This work uses deep learning methods for intraday directional movements prediction of Standard & Poor's 500 index using financial news titles and a set technical indicators as input. Deep can detect analyze complex patterns interactions in the data automatically allowing speed up trading process. paper focus on architectures such Convolutional Neural Networks (CNN) Recurrent (RNN), which have had good results traditional NLP tasks. Results has shown that CNN be better than RNN catching...

This work uses deep learning models for daily directional movements prediction of a stock price using financial news titles and technical indicators as input. A comparison is made between two different sets indicators, set 1: Stochastic %K, %D, Momentum, Rate change, William's %R, Accumulation/Distribution (A/D) oscillator Disparity 5; 2: Exponential Moving Average, Average Convergence-Divergence, Relative Strength Index, On Balance Volume Bollinger Bands. Deep methods can detect analyze...

In recent years, we have seen scientists attempt to model and explain human dynamics in particular movement. Many aspects of our complex life are affected by movement such as disease spread epidemics modeling, city planning, wireless network development, disaster relief, name a few. Given the myriad applications, it is clear that complete understanding how people move space can lead considerable benefits society. most works, focused on idea movements biased towards frequently-visited...

Abstract Community detection is one of the most important tasks in network analysis. It increasingly clear that quality measures are not sufficient for assessing communities and structural properties play a key hole understanding how nodes organized network. This work presents comparative study some representative state-of-the-art methods overlapping community from perspective identified by them. Experiments with synthetic real-world benchmark Ground-Truth networks show that, although able...

This study presents an automated method for objectively measuring rock heterogeneity via raw X-ray micro-computed tomography (micro-CT) images, thereby addressing the limitations of traditional methods, which are time-consuming, costly, and subjective. Unlike approaches that rely on image segmentation, proposed processes micro-CT images directly, identifying textural heterogeneity. The is partitioned into subvolumes, where attributes calculated each one, with entropy serving as a measure...

Precipitation nowcasting can predict and alert for any possibility of abrupt weather changes which may cause both human material risks. Most the conventional methods extrapolate radar echoes, but precipitation is still a challenge, mainly due to rapid in meteorological systems time required numerical simulations. Recently video prediction deep learning (VPDL) algorithms have been applied nowcasting. In this study, we use VPDL PredRNN++ sequences reflectivity images future sequence up 1-h...

Social structures influence human behavior, including their movement patterns. Indeed, latent information about an individual's can be present in the mobility patterns of both acquaintances and strangers. We develop a "colocation" network to distinguish ego's social ties from those not socially connected ego but who arrive at location similar time as ego. Using entropic measures, we analyze bound predictive pattern its flow types ties. While former generically provide more information,...

This paper presents two case studies of municipal solid waste site location using a decision-support system based on fuzzy logic. problem is very complex, as it requires the evaluation different criteria, which involve environmental, social and economic data. Such data deal with wide range information that not only quantitative, but also qualitative knowledge. In order to this characteristic, developed employs rules due its ability treat linguistic variables human way thinking. Conventional...

Defining and measuring spatial inequalities across the urban environment remains a complex elusive task which has been facilitated by increasing availability of large geolocated databases. In this study, we rely on mobile phone dataset an entropy-based metric to measure attractiveness location in Rio de Janeiro Metropolitan Area (Brazil) as diversity visitors’ residence. The results show that given measured entropy is important descriptor socioeconomic status location, can thus be used proxy...

Community structure detection is one of the major research areas network science and it particularly useful for large real networks applications. This work presents a deep study most discussed algorithms community based on modularity measure: Newman’s spectral method using fine-tuning stage Clauset, Newman, Moore (CNM) with its variants. The computational complexity analysed development high performance code to accelerate execution these without compromising quality results, according...

This paper presents the identification of nonlinear dynamical systems by recurrent fuzzy system (RFS) models. Two types RFS models are discussed: Takagi-Sugeno-Kang (TSK) type and linguistic or Mamdani type. Both equivalent latter model may be represented a finite-state automaton (FFA). An procedure is proposed based on standard general purpose genetic algorithm (GA). First, TSK rule parameters estimated and, in second step, converted into an model. The parameter evaluated some benchmark...

Rapid urbanization places increasing stress on already burdened transportation systems, resulting in delays and poor levels of service. Billions spatiotemporal call detail records (CDRs) collected from mobile devices create new opportunities to quantify solve these problems. However, there is a need for tools map data onto existing infrastructure. In this work, we propose system that leverages identify patterns road usage. First, develop an algorithm mine billions calls learn location...