Abstract

Abstract

Abstract
The COVID-19 pandemic highlighted the inadequate readiness of numerous nations to address diseases that could potentially evolve into epidemics or pandemics, posing risks to health systems and supply chains. Statistical analysis and predictive models were developed to manage COVID-19 and other diseases that harm public health. However, few public-policy decision-support tools are documented in the literature, although several governments have created them. In line with the previous developments, this tool uses socioeconomic features to model the COVID-19 province's mortality rates. This paper presents a tool to predict the mortality rate of a province using supervised learning techniques, named CuraZone. This tool was validated using 196 provinces in Peru for training and considering 31 characteristics. The tool displays the dataset's most essential characteristics, shows the country's mean square error (MSE), and predicts a province's mortality rate. In addition, the tool contributes to the field of Explainable AI (XAI), as it shows the importance of each feature. Highlighted contributions of this work include the support for the decision-making of governments or stakeholders in epidemics, providing the source code in an open and reproducible way, and the estimated mortality rate for specific populations of a neighborhood, city, or country.

Abstract
The main objectives of this chapter are to provide an overview of chatbot personality dimensions and to analyze the expected impacts on user behavior. To accomplish these objectives, the chapter provides a detailed review of the main contributions in the literature regarding this topic. It highlights the chatbot personality characteristics that are expected to foster user satisfaction, trust, loyalty, and engagement. This information is useful for both practitioners and researchers, particularly related to customer service, as it provides clear guidance on what characteristics to incorporate in chatbots and on what factors need to be further studied in the future.

Abstract
Anomaly detection in the industrial context, identifying defective products and their categorization, is a prevalent task. It is aimed to acknowledge if training and testing multilabel classification models on textures to deploy on an MCU is possible. The focus is deploying lightweight models on MCUs, performing a multilabel classification on textures for industrial usage. For this purpose, a Systematic Literature Review was conducted, which allows knowing the commonly used machine learning models in industrial products anomaly detection and what methods are used to defect detection on textures. Through the Systematic Literature Review, was possible to understand the range of different and combined methods, the methods used in multilabel classification, the most common hyper-parametrizations and popular inferences engines to train machine-learning models to deploy on MCUs, and some techniques applied to overcome the restricted resources of memory and inference time associated with MCUs.

Abstract
Versioned documents are common in many situations and play a vital part in numerous applications enabling an overview of the revisions made to a document or document collection. However, as documents increase in size, it gets difficult to summarize and comprehend all the changes made to versioned documents. In this paper, we propose a novel research problem of contrastive keyword extraction from versioned documents, and introduce an unsupervised approach that extracts keywords to reflect the key changes made to an earlier document version. In order to provide an easy-to-use comparison and summarization tool, an open-source demonstration is made available which can be found at https://contrastive-keyword-extraction.streamlit.app/.