The digital transformation faces tremendous security challenges. In particular, the growing number of cyber-attacks targeting Internet of Things (IoT) systems restates the need for a reliable detection of malicious network activity. This paper presents a comparative analysis of supervised, unsupervised and reinforcement learning techniques on nine malware captures of the IoT-23 dataset, considering both binary and multi-class classification scenarios. The developed models consisted of Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Isolation Forest (iForest), Local Outlier Factor (LOF) and a Deep Reinforcement Learning (DRL) model based on a Double Deep Q-Network (DDQN), adapted to the intrusion detection context. The most reliable performance was achieved by LightGBM. Nonetheless, iForest displayed good anomaly detection results and the DRL model demonstrated the possible benefits of employing this methodology to continuously improve the detection. Overall, the obtained results indicate that the analyzed techniques are well suited for IoT intrusion detection.

New records of 110 species of the acalyptrate Diptera family Agromyzidae are given from Portugal, including Madeira, Porto Santo and the Azores. A quarantine plant pest, Nemorimyza maculosa (Malloch, 1913), was detected in the Old World for the first time. Details on Phytobia xylem- miners and a parthenogenetic Phytomyza species are recorded together with new distribution data. For certain species morphological and taxonomic notes and discussions on known or new host plants are added. A complete checklist of Agromyzidae of Portugal is presented.

Periscelis (Periscelis) fugax sp. nov. is described and illustrated based on series of specimens from Portugal and the Czech Republic. It is probably widespread in Europe but has not been previously recognized from the closely related and externally very similar P. (P.) winnertzii Egger, 1862. The new species differs from P. winnertzii not only in structures of the male and female terminalia (illustrated) but also in colour pattern of pedicel and mesonotum. Morphology of the male and female terminalia is studied in detail in both these species and terminology of their structures is discussed and redefined. A new modern key to Palaearctic species of Periscelis (s. str.) species is constructed.

Seven species in the genus Medetera (Diptera: Dolichopodidae) are described here: Medetera aglaops sp. nov., Medetera corsicana sp. nov., Medetera gibbosipyga sp. nov., Medetera hispanica sp. nov., Medetera lusitana sp. nov., Medetera parva sp. nov., and Medetera rectipyga sp. nov. They all originate from the wider Mediterranean region in southwestern Europe (Iberian Peninsula, southern France, Corsica, Sardinia). The most striking feature in most of the species is the bi- or multicolour pattern of the eyes, most conspicuous in Medetera aglaops sp. nov. and Medetera lusitana sp. nov. This character is shared by the male and female sex and is thus not regarded as Male Secondary Sexual Character. All species belong to the Medetera apicalis species group sensu Bickel and are closely related based on shared characters in the hypandrium and cercus. Interestingly, three species with and four species without the basal pair of anterodorsal and posterodorsal bristles on the mid tibia are represented. This suggests that this diagnostic character has less phylogenetic relevance than previously anticipated. Unlike most Palaearctic Medetera, at least five of the new species are mostly found on rocky substrates in dry biotopes with a sclerophyllous vegetation. This further supports their relationship with species of the Medetera muralis subclade sensu Pollet, Germann and Bernasconi.

This paper explores the concept of the local energy markets and, in particular, the need for trust and security in the negotiations necessary for this type of market. A multi-agent system is implemented to simulate the local energy market, and a trust model is proposed to evaluate the proposals sent by the participants, based on forecasting mechanisms that try to predict their expected behavior. A cyber-attack detection model is also implemented using several supervised classification techniques. Two case studies were carried out, one to evaluate the performance of the various classification methods using the IoT-23 cyber-attack dataset; and another one to evaluate the performance of the developed trust mode.