Input Sequence Research Articles

Deep spatio-temporal feature fusion learning for multi-step building cooling load forecasting

Accurate forecasting of building cooling load is essential for optimizing control strategies and energy efficiency in heating, ventilation, and air conditioning (HVAC) systems. However, existing machine learning and deep learning algorithms encounter challenges in addressing the dynamics and uncertain nature of cooling load demand, as they solely focus on temporal analysis. To tackle this issue, the paper proposes CRG-Informer, a deep learning approach that integrates spatio-temporal features to enhance multi-step predictive performance. Firstly, a time series spatial control relationship graph set is constructed based on the interconnected structure, static attributes, and control characteristics of the cooling system. Then, a graph neural network (GNN) is designed for graph embedding to extract spatial correlation information at each instance. Additionally, the Informer model with the ProbSparse self-attention mechanism is utilized to capture long-term temporal dependencies among input sequences. This integration of components enables feature fusion, and the proposed approach is evaluated using data from an office building in Hangzhou, China, over an entire cooling season. Extensive experiments were conducted to compare our method with five superior baseline models for 6, 12, 24, 72, and 144-step cooling load forecasting across four types of data acquisition intervals. The results demonstrate that the CRG-Informer, through spatio-temporal feature fusion, outperforms existing methods and provides a novel solution for improving cooling load forecasting performance.

Enhancing source code classification effectiveness via prompt learning incorporating knowledge features

Researchers have investigated the potential of leveraging pre-trained language models, such as CodeBERT, to enhance source code-related tasks. Previous methodologies have relied on CodeBERT’s ‘[CLS]’ token as the embedding representation of input sequences for task performance, necessitating additional neural network layers to enhance feature representation, which in turn increases computational expenses. These approaches have also failed to fully leverage the comprehensive knowledge inherent within the source code and its associated text, potentially limiting classification efficacy. We propose CodeClassPrompt, a text classification technique that harnesses prompt learning to extract rich knowledge associated with input sequences from pre-trained models, thereby eliminating the need for additional layers and lowering computational costs. By applying an attention mechanism, we synthesize multi-layered knowledge into task-specific features, enhancing classification accuracy. Our comprehensive experimentation across four distinct source code-related tasks reveals that CodeClassPrompt achieves competitive performance while significantly reducing computational overhead.

Combining Multiple Text Representations for Improved Automatic Evaluation of Indonesian Essay Answers

multiple-choices, regarding students’ learning achievement. When the number of students in a class is huge, however, examinations using essay questions become hard to conduct and take a long evaluation time. Automatic essay evaluation has, therefore, become a potential approach in this situation. Various methods have been proposed, however, optimal solutions for such evaluation in the Indonesian language are less known. Furthermore, with the rapid development of machine learning approaches, in particular deep learning approaches, the investigation of such optimal solutions becomes more necessary. Method: To address the aforementioned issue, this study proposed the investigation of text representation approaches for optimal automatic evaluation of Indonesian essay answers. The investigation compared pre-trained word embedding methods such as Word2vec, GloVe, FastText, and RoBERTa, as well as compared text encoding methods such as long short-term memories (LSTMs) and transformers. LSTMs are able to capture temporal semantics by employing state variables, while transformers are able to capture long-term dependency between parts of their input sequences. Additionally, we investigated classification-based and similarity-based training to build text encoders. We expected that these training approaches allowed encoders to extract different views of information. We compared classification results produced by different text encoders and combinations of text encoders. Result: We evaluated various text representation approaches using the UKARA dataset. Our experiments showed that the FastText word embedding method outperformed the Word2vec, GloVe, and RoBERTa methods. The FastText method achieved an F1-score of 75.43% on validation sets, while the Word2vec, GloVe, and RoBERTa methods achieved F1-scores of 69.56%, 74.53%, and 72.87%, respectively. In addition, the experiments showed that combinations of text encoders outperformed individual encoders. The combination of the LSTM encoder, the transformer encoder, and the TF-IDF encoder obtained an F1-score of 77.22% in the best case, which is better than the best F1-scores of the individual LSTM encoders (75.35%), the best combination of transformer encoders (71.49%), and the individual TF-IDF encoder (76.69%). We observed that LSTM encoders produced better performance when they were built using classification-based training. Meanwhile, the transformer encoders obtained better performance when built using similarity-based training. Novelty: The novelty proposed in this research is the optimal combination of text encoders specifically constructed for the evaluation of essay answers in the Indonesian language. Our experiments showed that the combination of three encoders - namely the LSTM encoder built using classification-based training, the transformer encoder built using classification-based and similarity-based training, and the TF-IDF encoder - obtained the best classification performance.

