Explore the vast range of titles available.

Recurrent neural network (RNN) and convolutional neural network (CNN) are two prevailing architectures used in text classification. Traditional approaches combine the strengths of these two networks by straightly streamlining them or linking features extracted from them. In this article, a novel approach is proposed to maintain the strengths of RNN and CNN to a great extent. In the proposed approach, a bi-directional RNN encodes each word into forward and backward hidden states. Then, a neural tensor layer is used to fuse bi-directional hidden states to get word representations. Meanwhile, a convolutional neural network is utilized to learn the importance of each word for text classification. Empirical experiments are conducted on several datasets for text classification. The superior performance of the proposed approach confirms its effectiveness.

Various deep learning techniques have recently been developed in many fields due to the rapid advancement of technology and computing power. These techniques have been widely applied in finance for stock market prediction, portfolio optimization, risk management, and trading strategies. Forecasting stock indices with noisy data is a complex and challenging task, but it plays an important role in the appropriate timing of buying or selling stocks, which is one of the most popular and valuable areas in finance. In this work, we propose novel hybrid models for forecasting the one-time-step and multi-time-step close prices of DAX, DOW, and S&P500 indices by utilizing recurrent neural network (RNN)–based models; convolutional neural network-long short-term memory (CNN-LSTM), gated recurrent unit (GRU)-CNN, and ensemble models. We propose the averaging of the high and low prices of stock market indices as a novel feature. The experimental results confirmed that our models outperformed the traditional machine-learning models in 48.1% and 40.7% of the cases in terms of the mean squared error (MSE) and mean absolute error (MAE), respectively, in the case of one-time-step forecasting and 81.5% of the cases in terms of the MSE and MAE in the case of multi-time-step forecasting.

Plant diseases have a significant impact on global food security and the world's agricultural economy. Their early detection and classification increase the chances of setting up effective control measures, which is why the search for automatic systems that allow this is of major interest to our society. Several recent studies have reported promising results in the classification of plant diseases from RGB images on the basis of Convolutional Neural Networks (CNN). These studies have been successfully experimented on a large number of crops and symptoms, and they have shown significant advantages in the support of human expertise. However, the CNN models still have limitations. In particular, CNN models do not necessarily focus on the visible parts affected by a plant disease to allow their classification, and they can sometimes take into account irrelevant backgrounds or healthy plant parts. In this paper, we therefore develop a new technique based on a Recurrent Neural Network (RNN) to automatically locate infected regions and extract relevant features for disease classification. We show experimentally that our RNN-based approach is more robust and has a greater ability to generalize to unseen infected crop species as well as to different plant disease domain images compared to classical CNN approaches. We also analyze the focus of attention as learned by our RNN and show that our approach is capable of accurately locating infectious diseases in plants. Our approach, which has been tested on a large number of plant species, should thus contribute to the development of more effective means of detecting and classifying crop pathogens in the near future.

Researching public opinion can help us learn important facts. People may quickly and easily express their thoughts and feelings on any subject using social media, which creates a deluge of unorganized data. Sentiment analysis on social media platforms like Twitter and Facebook has developed into a potent tool for gathering insights into users' perspectives. However, difficulties in interpreting natural language limit the effectiveness and precision of sentiment analysis. This research focuses on developing a social media sentiment analysis (SMSA) framework, incorporating a custom-built emotion thesaurus to enhance the precision of sentiment analysis. It delves into the efficacy of various deep learning algorithms, under different parameter calibrations, for sentiment extraction from social media. The study distinguishes itself by its unique approach towards sentiment dictionary creation and its application to deep learning models. It contributes new insights into sentiment analysis, particularly in social media contexts, showcasing notable advancements over previous methodologies. The results demonstrate improved accuracy and deeper understanding of social media sentiment, opening avenues for future research and applications in diverse fields.

Agriculture is the most important sector and the backbone of a developing country’s economy. Accurate crop yield prediction models can provide decision-making tools for farmers to make better decisions. Crop yield prediction has challenged researchers due to dynamic, noisy, non-stationary, non-linear features and complex data. The factors that influence crop yield are changes in temperature and rainfall, plant disease, pests, fertilizer, and soil quality. The paper discusses the factors affecting crop yield, explores the features utilized, and analysis deep learning methodologies and performance metrics utilized in crop yield prediction.

