azevedo, Ângela Maria Pereira e SáO principal objetivo da presente investigação consiste em explorar a relação entre a utilização do instagram e a satisfação corporal nos homens, numa amostra composta por 205 sujeitos do sexo masculino, com idades compreendidas entre os 18 e os 35 anos de idade, dada a escassez de estudos no contexto português. Para tal, foram aplicados 3 instrumentos de avaliação: Questionário Sociodemográfico, Questionário de Adição ao Instagram e Escala de Avaliação da Satisfação com a Imagem Corporal. Os resultados demonstram existir uma correlação estatisticamente significativa entre a utilização do instagram e a satisfação com a aparência.
- The role of celebrity endorsement and its effect on purchase intentionPublication . Marques, Ana Carolina de Almeida; Romeiro, PauloThrough the years, it has become clear that celebrity endorsement is an essential part of any effective marketing plan for a company. Subsequently, one of the most crucial techniques to reach the target audience and differentiate the business from the competition is certainly creating successful communication strategies. As a result, the main goal of this dissertation is to understand how the use of celebrities in advertisements—or the lack thereof—can affect consumers' intentions to buy a brand's product. In this way, celebrity endorsement will be treated as an independent variable, while purchase intention will be handled as a dependent variable. Additionally, brand equity will be included in the conceptual framework as a mediator between the two previously mentioned variables. Later, through a survey in which participants faced one of six stimuli, quantitative data was gathered and, posteriorly, analyzed though IBM SPSS. This way, respondents were prompted to respond to some questions concerning celebrity selection, rate twentythree brand equity statements, and indicate on a fivepoint Likert scale whether they intended to purchase the brand that was being promoted. Contrary to what was concluded by the literature, using celebrities in advertisements does not appear to boost brand equity or purchase intention. This was confirmed by the completion of independent samples TTests, which demonstrated that there was no significant relationship between the variables. Additionally, it doesn't seem that brand equity significantly mediates celebrity endorsement and purchase intention.
- Sustainable savings accounts : opportunities, risks and relevant attributesPublication . Koller, Sandro; Moreira, Sérgio PauloThe climate crisis has emphasized the need for more sustainable practices in all sectors of society, including the financial industry. While there has been a significant amount of research on sustainable investing and funds, there has been relatively little attention given to the potential for sustainable savings to contribute to the transition to a more sustainable future. This master's thesis aims to address this gap by conducting a qualitative study through ten expert interviews with finance professionals from Swiss retail banks. The research focuses on exploring the opportunities, risks, and attributes of sustainable savings accounts. The results of the study show that the main opportunities for banks in offering sustainable savings accounts are differentiation from competitors, higher margins, and simplicity of product setup. The main opportunities for consumers are the ability to align their savings with their values and make a positive impact on the environment and society, with minimal effort required on their part. The study also identified risks for both banks and consumers, including competitive disadvantage for banks, increased effort and potential reputation risk, and financial disadvantage and the possibility of banks misusing savings for consumers. Key attributes of sustainable savings accounts were identified and include interest rates, account maintenance fees, and transparency. These attributes should be used in further research to understand consumer preferences and to maximize the potential of sustainable savings accounts to fight the climate crisis in the long run.
- Status and conspicuousness on sustainable fashion products : examining the role of conspicuousness between status motives and consumers’ sustainable fashion valuationsPublication . Martins, Carolina Santos; Colaço, Vera HerédiaThe fashion industry is believed to be the second-largest polluting industry worldwide. Hence, understanding the underlying reasons motivating socially and environmentally responsible consumption behaviors is crucial to implement the necessary changes in this industry. The present dissertation aims to examine the effect of social motivations, such as activating status motives, on consumers’ attitudes and perceptions towards sustainable fashion products. Based on an academic literature review regarding sustainable fashion, conspicuousness and status consumption, an experimental design study was conducted. The purpose was to examine the impact of status and prestige motivations on consumers’ sustainable fashion valuations, such as purchasing behaviors, product, and brand perceptions. Results show that activating status motives enhance status and prestige motivations. In turn, conspicuously signaling a sustainable fashion product moderates the effect that status and prestige motivations have on consumers’ valuations. Findings show that a desire for status is capable of inciting self-sacrifice and consequently motivating sustainable consumption decisions. Additionally, brand iconicity perceptions indirectly explain the process through which conspicuous consumption orientation impacts sustainable fashion valuations. Therefore, consumers tend to use the symbolic value of iconic brands to express their social identity and communicate their desires and aspirations with others. This study provides valuable consumer behavior insights, suggesting brands’ approach to sustainability should not disregard society’s predisposition for consumption. Forthwith, researchers, managers, and marketers trying to incite sustainable consumption might obtain more effective and long-lasting results if they factor in individual traits and predispositions since these are able to influence consumption behavioral patterns within the fashion context.
