Ivo Pereira

Abstract
The growing interest in learning more about consumer behaviors through analytical techniques requires the integration of innovative approaches that relate their needs to strategic marketing procedures. Multimodality and Affective Computing combined a series of robust optimizations for this challenge, implying the complexity of each application. However, the entanglement of different modalities demands new and tailored refinements to enhance adaptability and accuracy in the field. This paper outlines the implementation of a Multimodal Artificial Intelligence methodology with Affective Computing to enhance consumer insights and marketing strategies. The application combines different data modalities, such as textual, visual, and audio inputs, to tackle complex issues in dealing with consumer sentiment. The proposed approach uses advanced preprocessing techniques, including word embeddings, neural networks, and recurrent models, to extract information from diverse modalities. Fusion strategies, such as attention-based and late fusion procedures, are utilized to combine knowledge, facilitating robust sentiment detection. The implementation includes the analysis of real-time customer feedback on social media and product assessments, demonstrating improvements in predicting engagement and shaping consumer behavior. The results underscore the practical viability of the suggested method, promoting progress in multimodal sentiment analysis to extract actionable consumer insights in marketing.

Abstract
This review discusses the integration of intelligent technologies into customer interactions in organizations and highlights the benefits of using artificial intelligence systems based on a multimodal approach. Multimodal learning in marketing is explored, focusing on understanding trends and preferences by analyzing behavior patterns expressed in different modalities. The study suggests that research in multimodality is scarce but reveals that it is as a promising field for overcoming decision-making complexity and developing innovative marketing strategies. The article introduces a methodology for accurately representing multimodal elements and discusses the theoretical foundations and practical impact of multimodal learning. It also examines the use of embeddings, fusion techniques, and explores model performance evaluation. The review acknowledges the limitations of current multimodal approaches in marketing and encourages more guidelines for future research. Overall, this work emphasizes the importance of integrating intelligent technology in marketing to personalize customer experiences and improve decision-making processes.

Abstract
Recommendation systems can play an important role in today's digital content platforms by supporting the suggestion of relevant content in a personalised manner for each customer. Such content customisation has not been consistent across most media domains, and particularly on radio streaming and gaming aggregators, which are the two real-world application domains focused in this work. The challenges faced in these application areas are the dynamic nature of user preferences and the difficulty of generating recommendations for less popular content, due to the overwhelming choice and polarisation of available top content. We present the design and implementation of a Reinforcement Learning-based Recommendation System (RLRS) for web applications, using a Deep Deterministic Policy Gradient (DDPG) agent and, as a reward function, a weighted sum of the user Click Distribution (CD) across the recommended items and the Dwell Time (DT), a measure of the time users spend interacting with those items. Our system has been deployed in real production scenarios with preliminary but promising results. Several metrics are used to track the effectiveness of our approach, such as content coverage, category diversity, and intra-list similarity. In both scenarios tested, the system shows consistent improvement and adaptability over time, reinforcing its applicability.

Abstract
Phishing attacks aims to steal sensitive information and, unfortunately, are becoming a common practice on the web. Email phishing is one of the most common types of attacks on the web and can have a big impact on individuals and enterprises. There is still a gap in prevention when it comes to detecting phishing emails, as new attacks are usually not detected. The goal of this work was to develop a model capable of identifying phishing emails based on machine learning approaches. The work was performed in collaboration with E-goi, a multi-channel marketing automation company. The data consisted of emails collected from the E-goi servers in the electronic mail format. The problem consisted of a classification problem with unbalanced classes, with the minority class corresponding to the phishing emails and having less than 1% of the total emails. Several models were evaluated after careful data selection and feature extraction based on the email content and the literature regarding these types of problems. Due to the imbalance present in the data, several sampling methods based on under-sampling techniques were tested to see their impact on the model's ability to detect phishing emails. The final model consisted of a neural network able to detect more than 80% of phishing emails without compromising the remaining emails sent by E-goi clients.

Abstract
In the age of rapidly advancing machine learning capabilities, the pursuit of maximum performance encounters the practical limitations imposed by limited resources in several fields. This work presents a cost-effective proposal for feature selection, which is a crucial part of machine learning processes, and intends to partly solve this problem through computational time reduction. The proposed methodology aims to strike a careful balance between feature exploration and strict computational time concerns, by enhancing the quality and relevance of data. This approach focuses on the use of interim representations of feature combinations to significantly speed up a potentially slow and computationally expensive process. This strategy is evaluated in several datasets against other feature selection methods, and the results indicate a significant reduction in the temporal costs associated with this process, achieving a mean percentage decrease of 85%. Furthermore, this reduction is achieved while maintaining competitive model performance, demonstrating that the selected features remain effective for the learning task. These results emphasize the method's feasibility, confirming its ability to transform machine learning applications in environments with limited resources.