摘要 :
The Tangni landslide in Chamoli, India, has experienced several landslide incidents in the recent past. Due to the fatalities and injuries caused, it is essential to predict slope movements at this site. A recent approach to predi...
展开
The Tangni landslide in Chamoli, India, has experienced several landslide incidents in the recent past. Due to the fatalities and injuries caused, it is essential to predict slope movements at this site. A recent approach to predicting slope movements is via machine-learning algorithms. In machine learning literature, recurrent neural networks (simple LSTMs, stacked LSTMs, bidirectional LSTMs, convolutional LSTMs, CNN-LSTMs, and encoder-decoder LSTMs) and non-recurrent neural networks (multilayer perceptrons) have been proposed. However, evaluating recurrent and non-recurrent neural networks for real-world slope movements prediction has been less explored. This research's primary objective is to develop and evaluate novel recurrent and non-recurrent neural network algorithms in their ability to predict slope movements. We used two years' weekly data of slope movements from the Tangni landslide site in Chamoli, India. Different recurrent and non-recurrent neural networks were calibrated on the training data and then predicted the test data. Different hyperparameters (epochs; packet shuffle; look-back period; the number of nodes per layer; and the number of layers) were calibrated to training data. Later, the developed models were evaluated on test data. Results revealed that, during training, the recurrent stacked LSTMs and bidirectional LSTMs performed the best and second-best, respectively, compared to other recurrent and non-recurrent neural networks. However, during the test, the recurrent CNN-LSTMs and simple LSTMs performed best and second best, respectively, compared to other recurrent and non-recurrent neural networks. We discuss the implications of our results for predicting slope movements at real-world landslide sites.
收起
摘要 :
The literature argues that an accurate price prediction of agricultural goods is quintessential to ensure the economy's good functioning worldwide. Research reveals that studies with the application of deep learning in the tasks o...
展开
The literature argues that an accurate price prediction of agricultural goods is quintessential to ensure the economy's good functioning worldwide. Research reveals that studies with the application of deep learning in the tasks of agricultural price forecast on short historical agricultural price data are very scarce. The gap mentioned above is removed in this study by employing five versions of LSTM deep learning techniques for five agricultural commodities prices prediction on a univariate time series dataset of rice, wheat, gram, banana, and groundnut spanning January 2000 to July 2020. The study obtained good forecasting results for all five commodities employing the five LSTM models. The study validated the results with lower values of error metrics, MAE, MAPE, MSE, and RMSE, and two paired t-tests with hypothesis and confidence levels of 95% as a measure of robustness. The study predicted one month ahead future price and compared it with actual prices using LSTM models.
收起
摘要 :
Dialog state tracking (DST) maintains and updates dialog states at each time step as the dialog progresses. It is necessary to include dialog historical information in DST. Previous word-based DST models took historical utterances...
展开
Dialog state tracking (DST) maintains and updates dialog states at each time step as the dialog progresses. It is necessary to include dialog historical information in DST. Previous word-based DST models took historical utterances as a word sequence and used n-grams in the sequence as inputs of models. It suffered from the problem of data sparseness. This paper proposes a cascaded deep neural network framework for DST. It alleviates the problem of data sparseness by making use of the hierarchical structure in dialog. The bottom layer of the cascaded framework, implemented by an Long Short Term Memory (LSTM) or a Convolutional Neural Network (CNN), encodes the word sequence into a sentence embedding in each dialog turn, and the upper layer integrates the representation of each turn gradually to get the dialog state using an LSTM. The cascaded models integrate natural language understanding into DST, and the entire network is trained as a whole. The experimental results on the DSTC2 dataset indicate that the proposed models, LSTM+LSTM and CNN+LSTM, can achieve better performance than existing models.
收起
摘要 :
Studies have revealed that the failure rates of storage devices can often be as high as fourteen percent. To make matters worse, there are frequently no warning signs for precaution before catastrophic failure of storage devices o...
