Long Shot Term Memory Research Articles

Lung Image Classification Based On Long-Short Term Memory recurrent neural network

Due to various factors such as complicated lung imaging and rapidly growing amount of data, the task for imaging technicians is arduous. The emergence of artificial intelligence assisted diagnosis technology comes just in time.To effectively classify lung partial images and alleviate the burden of medical application, a deep learning method based on attention mechanism is therefore developed. The proposed model uses deep learning as the basic integrates Long-Shot term memory (LSTM) the recurrent neural network (RNN). Technology for lung imaging diagnosis based on artificial intelligence has evolved through time from combined diagnosis of multiple diseases to the diagnosis of a single specific disease. The suggested network’s overall classification accuracy, according to experiments, is 95.93%, which is 1.019% greater than that of the deep learning basic network. It also outperforms the VGG16 and VGG19 networks in terms of classification performance. Finally, the benefits and drawbacks of the suggested algorithm are explored, as well as the future development path.

Towards the estimation of quantity of fuel consumed in steam generation through predictive modelling of feedwater temperature

Fuel consumption is a critical parameter that every steam boiler operator seeks to minimize, due to its direct relationship with the operational cost of the steam boiler. Feedwater temperature has been identified as a key parameter that influences the fuel consumption of a steam boiler. In view of this, this study aimed to develop predictive models of feedwater temperature in order to estimate the quantity of fuel consumed by the boiler. Feedwater temperature data of a steam boiler has been acquired, examined, and found to be of time-dependent characteristic. Four time-dependent models, namely, Autoregressive Moving Average, Neural Network, Long-Shot Term Memory, and Autoregressive Integrated Moving Average, have been used in modelling the feedwater temperature and validated. The Autoregressive Moving Average, ARMA(1,1) successfully predicts the feedwater temperature within a margin of error of ± 3.967 at 95% confidence level with validation criteria R, RMSE and MSE values recorded as 0.827, 1.022 and 1.044 respectively. A negative linear correlation is found between the feedwater temperature and the quantity of fuel consumed by the steam boiler. Hence, feedwater temperature can be used to control the quantity of fuel consumed by a steam boiler.

I2M: Intelligent Information Management for Rendering IoE Services in Society 5.0

In this paper, we propose an intelligent information management scheme – I2M – an two level delay-tolerant profile matching scheme for provisioning Internet-of-Everything (IoE) services in a Society 5.0. The primary aim of Society 5.0 is to provide seamless real-time IoE services using artificial intelligence (AI) and next-generation networks (NGNs). In this context, we introduce Information-Centric Satellite Network (ICSN), which is expected to enable efficient in-network caching, content delivery, and large area coverage in Society 5.0. The existing works on ICSNs suggest caching the information, requested by the user, in the satellites. However, the satellites are storage-constraint in nature. Therefore, storing information in such satellites, in the long run, may not be a suitable solution. Moreover, the existing solutions increase the delay while a cache-miss occurs in ICSN, which is unacceptable in providing real-time IoE services in Society 5.0. On the other hand, data requested by the increased number of users results in reduced network lifetime and transmission efficiency for Society 5.0 without having a suitable load balancing mechanism in the existing ICSN. To address these issues, we design I2M, which is capable of managing information in a Society 5.0. In I2M, we introduce a tabular data structure to store, update, and replace in-network information in the resource-constrained satellites. Using this tabular data structure, I2M performs a 2-level profile matching scheme, which helps to efficiently access the in-network information and reduce the delay for cache-miss in satellites. Additionally, to increase the transmission efficiency and network lifetime, we apply a long-shot term memory (LSTM)-based intelligent load balancing mechanism in I2M. The extensive simulation results show that I2M outperforms as compared to the existing schemes for ICSNs, in terms of energy consumption, packet loss, and delay. We observe that I2M is capable of reducing delay by 35-55%, energy consumption by 34-54%, and packet loss by 36-53%.

Predictions for COVID-19 with deep learning models of LSTM, GRU and Bi-LSTM.

