Abstract
Interictal Epileptiform Discharge (IED) detection in EEG signals is widely used in the diagnosis of epilepsy. Visual analysis of EEGs by experts remains the gold standard, outperforming current computer algorithms. Deep learning methods can be an automated way to perform this task. We trained a VGG network using 2-s EEG epochs from patients with focal and generalized epilepsy (39 and 40 patients, respectively, 1977 epochs total) and 53 normal controls (110770 epochs). Five-fold cross-validation was performed on the training set. Model performance was assessed on an independent set (734 IEDs from 20 patients with focal and generalized epilepsy and 23040 normal epochs from 14 controls). Network visualization techniques (filter visualization and occlusion) were applied. The VGG yielded an Area Under the ROC Curve (AUC) of 0.96 (95% Confidence Interval (CI) = 0.95 - 0.97). At 99% specificity, the sensitivity was 79% and only one sample was misclassified per two minutes of analyzed EEG. Filter visualization showed that filters from higher level layers display patches of activity indicative of IED detection. Occlusion showed that the model correctly identified IED shapes. We show that deep neural networks can reliably identify IEDs, which may lead to a fundamental shift in clinical EEG analysis.

Abstract
Recommendation systems are usually evaluated through accuracy and classification metrics. However, when these systems are supported by crowdsourced data, such metrics are unable to estimate data authenticity, leading to potential unreliability. Consequently, it is essential to ensure data authenticity and processing transparency in large crowdsourced recommendation systems. In this work, processing transparency is achieved by explaining recommendations and data authenticity is ensured via blockchain smart contracts. The proposed method models the pairwise trust and system-wide reputation of crowd contributors; stores the contributor models as smart contracts in a private Ethereum network; and implements a recommendation and explanation engine based on the stored contributor trust and reputation smart contracts. In terms of contributions, this paper explores trust and reputation smart contracts for explainable recommendations. The experiments, which were performed with a crowdsourced data set from Expedia, showed that the proposed method provides cost-free processing transparency and data authenticity at the cost of latency.

Abstract

Abstract
The use of adequate relay selection techniques is crucial to improve the behavior of cooperation based approaches in Wireless Sensor Networks (WSN). The Optimized Relay Selection Technique (ORST) is a relay selection technique that may be reduced to the application on classic set-covering problem (SCP) to WSN. The SCP seeks to find a minimum number of sets that contain all elements of all data sets. The SCP can be solved with different types of algorithms. This paper assesses the performance and quality of three different algorithms to solve the SCP generated by the previously proposed ORST technique, considering performance metrics relevant within WSNs context. The analysis was performed by simulation using the OMNeT++ tool and the WSN framework Castalia. The simulation results show that the branch and bound algorithm excels when compared to other state-of-the-art approaches.

Abstract
Domestic energy policies destined to foster the use of end-use electric technologies could cause rapid penetration of new residential loads and, consequently, this could cause a significant increase in the demand for electricity in urban areas. This paper presents a spatial-temporal growth model for estimating the adoption of new end-use electric technologies encouraged by energy-efficiency policies. The proposed method consists of three modules: temporal, spatial and grouping. The temporal module calculates by districts or census tracts of a city, the percentage of homes in which residents are prospective buyers of a new end-use electric technology. Then, the spatial module adjusts the calculations made by the temporal module, considering the spatial interactions among the inhabitants of the districts. Finally, the grouping module discovers the low-voltage transformer where the prospective buyers are connected. The results of the proposed model are a spatial database with information related to the percentage of homes in which residents are prospective buyers of a new end-use electric technology, as well as the number of prospective buyers connected to each low-voltage transformer. The results can visualize through thematic maps to identify the districts where the new technology will have faster adoption. The proposed method was employed to estimate the adoption of induction heating cookers in a medium-sized Ecuadorian city. The Ecuadorian government has developed a program of economic subsidies to encourage its population to use this electrical appliance. The results from this application are an important tool to estimate the spatial increase in electricity demand, decide important issues related to the planning of distributed resources, and develop demand-side management programs. Furthermore, the results can be used to evaluate and manage energy policies formulated to achieve environmental and energy goals.

Abstract
Researchers in a variety of fields are increasingly adopting crowdsourcing as a reliable instrument for performing tasks that are either complex for humans and computer algorithms. As a result, new forms of collective intelligence have emerged from the study of massive crowd-machine interactions in scientific work settings as a field for which there is no known theory or model able to explain how it really works. Such type of crowd work uses an open participation model that keeps the scientific activity (including datasets, methods, guidelines, and analysis results) widely available and mostly independent from institutions, which distinguishes crowd science from other crowd-assisted types of participation. In this paper, we build on the practical challenges of crowd-AI supported research and propose a conceptual framework for addressing the socio-technical aspects of crowd science from a CSCW viewpoint. Our study reinforces a manifested lack of systematic and empirical research of the symbiotic relation of AI with human computation and crowd computing in scientific endeavors.