Recently, due to the advance and impressive results of deep learning techniques in the fields of image recognition, natural language processing and speech recognition for various long-standing artificial intelligence (AI) tasks, there has been a great interest in applying towards security tasks too. This article focuses on applying these deep taxonomy techniques to network intrusion detection system (N-IDS) with the aim to enhance the performance in classifying the network connections as either good or bad. To substantiate this to NIDS, this article models network traffic as a time series data, specifically transmission control protocol / internet protocol (TCP/IP) packets in a predefined time-window with a supervised deep learning methods such as recurrent neural network (RNN), identity matrix of initialized values typically termed as identity recurrent neural network (IRNN), long short-term memory (LSTM), clock-work RNN (CWRNN) and gated recurrent unit (GRU), utilizing connection records of KDDCup-99 challenge data set. The main interest is given to evaluate the performance of RNN over newly introduced method such as LSTM and IRNN to alleviate the vanishing and exploding gradient problem in memorizing the long-term dependencies. The efficient network architecture for all deep models is chosen based on comparing the performance of various network topologies and network parameters. The experiments of such chosen efficient configurations of deep models were run up to 1,000 epochs by varying learning-rates between 0.01-05. The observed results of IRNN are relatively close to the performance of LSTM on KDDCup-99 NIDS data set. In addition to KDDCup-99, the effectiveness of deep model architectures are evaluated on refined version of KDDCup-99: NSL-KDD and most recent one, UNSW-NB15 NIDS datasets.

Text is a significant tool for human communication, and text recognition in scene images becomes more and more important. In this paper, we propose a residual convolutional recurrent neural network for solving the task of scene text recognition. The general convolutional recurrent neural network (CRNN) is realized by combining convolutional neural network (CNN) with recurrent neural network (RNN). The CNN part extracts features and the RNN part encodes and decodes feature sequences. In order to improve the accuracy rate of scene text recognition based on CRNN, we explore different deeper CNN architectures to get feature descriptors and analyze the corresponding text recognition results. Specifically, VGG and ResNet are introduced to train these different deep models and obtain the encoding information of images. The experimental results on public datasets demonstrate the effectiveness of our method.

Heterogeneity is common in parallel and distributed environments used for extensive computations such as MapReduce. Stragglers are the tasks that are running on inferior performing nodes in the cluster. Early detection of stragglers is always challenging in such environments. In the previously proposed approaches, late detection of straggler tasks and estimation of time to end (TTE) for all the tasks running in a heterogeneous environment delays the entire job execution. Early straggler detection help to speculate a task at the early stages of task execution which indirectly improves the complete job execution. This article proposed early straggler detection by a recurrent neural network (ESDRNN) that collects the task and node information every three seconds from ApplicationMaster to train the RNN. It classifies the straggler tasks pretty early by RNN, between thirty to forty seconds of task execution, and transfers a list of classified tasks to an agent running on ResourceManager. RNN is a type of artificial neural network that is prevalent for processing sequential time-series data. Then, the agent predicts the TTE of these classified tasks by the Autoregressive integrated moving average (ARIMA) model. Finally, it sorts and refreshes the list with higher TTE after every ten seconds and speculates the tasks for the early completion of the MapReduce job. This proposed technique’s performance is evaluated on the HiBench benchmark suite of Hadoop’s most popular benchmark. Finally, compared with the default speculation technique and different techniques, the proposed speculation technique detects the stragglers early within 35 to 40 seconds of task execution. As a result, it decreases the job execution time by an average of 21% to 38% significantly for different workloads in a heterogeneous Hadoop cluster.

Rapid technological changes and disruptive innovations have resulted in a significant shift in people’s behavior and requirements. Electronic gadgets, including smartphones, notebooks, and other devices, are indispensable to everyday routines. Consequently, the demand for high-capacity batteries has surged, which has enabled extended device autonomy. An alternative approach to address this demand is battery swapping, which can potentially extend the battery life of electronic devices. Although battery sharing in electric vehicles has been well studied, smartphone applications still need to be explored. Crucially, assessing the batteries’ state of health (SoH) presents a challenge, necessitating consensus on the best estimation methods to develop effective battery swap strategies. This paper proposes a model for estimating the SoH curve of lithium-ion batteries using the state of charge curve. The model was designed for smartphone battery swap applications utilizing Gated Recurrent Unit (GRU) neural networks. To validate the model, a system was developed to conduct destructive tests on batteries and study their behavior over their lifetimes. The results demonstrated the high precision of the model in estimating the SoH of batteries under various charge and discharge parameters. The proposed approach exhibits low computational complexity, low cost, and easily measurable input parameters, making it an attractive solution for smartphone battery swap applications.

Plasticity theory aims at describing the yield loci and work hardening of a material under general deformation states. Most of its complexity arises from the nontrivial dependence of the yield loci on the complete strain history of a material and its microstructure. This motivated 3 ingenious simplifications that underpinned a century of developments in this field: 1) yield criteria describing yield loci location; 2) associative or nonassociative flow rules defining the direction of plastic flow; and 3) effective stress–strain laws consistent with the plastic work equivalence principle. However, 2 key complications arise from these simplifications. First, finding equations that describe these 3 assumptions for materials with complex microstructures is not trivial. Second, yield surface evolution needs to be traced iteratively, i.e., through a return mapping algorithm. Here, we show that these assumptions are not needed in the context of sequence learning when using recurrent neural networks, diverting the above-mentioned complications. This work offers an alternative to currently established plasticity formulations by providing the foundations for finding history- and microstructure-dependent constitutive models through deep learning.