In recent years, the ever-increasing demand of e-commerce has led to the creation and development of projects related to international freight transportation, using machine learning techniques in order to handle field challenges. The purpose of this work is to forecast the critical time interval cases from warehouses to delivery points between: (a) order and pickup, (b) pickup and ETA delivery and (c) pickup and ATA delivery of an international freight transportation company, utilizing and comparing several machine learning techniques suitable for regression problems. Four machine learning techniques are utilized: Random Forest, Gradient Boosting, WaveNet and Bagging. In order to decide for the method or methods with the best predictive performance, Monte-Carlo simulations were performed for each technique. Simulation results show that Gradient Boosting and Bagging techniques have achieved the best predictions of time intervals for most of the examined cases.

The fungal pathogen Botrytis cinerea (B. cinerea) attacks over 1400 plant species and results in estimated annual losses of $10–100 billion worldwide. In precision agriculture, deep learning (DL) provides reliable tools for rapid and objective plant disease detection. This study presents a unified two-stage DL solution for the automated detection of visible B. cinerea across three major vegetable crops—tomato, pepper, and cucumber—using standard RGB imagery. In the first stage, a YOLOv11-based instance segmentation model accurately localizes leaf regions, achieving a localization accuracy of 87.3% as measured by mAP₅₀. In the second stage, an ensemble of 13 MobileViT variant models analyzes the segmented leaf regions and performs per-crop classification into healthy and infected leaves. The proposed system achieves an overall detection accuracy of 84.05%, with per-class detection of infected leaves at 88.61% for pepper, 82.68% for tomato, and 70.55% for cucumber, measured using the F1-score. These results demonstrate that the proposed approach can reliably detect B. cinerea symptoms across different crops using only RGB data, offering a practical path toward smartphone-based field deployment and integration into decision support systems for timely, symptom-based disease management.

During the last few years, software security especially at the operating system level has been significantly improved for consumer electronics devices. However, few would argue that very little attention has been paid to software intrusiveness when it comes to preserve end-users' privacy and security. In fact, this situation is in contrast to the security incidents leading to data leakages that are constantly on the rise. Furthermore, software developers in majority fail to offer strong warranties regarding the software they implement, meaning that they often do not bear any responsibility for software insecurities and misconfigurations. So, a major open question is whether end-users are in a position to become informed and cope with the various forms of software intrusiveness threatening their private sphere as a consequence of security incident. This is especially important when using their mobile devices as they provide a fine-grained access to private data sources. Note that nowadays mobile devices provide end-users equivalent or even more advanced services than the ones provided by PCs, as they incorporate a variety of sensors, for example, GPS and accelerometer, that can be accessed through operating system services. In this way, mobile apps that get access to end-users' data might invade their privacy and anonymity as a trade-off for offering the app free of charge. Moreover, end-users do not usually notice this, as in most cases such applications get access to this information “silently.” In this direction, end-users and even organizations lack the appropriate tools for accomplishing software privacy risk assessment. For instance, current risk management methodologies (e.g., CRAMM) are basically not considering software “features” in their risk models. While beneficial, these approaches in their current form can be used only for assessing a possible impact in terms of cost, as they do not assess software intrusiveness. So, there is not a way both for end-users and for organizations to use a risk assessment methodology in order to assess mobile application's intrusiveness before application “employment.” In this chapter, we study Android apps' features that can be used to assess their intrusiveness to end-users' private sphere, so end-users can be informed about possible privacy exposure in the direction of a mobile app's risk assessment methodology. To do so, we introduce a practical and an easily employable approach that does not require access to the app's original source code. We evaluate our approach using both goodware and malware.