Multi-sensor fusion refers to methods used for combining information coming from several sensors (in some cases, different ones) with the aim make one sensor compensate weaknesses of others or improve overall accuracy reliability a decision-making process. Indeed, this area has made progress, and combined use been so successful that many authors proposed variants methods, point it is now hard tell which them best given set application context. To address issue choosing an adequate method, we...

Upon an intrusion, security staff must analyze the IT system that has been compromised, in order to determine how attacker gained access it, and what he did afterward. Usually, this analysis reveals run exploit takes advantage of a vulnerability. Pinpointing, given log file, execution one such exploit, if any, is very valuable for computer security. This both because it speeds up process gathering evidence helps taking measures prevent further e.g., by building applying appropriate attack...

Distributed denial of service (DDoS) attacks aim to deplete the network bandwidth and computing resources targeted victims. Low-rate DDoS exploit protocol features such as transmission control (TCP) three-way handshake mechanism for connection establishment TCP congestion-control induced backoffs attack at a much lower rate still effectively bring down computer systems. Most statistical machine/deep learning-based detection methods proposed in literature require keeping track packets by...

Anomaly detection is a well-known topic in cybersecurity. Its application to the Internet of Things can lead suitable protection techniques against problems such as denial service attacks. However, Intrusion Detection Systems based on Artificial Intelligence, defense mechanism, need robust data sources achieve strong generalization levels from knowledge domain interest. Therefore, this research we present LATAM-DDoS-IoT dataset, which results collaboration among Aligo, Universidad de...

This research aims at applying the Artificial Organic Network (AON), a nature-inspired, supervised, metaheuristic machine learning framework, to develop new algorithm based on this class. The focus of is model and predict stock markets Index Tracking Problem (ITP). In work, we present algorithm, AON that call Halocarbon Compounds, or AHC for short. study, compare against genetic algorithms (GAs), by forecasting eight market indices. Additionally, performed cross-reference comparison results...

Most studies in masquerade detection focus mainly on the user action, ignoring object upon which that action is performed. This may yield limited models, since, for example, command execution (an action) usually ends up transformation of a file (the object). The overall goal this paper to prove paramount distinguishing from masquerade. With mind, we have developed new approach detection, called system navigation, and tested our ideas using Windows-Users Windows-Intruder simulations Logs Data...

DNS DDoS attacks may severely affect the operation of computer networks, prompting need for methods able to timely detect them, and then apply mitigation countermeasures. Visual models have been used an ongoing attack, but often demand continuous attention from IT staff. However, machine learning techniques could complement a visual model with further information on-time alerts that help officers give only when attack is in progress at its very early stage. In this paper, we present...

In this work, we evaluate the effectiveness of a multicomponent program that includes psychoeducation in academic stress, mindfulness training, and biofeedback-assisted mindfulness, while enhancing Resilience to Stress Index (RSI) students through control autonomic recovery from psychological stress. Participants are university enrolled excellence granted an scholarship. The dataset consists intentional sample 38 undergraduate with high performance, 71% (27) women, 29% (11) men, 0% (0)...

We define personal risk detection as the timely identification of when someone is in midst a dangerous situation, for example, health crisis or car accident, events that may jeopardize person's physical integrity. work under hypothesis risk-prone situation produces sudden and significant deviations standard physiological behavioural user patterns. These changes can be captured by group sensors, such accelerometer, gyroscope, heart rate. introduce dataset, called PRIDE, which provides...

This study introduces the One-Class K-means with Randomly-projected features Algorithm (OCKRA). OCKRA is an ensemble of one-class classifiers built over multiple projections a dataset according to random feature subsets. Algorithms found in literature spread wide range applications where ensembles have been satisfactorily applied; however, none oriented area under our study: personal risk detection. has designed aim improving detection performance problem posed by Personal RIsk...

