This study aims to explore the potential of Internet Things (IoT) devices and explainable Artificial Intelligence (AI) techniques in predicting biomarker values associated with GDM when measured 13–16 weeks prior diagnosis. We developed a system that forecasts biomarkers such as LDL, HDL, triglycerides, cholesterol, HbA1c, results from Oral Glucose Tolerance Test (OGTT) including fasting glucose, 1-hour, 2-hour post-load glucose values. These are predicted based on sensory measurements...

After setting the performance benchmarks for image, video, speech and audio processing, deep convolutional networks have been core to greatest advances in image recognition tasks recent times.This raises question of whether there are any benefit targeting these remarkable architectures with unattempted task recognising human rights violations through digital images.Under this perspective, we introduce a new, well-sampled rights-centric dataset called Human Rights Understanding (HRUN).We...

Abstract The vision of future electronic marketplaces (e-markets) is that markets being populated by autonomous intelligent entities—software, trading, e-agents—representing their users or owners and conducting business on behalf. For this to materialize, one fundamental issue needs be addressed trust. First, need able trust the agents will do what they say do. Second, confident privacy protected security risks involved in entrusting perform transactions behalf are minimized. Finally,...

While deep learning and backpropagation continue to dominate the field of machine in terms benchmarks versatility, recent neuroscientific advances shed light on more biologically plausible approaches. Spiking neural networks (SNNs), modelled after action potential dynamics, offer inherent time sensitivity efficiency performance complexity. investigating paradigms support such alternatives, we attempt answer whether reservoir computing can benefit from a spiking network based implementation...

This paper presents a software tool that supports teaching, assessment and student directed learning in AI multi-agent systems courses. E-game is configurable on-line auction server promotes active through the use of simulation exercises. The system can be used to teach demonstrate principles negotiation protocols, market mechanisms agent strategies. Moreover, it provides facilities for instructors develop exercises teaching or providing students with challenging interactive environment...

General Video Game Playing (GVGP) allows for the fair evaluation of algorithms and agents as it minimizes ability an agent to exploit apriori knowledge in form game specific heuristics. In this paper we compare four possible combinations evolutionary learning using Separable Natural Evolution Strategies our algorithm choice; linear function approximation with Softmax search e-greedy policies neural networks same policies. The explored research play each games during a sequence 1000 matches,...

Fiction, a prime form of entertainment, has evolved into multiple genres which one can broadly attribute to different forms stories. In this paper, we examine the hypothesis that works fiction be characterised by emotions they portray. To investigate hypothesis, use work fictions in Project Gutenberg and basic emotional content each individual sentence using Ekman's model. A time-smoothed version for emotion is used train extremely randomized trees. We show through 10-fold Cross-Validation...

While much research has been performed on query logs collected for major Web search engines, log analysis to enhance smaller and more focused collections attracted less attention. Our hypothesis is that an intranet engine can be enhanced by adapting the system real users' behaviour through exploiting its logs. In this work we describe how a constantly domain model used identify capture changes in requirements over time. We employ algorithm dynamically builds from modifications taken employs...

Despite high expectations, shopbots have yet to significantly facilitate a richer, more satisfying online shopping experience for users. By taking advantage of semantic Web and services technologies, however, researchers can overcome current technological limitations finally realize the shopbot's significant potential

Identifying potential abuses of human rights through imagery is a novel and challenging task in the field computer vision, which will enable to expose violations over large-scale data that may otherwise be impossible. While standard databases for object scene categorization contain hundreds different classes, largest available dataset contains only four classes. Here, we introduce archive (HRA) database, verified-by-experts repository 3050 photographs, labeled with semantic categories,...

Detecting events from streams of data is challenging due to the characteristics such streams: elements arrive in real-time and at high velocity, size typically unbounded while it not possible backtrack over past or maintain review entire history. Social networks are a good source for event identification as they generate huge amount timely information representing what users posting discussing. In this research, we developing methods detection data. More specifically, presenting framework...

The Web services paradigm has enabled an increasing number of providers to host remotely accessible services. However, the true potential such a distributed infrastructure can only be reached when autonomic combined together as parts workflow, in order collectively achieve functionality. In this article, we present our work area automatic workflow composition among with semantically described functionality capabilities. For purpose, use set heuristics derived from connectivity structure...