展开
Studies have revealed that the failure rates of storage devices can often be as high as fourteen percent. To make matters worse, there are frequently no warning signs for precaution before catastrophic failure of storage devices occurs. A real-time predictive maintenance system that provides an automatic means for predicting when maintenance should be performed to ultimately eliminate unexpected breakdowns needs to be developed. Unlike traditional regression predictive modeling, the failure detection of storage devices is a problem of time series prediction, which adds the complexity of a sequence dependence among the input variables. The proposed LSTM (Long Short-Term Memory) network is a branch of RNN (Recurrent Neural Network) used in deep learning, which presents a very large architecture that can be successfully trained. LSTM is good at extracting patterns in input feature space, where the input data spans over long sequences. With the gated architecture of LSTM, it is capable of learning the context required to make predictions in time series forecasting. It is ideal for generating responses that depend on a time-evolving state; for example detecting the condition of storage devices over time. This paper describes our development of an LSTM (Long short-term memory), a special kind of RNN (Recurrent Neural Network)-based real-time predictive maintenance system (RPMS) built on top of Apache Spark for detecting storage device failure. By streaming real-time data into a RPMS directly from the device itself, the issues can be revealed and addressed early before they cause costly downtime.
收起
摘要 :
The agricultural industry has enormous potential to meet the world's growing need for nutritious and healthful food. However, pests destroy a significant portion of the harvest and reduce quality; thus, farmers struggle to discove...
展开
The agricultural industry has enormous potential to meet the world's growing need for nutritious and healthful food. However, pests destroy a significant portion of the harvest and reduce quality; thus, farmers struggle to discover pests on their farms. Traditional pest identification methods require scientists with extensive field experience correctly identify pests based on their physical characteristics. Pesticides harm food and agriculture. IoT technology uses many affordable sensor devices to collect real-time data on pest-related agricultural development characteristics. The research paper aims to develop an Al-enabled real-time, IoT-based automatic pest detection system using sound pest analytics and IoT networks in the sizeable agricultural field. The proposed method used audio pre-processing techniques in sound analytics such as HPF, Hann window, hop window, FFT, DFT, STFT, and MFCC algorithm for denoising the pest sound, removing the spectral leakage, converting overlapping to non-overlapping frames, converting the time to the frequency domain, determining frequency spectrum, detecting sinusoidal frequency and internal component, extracting the MFCCs feature respectively. The characteristics and other statistical metrics were collected from 500 pest sounds and trained, validated, and evaluated using the CNN, LSTM, Bi-LSTM, and CNN-Bi-LSTM network models. The extracting features and different statistical measurements were compared during the testing process. The proposed system integrated the CNN and Bi-LSTM techniques called CNN-Bi-LSTM model for training, validating, and testing. From the experimental results, it has been observed that, the proposed CNN-Bi-LSTM model achieved 98.91 % accuracy, 96.8 % sensitivity, 97.96 % specificity, 98.54 % recall, 98.63 % precision, and 98.58 % F1 score, which is better compared to CNN, LSTM, Bi-LSTM and also existing ANN, DCNN, and VGG-16 state-of-the-art techniques.
收起
摘要 :
Accurate predicting the air quality trend can provide a theoretical basis for environmental protection manage-ment and decision-making. This study proposed the convolutional neural networks-long short-term memory (CNN-LSTM) model,...
展开
Accurate predicting the air quality trend can provide a theoretical basis for environmental protection manage-ment and decision-making. This study proposed the convolutional neural networks-long short-term memory (CNN-LSTM) model, which was proposed to improve the air quality prediction accuracy. Firstly, CNN's efficient feature extraction function was used to extract data features. Then the feature vectors were constructed into the sequence form, which was transmitted to the LSTM network. The LSTM layer learned the changing rules of air quality data to predict future data. Taking Beijing's air quality index as an example, the prediction results of the CNN-LSTM model were compared with those of auto-regressive moving average (ARMA), seasonal auto-regression integrated moving average (SARIMA), recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU) models. The results show that, compared with other single prediction models, the CNN-LSTM achieved the highest prediction accuracy. In particular, CNN-LSTM was compared with the SARIMA model, which is a time series representative model. The indicators of the CNN-LSTM model have been well improved. The mean ab-solute error (MAE) and root mean square error (RMSE) of the CNN-LSTM were reduced respectively 3.17% and 5.46%, and R2 was improved 8.45%.
