Abstract
Automated image classification is a thriving field of machine learning, and various successful applications dealing with biological images have recently emerged. In this work, we address the ability of these methods to identify species that are difficult to tell apart by humans due to their morphological similarity. We focus on distinguishing species of wall lizards, namely those belonging to the Podarcis hispanicus species complex, which constitutes a well-known example of cryptic morphological variation. We address two classification experiments: (1) assignment of images of the morphologically relatively distinct P. bocagei and P. lusitanicus; and (2) distinction between the overall more cryptic nine taxa that compose this complex. We used four datasets (two image perspectives and individuals of the two sexes) and three deep-learning models to address each problem. Our results suggest a high ability of the models to identify the correct species, especially when combining predictions from different perspectives and models (accuracy of 95.9% and 97.1% for females and males, respectively, in the two-class case; and of 91.2% to 93.5% for females and males, respectively, in the nine-class case). Overall, these results establish deep-learning models as an important tool for field identification and monitoring of cryptic species complexes, alleviating the burden of expert or genetic identification.

Abstract
Web 2.0 has allowed collaboration, interaction and sharing of information online, such as online review platforms. Consequently, these short and straightforward opinions have increasingly proved to be essential sources of information not only for consumers but also for companies, as they represent the consumer's sincere evaluation, free from any kind of bias. In this sense, there should be an interest in the analysis and monitoring of online reviews by companies, as the result of these actions may provide guidelines to readjust their strategy, support decision-making and ensure the satisfaction of their consumers. To generate useful information to assist decision-making and strategies' implementation by retailers in the Municipality of Porto, online reviews from the GoogleMyBusiness platform were organised, classified, and analysed. 9945 online reviews were extracted, directed to 246 retail adaptations of the Municipality of Porto, from 2017 to 2020, which were later classified by the polarity of sentiment (positive, negative, neutral, or mixed). Sentiment analysis was conducted, combined with statistical tests and frequency distribution tables to discover relevant information for retailers. With sentiment analysis, retailers can understand their consumers and their behaviour to adapt their strategies and make the right decisions to ensure their customers' satisfaction. With the results obtained, this study proves that it is possible to extract useful information from online reviews and reveals that it is still an area of little interest for retailers in the Municipality of Porto.

Abstract
This research establishes the relationship between the digital channels that organizations use to communicate with their audience and the stages of the consumer buying decision process in online stores. Researchers have not treated this relationship in much detail and little-known empirical research has focused on exploring relationships between the two subjects. Establishing this relationship is of crucial importance for organizations and consumers, as it ensures organizations use the digital channels that consumers want. A literature review of digital channels and consumer behavior models was performed, which allowed us to define which are the digital channels and to identify the different models of consumer behavior appropriate for the digital age. A quantitative methodology was used, supported on a questionnaire that allowed us to find out which digital channels are the most appropriate for each stage of the buying decision process. The results show that consumers use more than one digital channel at each stage of the buying decision process and for each stage, a set of digital channels is identifiable that is most preferred. In light of the above, those who are responsible for defining the digital marketing strategy know what types of content they should produce for each digital channel, allowing them to guarantee efficiency in the use of resources while ensuring that consumers get what they want.

Abstract
Body-positive content on social media offers a promising approach to promote positive body image (PBI). However, we need further research in order to better characterize and understand its nature. This study provides a content analysis of adolescents' image-based posts on Facebook. We aimed to determine whether the theoretical six -facet conceptualization of PBI was reflected in adolescents' posts, and whether there were different trends according to gender and time, over a 3-year period. A set of 6,503 images posted by 66 adolescents (51.5% male), were coded for PBI attributes. The results indicate that inner positivity and appreciation of body functionality through involvement in sports and fun activities were the most represented PBI facets. Conversely, imagery representing taking care of the body via healthy food/beverage choices, embracing body diversity, and filtering information in a body-preserving manner, was rarely used to project self-image on Facebook. Gender differences were only found in the appreciation of body functionality via sports activities (more prevalent in boys) and investment in appearance using benign methods, such as the use of cosmetics (more prevalent in girls). Posts addressing appearance and health -promoting self-care behaviors tended to increase in mid-adolescence. We conclude that the adolescents' posts on Facebook reflected several PBI characteristics, with a particular focus on those that enhance a functional view of the body. Future social media and school-level initiatives should prioritize the development of attuned self-care as well as mechanisms to filter messages that could endanger PBI, while also increasing the visibility of the diverse bodies that exist in the world.

Abstract
Body-positive content on social media offers a promising approach to promote positive body image (PBI). However, we need further research in order to better characterize and understand its nature. This study provides a content analysis of adolescents’ image-based posts on Facebook. We aimed to determine whether the theoretical six-facet conceptualization of PBI was reflected in adolescents’ posts, and whether there were different trends according to gender and time, over a 3-year period. A set of 6,503 images posted by 66 adolescents (51.5% male), were coded for PBI attributes. The results indicate that inner positivity and appreciation of body functionality through involvement in sports and fun activities were the most represented PBI facets. Conversely, imagery representing taking care of the body via healthy food/beverage choices, embracing body diversity, and filtering information in a body-preserving manner, was rarely used to project self-image on Facebook. Gender differences were only found in the appreciation of body functionality via sports activities (more prevalent in boys) and investment in appearance using benign methods, such as the use of cosmetics (more prevalent in girls). Posts addressing appearance and health-promoting self-care behaviors tended to increase in mid-adolescence. We conclude that the adolescents’ posts on Facebook reflected several PBI characteristics, with a particular focus on those that enhance a functional view of the body. Future social media and school-level initiatives should prioritize the development of attuned self-care as well as mechanisms to filter messages that could endanger PBI, while also increasing the visibility of the diverse bodies that exist in the world.

Abstract
The management of buildings responsible for the energy storage and control can be optimized with the support of forecasting techniques. These are essential on the finding of load consumption patterns being these last involved in decisions that analyze which forecasting technique results in more accurate predictions in each context. This paper considers two forecasting methods known as artificial neural network and k-nearest neighbor involved in the prediction of consumption of a building composed by devices recording consumption and sensors data. The forecasts are performed in five minutes periods with the forecasting technique taken into account as a potential to improve the accuracy of predictions. The decision making considers the Multi-armed Bandit in reinforcement learning context to find the best suitable algorithm in each five minutes period thus improving the predictions accuracy in forecasting. The reinforcement learning has been tested in upper confidence bound and greedy algorithms with several exploration alternatives. In the case-study, three contexts have been analyzed. (C) 2022 The Author(s). Published by Elsevier Ltd.