Multi-Scale Spatio-Temporal Attention Networks for Network-Scale Traffic Learning and Forecasting.

Accurate and timely forecasting of traffic on local road networks is crucial for deploying effective dynamic traffic control, advanced route planning, and navigation services. This task is particularly challenging due to complex spatio-temporal dependencies arising from non-Euclidean spatial relations in road networks and non-linear temporal dynamics influenced by changing road conditions. This paper introduces the spatio-temporal network embedding (STNE) model, a novel deep learning framework tailored for learning and forecasting graph-structured traffic data over extended input sequences. Unlike traditional convolutional neural networks (CNNs), the model employs graph convolutional networks (GCNs) to capture the spatial characteristics of local road network topologies. Moreover, the segmentation of very long input traffic data into multiple sub-sequences, based on significant temporal properties such as closeness, periodicity, and trend, is performed. Multi-dimensional long short-term memory neural networks (MDLSTM) are utilized to flexibly access multi-dimensional context. Experimental results demonstrate that the STNE model surpasses state-of-the-art traffic forecasting benchmarks on two large-scale real-world traffic datasets.

Cooperative control of virtually coupled trains considering train-to-train communication cycle: a tube-based model predictive control approach

In this paper, a distributed tube-based model predictive control (TMPC) method is proposed to address the cooperative control of virtually coupled high-speed train (HSTs), which consists of an optimisation control part and a feedback control part. The novelty of this work lies in its pioneering consideration of the inconsistency between the train-to-train (T2T) communication cycle and the train operation control cycle in the cooperative control of virtually coupled trains. Firstly, for the optimisation control part, a distributed MPC scheme is designed based on the nominal train dynamics model to address the cooperative control of trains with different initial states, state constraints and control input constraints. By solving the optimisation control problem of the MPC, the optimal control input sequence and operating state sequence of the train for the next T2T communication cycle can be obtained. Secondly, the recursive feasibility and convergence of the MPC are analysed theoretically. Thirdly, for the feedback control part, an adaptive feedback control law is designed based on the actual train dynamics model, auxiliary dynamic system (ADS) and neural network techniques. In the train operation control cycle, the feedback control part utilises the optimal state sequence from the MPC as a reference for feedback control to deal with the external disturbance and train dynamics modelling inaccuracy. Fourthly, it is theoretically proved that the actual operating state of the train can converge to a small region around the optimal state of the MPC in a finite time. Finally, the effectiveness of the proposed distributed cooperative control scheme is verified by simulations.

Construction of HaloTag-based macromolecular probe for multiple logic gates and photoactivatable bioimaging