收起
摘要 :
This research paper investigates the use of Long Short-Term Memory (LSTM) and Grid Search Algorithm (GSA)-LSTM methods to forecast PV power output in different horizons. The study proposes precise hyperparameters for the LSTM netw...
展开
This research paper investigates the use of Long Short-Term Memory (LSTM) and Grid Search Algorithm (GSA)-LSTM methods to forecast PV power output in different horizons. The study proposes precise hyperparameters for the LSTM network for improved performance prediction. Initially, LSTM was evaluated solely, and three different scenarios were investigated to obtain the best LSTM network hyperpa-rameter settings for various horizons using a sensitivity study. Afterwards, LSTM was coupled with GSA to optimize LSTM network hyperparameters, which enhanced prediction accuracy and minimized error. The proposed methodology identified the most effective values for each hyperparameter based on different forecasting horizons. Additionally, Spearman Correlation Coefficient (SCC) and Pearson Correlation Coefficient (PCC) were deployed to determine the relationship between input data and target as well as identifying optimum lag value which has substantial effect on LSTM network. Optimal lag values of 150 and 280 were determined following the strong correlation between the data and target. LSTM and GSA-LSTM methods were compared with each other at various horizons and it was demonstrated that GSA-LSTM is superior by improving MSE at peak points by 10%, 30%, and 34% for 12 hours, 3 days, and 2 weeks horizons respectively. Comparing the proposed methodology with other studies in the literature revealed that the current study is capable of predicting PV power output for the 1 hour ahead horizon with significantly higher accuracy and with maximum improvement of about 28%.
收起
摘要 :
We describe an architecture of Time-Varying Long Short-Term Memory recurrent neural networks (TV-LSTMs) for human action recognition. The main innovation of this architecture is the use of hybrid weights, shared weights and non-sh...
展开
We describe an architecture of Time-Varying Long Short-Term Memory recurrent neural networks (TV-LSTMs) for human action recognition. The main innovation of this architecture is the use of hybrid weights, shared weights and non-shared weights which we refer to as varying weights. The varying weights can enhance the ability of LSTMs to represent videos and other sequential data. We evaluate TV-LSTMs on UCF-11, HMDB-51, and UCF-101 human action datasets and achieve the top-1 accuracy of 99.64%, 57.52%, and 85.06% respectively. This model performs competitively against the models that use both RGB and other features, such as optical flows, improved Dense Trajectory, etc. In this paper, we also propose and analyze the methods of selecting varying weights.
收起
摘要 :
Fifth generation (5G) wireless networks' system performance is dependent on having perfect knowledge of the channel state information (CSI). Deep learning (DL) has helped improve both the end-to-end reliability of 5G and beyond fi...
展开
Fifth generation (5G) wireless networks' system performance is dependent on having perfect knowledge of the channel state information (CSI). Deep learning (DL) has helped improve both the end-to-end reliability of 5G and beyond fifth generation (B5G) networks and the computational complexity of these networks. This work uses the Bi-linear long short-term memory (Bi-LSTM) scheme to examine the overall performance of the 5G orthogonal frequency division multiplexing (OFDM) technology. The least squares (LS) channel estimation scheme is a famous scheme employed to estimate the fading channel coefficients due to their lower complexity without the prior CSI. However, this scheme has an exceedingly high CSI error. Using pilot symbols (PS) and loss functions, this work has proposed the Bi-LSTM 5G OFDM estimators to improve the channel estimation obtained by the LS approach. All simulation analysis uses convex optimization (CO) software (CVX software) and stochastic gradient descent (SGD). When combined with many PS (72) and a cross-entropy loss function, the proposed Bi-LSTM outperforms the long-short-term memory (LSTM) cross-entropy, LS, and minimum mean square error (MMSE) estimators in low, medium, and high signal-to-noise ratio (SNR) regimes. The computational and training times of Bi-LSTM and LSTM DL estimators are also compared. Because of its DNN design, it can evaluate massive datasets, find hidden statistical patterns and characteristics, establish underlying relationships, and transfer what it has learnt to other contexts. Statistical analysis of the bit error rate (BER) reveals that Bi-LSTM outperforms the MMSE in terms of accurate channel prediction.
收起
原文获取