- Predicting hourly demand for shared bicycles with weather data and machine learning modelsPublication . Peja, Dijora; Bertani, NicolòThis thesis aims to analyze the bike sharing system in Chicago and apply predictive models that accurately predict the hourly demand for shared bicycles by using timerelated and weatherrelated features. The dependent variable is Count, expressing the sum of the number of bicycles used per hour. Predictive models that are used for this regression problem are Linear Regression, Random Forest, Gradient Boosting, Light Gradient Boosting, Extreme Gradient Boosting, and MultiLayer Perceptron. Accuracies of these predictive models are measured by R2_score, Root Mean Square Error and Mean Absolute Error. For better predictions, different hyperparameters are used in predictive models. Without hyperparameters, Random Forest achieves the best accuracy measures. However, after using hyperparameters, Gradient Boosting predicts the most accurate results. The accuracy of Gradient Boosting boosts with hyperparameters, whereas Random Forest is almost unaffected by them for this regression problem. The secondbest model when using hyperparameters is Extreme Gradient Boosting. The neural network model, MultiLayer Perceptron presents less accurate results than the Random Forest and the Boosting models for this type of problem. Features that are most important for predictive models to forecast accurately were Temperature, Hour, Weekend, Pressure, Uv_Index, and Day.
- Replenishment support decision model for a try-before-you-buy retail fashion ecommercePublication . Canessa, Andrea; Gijsbrechts, JorenA try before you buy business model is a type of sales strategy in which customers are allowed to test a product before making a purchase. As the try before you buy online business model is a topic on which there is limited public scholarly research, the purpose of this research is to provide an initial approach to the subject by presenting a tool to support the replenishment strategy of Curve Catch, a fashion ecommerce retailer. A simulation engine characterized by two main components has been built: replenishment and a demand generator. Model development is built on artificially generated data based on real data of Curve Catch. Based on the literature inherent to inventory management and through the use of simulation optimization, the model provides managerial guidance on how to manage r,Q policy in a system where most goods shipped to customers are returned. The tool highlights the need to optimize the use of reorder point and economic order quantity to achieve better business performance. This is because conventional formulas, if not adjusted to the specific setting, perform sub optimally. The model also provides insights into the levels of lost sales due to out stocking and the quality of service provided to customers. Studying the relationships among these three KPIs provides insight into what tradeoffs are relevant in planning a continuous review replenishment strategy.
- The impact of video streaming service’s brand image on consumer loyalty : does word-of-mouth make a difference?Publication . Costa, Ana Rita Abreu; Romeiro, PauloVideo streaming services have grown substantially in the past few years, due to their convenience, low cost and variety of original movies and tv shows. Consumers have switched from traditional broadcasting TV to streaming service platforms, which has led to the entrance of several players in the industry. The purpose of this research is to investigate how the video streaming service’s brand image, with a particular focus on Netflix, Disney Plus and Amazon Prime Video, impact customer loyalty, while also studying the moderation effect of wordofmouth. Furthermore, the mediation effect of expectations and perceived quality were also analyzed. In order to study the proposed model, six different stimuli were created based on online reviews. The stimuli presented either negative wordofmouth or positive wordofmouth, and were then used on an online survey, randomly assigned to each participant. Results show that the video streaming service’s brand image positively impacts customer loyalty. It was also concluded that both expectations and perceived quality media te the relationship between brand image and customer loyalty. Moreover, it was also proven that wordofmouth also impacts this relationship.