Search

High-frequency financial data is more difficult to predict than low-frequency data because it possesses nonlinearity, nonstationarity, higher volatility, and long memory and is frequently accompanied by the jump phenomena. In this paper, the nonparametric regression (NR) model based on kernel function is used to fit the nonlinear relationship between the nonstationary series Y t and its lagging series to model the trend of high frequency financial time series. Furthermore, the deep learning LSTM (long short-term memory) model is applied to capture the high volatility and frequent jumps of high frequency financial data and to improve the forecasting accuracy of the residual series. The results demonstrate that the hybrid NR and LSTM model has greatly improved the forecasting accuracy in several evaluation criteria. In comparison to NR, support vector machine (SVM), LSTM, ARIMA and NR-SVM models, the mean absolute error (MAE) of NR-LSTM has reduced by 89.78%, 97.85%, 86.48%, 32.47% and 89%, respectively. In addition, we have constructed the trading strategy for the Shanghai-Shenzhen 300 index by using the NR-LSTM model. The NR-LSTM model can continue to provide good returns even during a bear market, which can serve as a guide for investors. Furthermore, the NR-LSTM model also exhibits the best forecasting effect when we model the high-frequency data of Ping An bank in China, the FTSE 100 index in the UK, and the S&P 500 index in the US. [ABSTRACT FROM AUTHOR]

Most existing deep learning based methods for vessel segmentation neglect two important aspects of retinal vessels: The orientation information of vessels and the contextual information of the whole fundus region. In this paper, we propose a robust orientation and context entangled network (OCE-Net), which can extract complex orientation and context information from blood vessels. To achieve complex orientation-aware convolution, we propose a dynamic complex orientation-aware convolution (DCOA Conv) to extract complex vessels with multiple orientations for improving vessel continuity. To simultaneously capture the global context information and emphasize the important local information, we propose a global and local fusion module (GLFM) to simultaneously model the long-range dependency of vessels and give sufficient attention to local thin vessels. A novel orientation and context entangled nonlocal (OCE-NL) module is also proposed to entangle the orientation and context information together. In addition, an unbalanced attention refining module (UARM) is proposed to deal with the unbalanced pixel numbers of the background and thick and thin vessels. Extensive experiments were performed on several commonly used datasets (DRIVE, STARE, and CHASEDB1) and some more challenging datasets (AV-WIDE, UoA-DR, RFMiD, and UK Biobank). The ablation study results show the proposed method's good performance in maintaining the continuity of thin vessels, and the comparative experimental results show OCE-Net's good performance in retinal vessel segmentation. Thus, the proposed framework can effectively carry out retinal vessel segmentation. • An Orientation and Context Entangled Network is proposed for vessel segmentation. • A DCOA convolution is proposed to capture the orientation of vessels. • A Global and Local Fusion Module is proposed to modeling the global dependencies. • An OCE-DNL is proposed to entangle the orientation and context. • An Unbalanced Attention Refine Module is designed to refine the output features. [ABSTRACT FROM AUTHOR]

• A novel method for investment builds from classic statistics and deep-learning. • Portfolio formation by mean-variance and deep learning for predictive finance. • Emphasis on preselection of high quality assets. • Considering long-term dependences in time-series fluctuation. • The results outperform other strategies with respect to potential returns and risks. Portfolio theory is an important foundation for portfolio management which is a well-studied subject yet not fully conquered territory. This paper proposes a mixed method consisting of long short-term memory networks and mean-variance model for optimal portfolio formation in conjunction with the asset preselection, in which long-term dependences of financial time-series data can be captured. The experiment uses a large volume of sample data from the UK Stock Exchange 100 Index between March 1994 and March 2019. In the first stage, long short-term memory networks are used to forecast the return of assets and select assets with higher potential returns. After comparing the outcomes of the long short-term memory networks against support vector machine, random forest, deep neural networks, and autoregressive integrated moving average model, we discover that long short-term memory networks are appropriate for financial time-series forecasting, to beat the other benchmark models by a very clear margin. In the second stage, based on selected assets with higher returns, the mean-variance model is applied for portfolio optimisation. The validation of this methodology is carried out by comparing the proposed model with the other five baseline strategies, to which the proposed model clearly outperforms others in terms of the cumulative return per year, Sharpe ratio per triennium as well as average return to the risk per month of each triennium. i.e. potential returns and risks. [ABSTRACT FROM AUTHOR]

• Proposed an intelligent diagnosis framework for skin lesion classification. • Designed a multi-class multi-level algorithm to enhance the accuracy. • Proposed improved techniques for noise removal from the images. • Deep learning and other machine learning approaches developed and compared. • So far, the best multi-class result (∼96.5% accuracy) achieved using deep learning. Skin diseases remain a major cause of disability worldwide and contribute approximately 1.79% of the global burden of disease measured in disability-adjusted life years. In the United Kingdom alone, 60% of the population suffer from skin diseases during their lifetime. In this paper, we propose an intelligent digital diagnosis scheme to improve the classification accuracy of multiple diseases. A Multi-Class Multi-Level (MCML) classification algorithm inspired by the "divide and conquer" rule is explored to address the research challenges. The MCML classification algorithm is implemented using traditional machine learning and advanced deep learning approaches. Improved techniques are proposed for noise removal in the traditional machine learning approach. The proposed algorithm is evaluated on 3672 classified images, collected from different sources and the diagnostic accuracy of 96.47% is achieved. To verify the performance of the proposed algorithm, its metrics are compared with the Multi-Class Single-Level classification algorithm which is the main algorithm used in most of the existing literature. The results also indicate that the MCML classification algorithm is capable of enhancing the classification performance of multiple skin lesions. [ABSTRACT FROM AUTHOR]