COVID-19, responsible of infecting billions of people and economy across the globe, requires detailed study of the trend it follows to develop adequate short-term prediction models for forecasting the number of future cases. In this perspective, it is possible to develop strategic planning in the public health system to avoid deaths as well as managing patients. In this paper, proposed forecast models comprising autoregressive integrated moving average (ARIMA), support vector regression (SVR), long shot term memory (LSTM), bidirectional long short term memory (Bi-LSTM) are assessed for time series prediction of confirmed cases, deaths and recoveries in ten major countries affected due to COVID-19. The performance of models is measured by mean absolute error, root mean square error and r2_score indices. In the majority of cases, Bi-LSTM model outperforms in terms of endorsed indices. Models ranking from good performance to the lowest in entire scenarios is Bi-LSTM, LSTM, GRU, SVR and ARIMA. Bi-LSTM generates lowest MAE and RMSE values of 0.0070 and 0.0077, respectively, for deaths in China. The best r2_score value is 0.9997 for recovered cases in China. On the basis of demonstrated robustness and enhanced prediction accuracy, Bi-LSTM can be exploited for pandemic prediction for better planning and management.

Disruption predictor based on neural network and anomaly detection on J-TEXT

Disruption prediction is essential for the safe operation of a large scale tokamak. Existing disruption predictors based on machine learning techniques have good prediction performance, but all these methods need large training datasets including many disruptions to develop their successful prediction capability. Future machines are unlikely to provide enough disruption samples since these cause excessive machine damage and the prediction models used are difficult to extrapolate to a machines that the predictor was not trained on. In this paper, a disruption predictor based on a deep learning and anomaly detection technique has been developed. It regards the disruption as an anomaly, and can learn on non-disruptive shots only. The model is trained to extract the hidden features of various non-disruptive shots with a convolutional neural network and a long-shot term memory (LSTM) recurrent neural network. It will predict the future trend of selected diagnostics, then using the predicted future trend and the measured signal to calculate an outlier factor to determine if a disruption is coming. It was tested with J-TEXT discharges in flat top phase and can demonstrate comparable performance to current machine learning disruption prediction techniques, without requiring a disruption data set. This could be applied to future tokamaks and reduce the dependency on disruptive experiments.

Exemplar based regular texture synthesis using LSTM

Exemplar based texture synthesis is an important technique for image processing and computer graph in texture mapping. So far, great achievements have been made in this field. However, both traditional and modern methods based on deep learning are making errors in synthesizing patterned texture due to the failure for catching the regularity of texture. To obtain a better synthesized result, a new framework for regular texture synthesis is proposed in this paper. Besides, we use recurrent neural network (RNN) of long-shot term memory (LSTM) to produce a regular texture based on exemplar. Our method can generate at any size texture without errors, which is an improvement for texture synthesis with deep learning techniques.Compared with traditional method as well as deep learning method, our method is obviously better in synthesizing regular texture.

Sensor Drift Compensation Based on the Improved LSTM and SVM Multi-Class Ensemble Learning Models.

Drift is an important issue that impairs the reliability of sensors, especially in gas sensors. The conventional method usually adopts the reference gas to compensate for the drift. However, its classification accuracy is not high. We propose a supervised learning algorithm that is based on multi-classifier integration for drift compensation in this paper, which incorporates drift compensation into the classification process, motivated by the fact that the goal of drift compensation is to improve the classification performance. In our method, with the obtained characteristics of sensors and the advantage of Support Vector Machine (SVM) in few-shot classification, the improved Long Shot Term Memory (LSTM) is integrated to build the multi-class classifier model. We tested the proposed approach on the publicly available time series dataset that was collected over three years by the metal-oxide gas sensors. The results clearly indicate the superiority of multiple classifier approach, which achieves higher classification accuracy as compared with different approaches during testing period with an ensemble of classifiers in the presence of sensor drift over time.

Image-to-Video Person Re-Identification With Temporally Memorized Similarity Learning

With the development of video surveillance in public safety field, there is an increasing research on person re-identification (re-id). In this paper, we address the image-to-video person re-id, in which the probe is an image and the gallery is consists of videos captured by nonoverlapping cameras. Compared with image, video sequence contains more temporal information that can be explored to improve the performance of re-identification system. However, it is challenging to model temporal information in the matching process of image-to-video person re-id. In this paper, we proposed a novel temporally memorized similarity learning neural network for this problem. In specific, the proposed network mainly consisted of two parts, including feature representation sub-network and similarity sub-network. In the first part, we adopted a convolutional neural network (CNN) to extract features from the input image. Given a video sequence of a person, features were first extracted from each its frame by using CNN and further forward to a long shot term memory (LSTM) network to encode the temporal information of video sequence. The outputs of LSTM were concatenated together as the feature vector of video sequences. Finally, the feature vectors of probe image and the video sequence were further forward to the similarity sub-network for distance metric learning. In the proposed framework, the feature representation and the similarity metric learning can be learned and optimized simultaneously. We evaluated the proposed framework on three public person re-id data sets, and the experimental results showed that the proposed approach is effective for the image-to-video person re-id.