Protein bioconjugation has emerged as one of the most valuable tools for the development of protein-based biochemical assays. Herein, we report a fluorescent macromolecular probe RF12_POI, in which the coumarin derivative RF12 is specifically conjugated onto the HaloTag fused protein of interest (POI) to achieve a dual stimuli-mediated fluorescence response. RF12 is first obtained by installing a photo-cleavable 1-ethyl-2-nitrobenzyl group onto the C7 hydroxy moiety of coumarin fluorophore with a HaloTag ligand attaching to the acid-labile 1,3-dioxane moiety. Upon stimulation, RF12_Halo exhibits a sequential fluorescence response to photon/H+ on both liquid and solid interfaces. Through the conjugation of RF12 onto the GFP_Halo protein, RF12_GFP_Halo presents a fluorescence resonance energy transfer (FRET) from photo-cleaved RF12 to GFP in the protein complex. Furthermore, by utilizing the stimuli-responsive fluorescence characteristics of coumarin derivatives RF12 (photon/H+) and RF16 (H2O2/H+), we construct RF12/RF16_POI based protein films and achieve multiple applications of logic circuits, including AND, OR, XOR, INHIBIT, Half-adder or Half-subtractor. In these circuits, the output value of I/I0 is dependent on the input sequence of photon, H2O2, and H+. Additionally, we evaluate the fluorescence labeling ability of RF12 to intracellular IRE1_Halo protein and demonstrate that RF12 containing the HaloTag ligand could be precisely retained in cells to track IRE1_Halo protein. Hence, we provide a unique structural design strategy to construct fluorescence dual-responsive macromolecules for information encryption and cellular protein visualization.

TIANA: transcription factors cooperativity inference analysis with neural attention

BackgroundGrowing evidence suggests that distal regulatory elements are essential for cellular function and states. The sequences within these distal elements, especially motifs for transcription factor binding, provide critical information about the underlying regulatory programs. However, cooperativities between transcription factors that recognize these motifs are nonlinear and multiplexed, rendering traditional modeling methods insufficient to capture the underlying mechanisms. Recent development of attention mechanism, which exhibit superior performance in capturing dependencies across input sequences, makes them well-suited to uncover and decipher intricate dependencies between regulatory elements.ResultWe present Transcription factors cooperativity Inference Analysis with Neural Attention (TIANA), a deep learning framework that focuses on interpretability. In this study, we demonstrated that TIANA could discover biologically relevant insights into co-occurring pairs of transcription factor motifs. Compared with existing tools, TIANA showed superior interpretability and robust performance in identifying putative transcription factor cooperativities from co-occurring motifs.ConclusionOur results suggest that TIANA can be an effective tool to decipher transcription factor cooperativities from distal sequence data. TIANA can be accessed through: https://github.com/rzzli/TIANA.

Deep Learning Regression with Sequences of Different Length: An Application for State of Health Trajectory Prediction and Remaining Useful Life Estimation in Lithium-Ion Batteries

This study presents methods to handle deep learning regressions with input and output sequences of different lengths. We discuss the Autoregressive one-step prediction framework and introduce an innovative one-time multi-step (OTMS) prediction approach, based on a custom loss function, that predicts all future steps in a single shot. The presented methodologies are then applied to simultaneously predict the State of Health (SoH) trajectory and estimate the Remaining Useful Life (RUL) of lithium-ion battery cells. Accurate estimates of SoH trajectory and RUL are essential for Battery Management Systems (BMSs), electronic systems that guarantee safety while maximizing performance and extending battery lifespan. In this context, the studied methodologies were compared using a rigorous cross-validation approach. The OTMS model showed better predictions in early cycles, while the Autoregressive model performed better in later cycles, suggesting a hybrid approach between these two methodologies as an optimal solution.

Decoupled Temperature–Pressure Sensing System for Deep Learning Assisted Human–Machine Interaction

AbstractWith the rapid development of intelligent wearable technology, multimodal tactile sensors capable of data acquisition, decoupling of intermixed signals, and information processing have attracted increasing attention. Herein, a decoupled temperature–pressure dual‐mode sensor is developed based on single‐walled carbon nanotubes (SWCNT) and poly(3,4‐ethylenedioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) decorated porous melamine foam (MF), integrating with a deep learning algorithm to obtain a multimodal input terminal. Importantly, the synergistic effect of PEDOT:PSS and SWCNT facilitates the sensor with ideal decoupling capability and sensitivity toward both temperature (38.2 µV K−1) and pressure (10.8% kPa−1) based on the thermoelectric and piezoresistive effects, respectively. Besides, the low thermal conductivity and excellent compressibility of MF also endow it with the merits of a low‐temperature detection limit (0.03 K), fast pressure response (120 ms), and long‐term stability. Benefiting from the outstanding sensing characteristics, the assembled sensor array showcases good capacity for identifying spatial distribution of temperature and pressure signals. With the assistance of a deep learning algorithm, it displays high recognition accuracy of 99% and 98% corresponding to “touch” and “press” actions, respectively, and realizes the encrypted transmission of information and accurate identification of random input sequences, providing a promising strategy for the design of high‐accuracy multimodal sensing platform in human–machine interaction.