People often face risk-prone situations, that range from a mild event to severe, life-threatening scenario. Risk situations stem number of different scenarios: health condition, hazard situation due natural disaster, dangerous because one is being subject crime or physical violence, among others. The lack prompt response, calling for assistance, may severely worsen the consequences. In this paper, we propose novel visualisation method track and identify, in real-time, when person under...

The Domain Name System (DNS) is a critical infrastructure of any network, and, not surprisingly common target cybercrime. There are numerous works that analyse higher level DNS traffic to detect anomalies in the or other network service. By contrast, few efforts have been made study and protect recursive level. In this paper, we introduce novel abstraction flooding attack, kind Distributed Denial Service (DDoS). crux our lies on simple observation: Recursive queries, from IP addresses domain...

Early identification of mental disorder symptoms is crucial for timely treatment and reduction recurring disabilities. A tool to help individuals recognize warning signs important. We posit that such a would have rely on longitudinal analysis patterns trends in the individual's daily activities mood, which can now be captured through data from wearable activity trackers, speech recordings mobile devices, own description their state. In this paper, we describe developed by our team detect...

This study examines the communications of English- and Spanish-speaking Twitter users through traditional deep learning algorithms to automatically recognize whether they live with one nine mental health conditions. We created two datasets in English Spanish. The "diagnosed" set comprises timeline 1,500 who explicitly reported or more their posts having been diagnosed following: ADHD, Anxiety, Autism, Bipolar, Depression, Eating disorders, OCD, PTSD, Schizophrenia. "control" 1,700 randomly...

Sensors are becoming more and ubiquitous as their price availability continue to improve, they the source of information for many important tasks. However, use sensors has deal with noise failures. The lack reliability in led forms redundancy, but simple solutions not always best, precise way which several combined a big impact on overall result. In this paper, we discuss how combination coming from different sensors, acting thus "virtual sensors", context human activity recognition,...

This study proposes a new index to measure the resilience of an individual stress, based on changes specific physiological variables. These variables include electromyography, which is muscle response, blood volume pulse, breathing rate, peripheral temperature, and skin conductance. We measured data with biofeedback device from 71 individuals subjected 10-min psychophysiological stress test. The exploration revealed that features' variability among test phases could be observed in...

Predicting mental health conditions from speech has been widely explored in recent years. Most studies rely on a single sample each subject to detect indicators of particular disorder. These ignore two important facts: certain disorders tend co-exist, and their severity tends vary over time. This work introduces longitudinal dataset labeled with depression, anxiety, stress scores using the DASS-21 self-report questionnaire, describes machine-learning pipeline determine three acoustic...

Introduction The rise in global temperatures due to climate change has escalated the frequency and intensity of wildfires worldwide. Beyond their direct impact on physical health, these can significantly mental health. Conventional health studies predominantly rely surveys, often constrained by limited sample sizes, high costs, time constraints. As a result, there is an increasing interest accessing social media data study effects Methods In this study, we focused Twitter users affected...

<sec> <title>BACKGROUND</title> Depression is a mental health condition that affects millions of people worldwide. Although common, it remains difficult to diagnose due its heterogeneous symptomatology. Mental questionnaires are currently the most used assessment method screen depression; these, however, have subjective nature their dependence on patients' self-assessments. Researchers been interested in finding an accurate way identifying depression through objective biomarker. Recent...

In this work, we present a new and efficient algorithm to perform short-term market trend forecast, based on the Artificial Organic Networks (AON) metaheuristic machine learning framework. Regarding goal, concept of Halocarbon Compounds (AHC) or AHC-algorithm as bio-inspired supervised AON Through our research, contrast forecast acquired with proposed AHC model, previously reported outcomes using Hydrocarbon (AHN) in similar tasks. The AHN is first formally defined topology AON, making vital...

: Access to a primary care provider is not guaranteed for many living in rural settings. Notably, populations experience higher degree of burden from chronic diseases compared urban-dwellers. For example, diabetes can go undiagnosed and undertreated with lack care. To address these gaps at large, family medicine practice western North Carolina, multidisciplinary pharmacist-led clinic was developed.