Search

Despite extensive research, measurable benefits of predictive policing are scarce. We argue that powerful models might not always help the work of officers. Furthermore, developed models are often unexplainable, leading to trust issues between police intuition and machine-made prediction. We use a joint approach, mixing criminology and data science knowledge, to design an explainable predictive policing model. The proposed model (a set of explainable decision trees) can predict police resource requirement across the city and explain this prediction based on human-understandable cues (i.e., past event information, weather, and socio-demographic information). The explainable decision tree is then compared to a non-explainable model (i.e., a neural network) to compare performance. Analyzing the decision tree behaviour revealed multiple relations with established criminology knowledge. Weather and recent event distribution were found to be the most useful predictors of police workforce resource. Despite wide research showing relationships between socio-demographic information and police activity, socio-demographic information did not contribute much to the model's performance. Though there is a lack of research on measurable effects of predictive policing applications, we argue that combining human instinct with machine prediction reduces risks of human knowledge loss, machine bias, and lack of confidence in the system. [ABSTRACT FROM AUTHOR]

The ability to identify reliable and sensitive physiological signatures of psychological dimensions is key to developing intelligent adaptive systems that may in turn help to mitigate human error in complex operations. The challenge of this endeavor lies with diagnosticity. Despite different underlying causes, the physiological correlates of workload and acute psychological stress manifest in rather similar ways and can be easily confounded. The current work aimed to build a diagnostic model of mental state through the simultaneous classification of mental workload (varied through three levels of the n-back task) and acute stress (the presence/absence of aversive sounds) with machine learning. Using functional near infrared spectroscopy (fNIRS) and electrocardiography (ECG), the model's classifiers was above-chance to disentangle variations of mental workload from variations of acute stress. Both ECG and fNIRS could predict mental workload level, the best accuracy resulted from the two measures in combination. Stress level could not be accurately diagnosed through ECG alone, only with fNIRS or ECG and fNIRS combined. Individual calibration may be important since stress classification was more accurate for those with higher subjective state anxiety, perhaps due to a greater sensitivity to stress. Mental workload and stress were both better classified with activity in lateral prefrontal regions of the cortex than the medial areas, and the HbO2 signal generally lead to better classification than HHB. The current model represents a step forward to finely discriminate different mental states despite their rather analog physiological correlates. • Cerebral and cardiac measures were able to detect mental workload variations. • Models disentangled physiological variations from 2 combined mental states. • Lateral (versus medial) prefrontal activity best predicted mental states. • Stress classification was more accurate for participants with higher state anxiety. [ABSTRACT FROM AUTHOR]

The rare occurrence of a sound deviating from the auditory background tends to trigger attentional orienting. While some sympathetic physiological responses can be used to index this orienting response, findings surrounding the pupillary dilation response (PDR) as a proxy for the orienting response are conflicting. The current study was tailor-designed to examine whether the PDR satisfies specific criteria of an orienting response index, namely the classic habituation pattern and a sensitivity to the size of the deviation. The PDR decrement to a repeated standard sound, recovery to a deviant sound, and dishabituation to the re-presentation of the standard were assessed for small and large deviations embedded in irrelevant auditory sequences. The PDR not only showed habituation and dishabituation, but also recovered in correspondence with the magnitude of the acoustic deviation. This consistency between variations of the PDR and orienting response's properties indicates that the PDR is a valid index of the auditory orienting response. [ABSTRACT FROM AUTHOR]

A design concept is presented that achieves ultrawideband (UWB) array performance with significantly fewer elements than the traditional approach of using a single wideband antenna element type to fully populate the array. Starting from a conventional 8:1 bandwidth array design of a given aperture size, an array of equivalent aperture and bandwidth is created using scaled elements of three different sizes. This wavelength-scaled equivalent array has fewer than 18% of the original element count, i.e., roughly 6-times fewer elements, a similar reduction in weight, and most importantly, a significant reduction in electronics required to feed the array. If proven viable, array architectures of this type could make UWB arrays significantly more cost effective. In this preliminary numerical study, rigorous full-wave simulation tools are used to test the performance of small but informative wavelength-scaled array configurations of flared-notch radiators for the single-polarization case. [ABSTRACT FROM AUTHOR]

Descriptions of the experimental design and research highlights obtained from a series of four multiagency field projects held near Cape Canaveral, Florida, are presented. The experiments featured a 3 MW, dual-polarization, C-band Doppler radar that serves in a dual capacity as both a precipitation and cloud radar. This duality stems from a combination of the radar's high sensitivity and extremely small-resolution volumes produced by the narrow 0.22° beamwidth and the 0.543 m along-range resolution. Experimental highlights focus on the radar's real-time aircraft tracking capability as well as the finescale reflectivity and eddy structure of a thin nonprecipitating stratus layer. Examples of precipitating storm systems focus on the analysis of the distinctive and nearly linear radar reflectivity signatures (referred to as "streaks") that are caused as individual hydrometeors traverse the narrow radar beam. Each streak leaves a unique radar reflectivity signature that is analyzed with regard to estimating the underlying particle properties such as size, fall speed, and oscillation characteristics. The observed along-streak reflectivity oscillations are complex and discussed in terms of diameter-dependent drop dynamics (oscillation frequency and viscous damping time scales) as well as radar-dependent factors governing the near-field Fresnel radiation pattern and inferred drop–drop interference. [ABSTRACT FROM AUTHOR]

Mental workload assessment is crucial in many real life applications which require constant attention and where imbalance of mental workload resources may cause safety hazards. As such, mental workload and its relationship with heart rate variability (HRV) have been well studied in the literature. However, the majority of the developed models have assumed individuals are not ambulant, thus bypassing the issue of movement-related electrocardiography (ECG) artifacts and changing heart beat dynamics due to physical activity. In this work, multi-scale features for mental workload assessment of ambulatory users is explored. ECG data was sampled from users while they performed different types and levels of physical activity while performing the multi-attribute test battery (MATB-II) task at varying difficulty levels. Proposed features are shown to outperform benchmark ones and further exhibit complementarity when used in combination. Indeed, results show gains over the benchmark HRV measures of 24.41 % in accuracy and of 27.97 % in F1 score can be achieved even at high activity levels. [ABSTRACT FROM AUTHOR]