The rise in the use of social media networks has increased the prevalence of cyberbullying, and time is paramount to reduce the negative effects that derive from those behaviours on any social media platform. This paper aims to study the early detection problem from a general perspective by carrying out experiments over two independent datasets (Instagram and Vine), exclusively using users’ comments. We used textual information from comments over baseline early detection models (fixed, threshold, and dual models) to apply three different methods of improving early detection. First, we evaluated the performance of Doc2Vec features. Finally, we also presented multiple instance learning (MIL) on early detection models and we assessed its performance. We applied $timeawareprecision$ ($TaP$) as an early detection metric to asses the performance of the presented methods. We conclude that the inclusion of Doc2Vec features improves the performance of baseline early detection models by up to 79.6%. Moreover, multiple instance learning shows an important positive effect for the Vine dataset, where smaller post sizes and less use of the English language are present, with a further improvement of up to 13%, but no significant enhancement is shown for the Instagram dataset.

With the increase in the use of Internet interconnected systems, security has become of utmost importance. One key element to guarantee an adequate level of security is being able to detect the threat as soon as possible, decreasing the risk of consequences derived from those actions. In this paper, a new metric for early detection system evaluation that takes into account the delay in detection is defined. Time aware F-score (TaF) takes into account the number of items or individual elements processed to determine if an element is an anomaly or if it is not relevant to be detected. These results are validated by means of a dual approach to cybersecurity, Operative System (OS) scan attack as part of systems and network security and the detection of depression in social media networks as part of the protection of users. Also, different approaches, oriented towards studying the impact of single item selection, are applied to final decisions. This study allows to establish that nitems selection method is usually the best option for early detection systems. TaF metric provides, as well, an adequate alternative for time sensitive detection evaluation.

Early detection is a matter of growing importance in multiple domains as network security, health conditions over social network services or weather forecasts related disasters. It is not enough to make a good decision but it also needs to be made on time. In this paper, we define a method to evaluate detection of anomalies in time-aware systems. To do so, we present the early detection problem from a generic perspective, examine the evaluation metrics available and propose a new metric, named TaP (Time aware Precision) . A set of experiments using three different datasets from different fields are performed in order to compare the behaviour of the different metrics. Two different approaches were followed, first a batch evaluation is performed, followed by a streaming evaluation which allows to present a more realistic behaviour of the systems. For both steps, we propose two sets of experiments. The first one using baseline models, followed by the evaluation of a set of Machine Learning algorithms results. The presented metric allows the amount of items needed to take a decision to be taken into account, not depending on the specific dataset but on the nature of the problem to solve.

In these days, organizations rely on the availability and security of their communication networks to perform daily operations. As a result, network data must be analyzed in order to provide an adequate level of security and to detect anomalies or malfunctions in the systems. Due to the increase of devices connected to these networks, the complexity to analyze data related to its communications also grows. We propose a method, based on Self-Organized Maps, which combine numerical and categorical features, to ease communication network data analysis. Also, we have explored the possibility of using different sources of data.

Objective: Early detection of infectious diseases using physiological and behavioural signals from wearable devices could enable earlier intervention and help reduce disease transmission. This study proposes and evaluates a framework specifically designed for sequential early-warning aimed at detecting infectious disease episodes prior to clinical diagnosis. Methods: In this retrospective observational cohort study, we used the Evidation 2019–2020 Fitbit dataset to formulate infection detection as an episode-level prediction problem. We designed augmented features that capture short-term deviations from each individual’s healthy baseline over sliding windows and evaluated several machine learning models. The framework incorporates probability calibration, validation-based threshold selection, and an optional persistent two-day alarm rule to reduce false-positive alerts. Results: Experiments on data from 435 participants showed that personalized deviation-based features substantially outperformed basic wearable features. The best-performing configuration achieved an episode-level F1-score of 0.784 with an Average Earliness (AE) of 0.555, corresponding to approximately 5.5 days before diagnosis. Additionally, combining step and sleep features provided a favourable trade-off between timeliness (AE = 0.69) and accuracy (F1 = 0.75). Conclusions: Personalized baseline-deviation features, combined with calibrated probabilities and persistent alarm mechanisms, enable accurate and timely early detection of infectious disease episodes from wearable data. The results indicate that recent individualized baselines and step-derived signals provide the most robust predictive information, supporting the potential of wearable devices as scalable tools for population-level infectious disease surveillance.

Communication network data has been growing in the last decades and with the generalisation of the Internet of Things (IoT) its growth has increased. The number of attacks to this kind of infrastructures have also increased due to the relevance they are gaining. As a result, it is vital to guarantee an adequate level of security and to detect threats as soon as possible. Classical methods emphasise in detection but not taking into account the number of records needed to successfully identify an attack. To achieve this, time-aware techniques both for detection and measure may be used. In this work, well-known machine learning methods will be explored to detect attacks based on public datasets. In order to obtain the performance, classic metrics will be used but also the number of elements processed will be taken into account in order to determine a time-aware performance of the method.