A label-free battery state of health estimation method based on adversarial multi-domain adaptation network and relaxation voltage

The state of health (SOH) estimation of lithium-ion batteries is crucial for the operational reliability and safety of electric vehicles. However, traditional data-driven methods face problems of label shortage and domain discrepancy caused by diverse battery types and operating conditions. This paper proposes a label-free SOH estimation method based on adversarial multi-domain adaptation technique and relaxation voltage (RV). Firstly, the raw RV and integral voltage are proposed to construct the two-dimensional input sequence to ensure high adaptability across various target domains. Secondly, a two-dimensional convolutional neural network integrated with fully connected layers is developed to establish the feature extractor and SOH estimator, which paves the way for effective knowledge transfer. Furthermore, an adversarial multi-domain adaptation method with a domain discriminator is developed to optimize the extraction of domain-invariant features and enable accurate SOH estimation without SOH labels. Datasets of 50 batteries with different temperatures and materials are used for the validation. The proposed method shows outstanding performance, achieving the average RMSE and MAE of 0.0274 and 0.0240 across various working temperatures, and 0.0177 and 0.0148 across different battery types. It suggests that the proposed method can achieve robust adaptive battery SOH estimation across multiple target domains without using SOH labels.

Benchmarking DNA Foundation Models for Genomic Sequence Classification.

The rapid advancement of DNA foundation language models has revolutionized the field of genomics, enabling the decoding of complex patterns and regulatory mechanisms within DNA sequences. However, the current evaluation of these models often relies on fine-tuning and limited datasets, which introduces biases and limits the assessment of their true potential. Here, we present a benchmarking study of three recent DNA foundation language models, including DNABERT-2, Nucleotide Transformer version-2 (NT-v2), and HyenaDNA, focusing on the quality of their zero-shot embeddings across a diverse range of genomic tasks and species through analyses of 57 real datasets. We found that DNABERT-2 exhibits the most consistent performance across human genome-related tasks, while NT-v2 excels in epigenetic modification detection. HyenaDNA stands out for its exceptional runtime scalability and ability to handle long input sequences. Importantly, we demonstrate that using mean token embedding consistently improves the performance of all three models compared to the default setting of sentence-level summary token embedding, with average AUC improvements ranging from 4.3% to 9.7% for different DNA foundation models. Furthermore, the performance differences between these models are significantly reduced when using mean token embedding. Our findings provide a framework for selecting and optimizing DNA language models, guiding researchers in applying these tools effectively in genomic studies.

Average Voltage Prediction of Battery Electrode Materials Using Transformer Deep Learning Architecture

Batteries are ubiquitous, with their presence ranging from electric vehicles to portable electronics. Research focused on increasing average voltage, improving stability, and extending cycle longevity of batteries is pivotal for the advancement of battery technology. These advancements can be accelerated through research into battery chemistries. The traditional approach, which examines each material combination individually, poses significant challenges in terms of resources and financial investment. Physics-based simulations, while detailed, are both time-consuming and resource-intensive. Researchers are aiming to mitigate these concerns by employing Machine Learning (ML) techniques. We propose the use of a Transformer ML model for predicting the average voltage of battery electrode materials. A Transformer is a Deep Learning (DL) architecture based on the attention mechanism; it dynamically focuses on different parts of the input sequence, thus capturing complex relationships and dependencies among input features. The absence of recurrent units allows the model to require less training time compared to other DL models. We have adapted the Transformer model for tabular data and regression tasks. Our dataset was sourced from the Materials Project Database. To ensure the model's compatibility with our voltage prediction regression task, we successfully trained and tested it using four features across 99 samples. Our ongoing research focuses on incorporating more relevant features and expanding the training and testing of the model to the entire dataset. Our work demonstrates the effectiveness of ML, especially the Transformer model, in predicting the average voltage of battery electrode materials.

Human pose estimation in complex background videos via Transformer-based multi-scale feature integration

Human posture estimation is still a hot research topic. Previous algorithms based on traditional machine learning have difficulties in feature extraction and low fusion efficiency. To address these problems, we proposed a Transformer-based method. We combined three techniques, namely the Transformer-based feature extraction module, the multi-scale feature fusion module, and the occlusion processing mechanism, to capture the human pose. The Transformer-based feature extraction module uses the self-attention mechanism to extract key features from the input sequence, the multi-scale feature fusion module fuses feature information of different scales to enhance the perception ability of the model, and the occlusion processing mechanism can effectively handle occlusion in the data and effectively remove background interference. Our method has shown excellent performance through verification on the standard dataset Human3.6M and the wild video dataset, achieving accurate pose prediction in both complex actions and challenging samples.

Maptcha: an efficient parallel workflow for hybrid genome scaffolding

BackgroundGenome assembly, which involves reconstructing a target genome, relies on scaffolding methods to organize and link partially assembled fragments. The rapid evolution of long read sequencing technologies toward more accurate long reads, coupled with the continued use of short read technologies, has created a unique need for hybrid assembly workflows. The construction of accurate genomic scaffolds in hybrid workflows is complicated due to scale, sequencing technology diversity (e.g., short vs. long reads, contigs or partial assemblies), and repetitive regions within a target genome.ResultsIn this paper, we present a new parallel workflow for hybrid genome scaffolding that would allow combining pre-constructed partial assemblies with newly sequenced long reads toward an improved assembly. More specifically, the workflow, called Maptcha, is aimed at generating long scaffolds of a target genome, from two sets of input sequences—an already constructed partial assembly of contigs, and a set of newly sequenced long reads. Our scaffolding approach internally uses an alignment-free mapping step to build a ⟨\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\langle $$\\end{document}contig,contig⟩\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\rangle $$\\end{document} graph using long reads as linking information. Subsequently, this graph is used to generate scaffolds. We present and evaluate a graph-theoretic “wiring” heuristic to perform this scaffolding step. To enable efficient workload management in a parallel setting, we use a batching technique that partitions the scaffolding tasks so that the more expensive alignment-based assembly step at the end can be efficiently parallelized. This step also allows the use of any standalone assembler for generating the final scaffolds.ConclusionsOur experiments with Maptcha on a variety of input genomes, and comparison against two state-of-the-art hybrid scaffolders demonstrate that Maptcha is able to generate longer and more accurate scaffolds substantially faster. In almost all cases, the scaffolds produced by Maptcha are at least an order of magnitude longer (in some cases two orders) than the scaffolds produced by state-of-the-art tools. Maptcha runs significantly faster too, reducing time-to-solution from hours to minutes for most input cases. We also performed a coverage experiment by varying the sequencing coverage depth for long reads, which demonstrated the potential of Maptcha to generate significantly longer scaffolds in low coverage settings (1×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$1\ imes $$\\end{document}–10×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$10\ imes $$\\end{document}).

PMANet: a time series forecasting model for Chinese stock price prediction

Forecasting stock movements is a crucial research endeavor in finance, aiding traders in making informed decisions for enhanced profitability. Utilizing actual stock prices and correlating factors from the Wind platform presents a potent yet intricate forecasting approach. While previous methodologies have explored this avenue, they encounter challenges including limited comprehension of interrelations among stock data elements, diminished accuracy in extensive series, and struggles with anomaly points. This paper introduces an advanced hybrid model for stock price prediction, termed PMANet. PMANet is founded on Multi-scale Timing Feature Attention, amalgamating Multi-scale Timing Feature Convolution and Ant Particle Swarm Optimization. The model elevates the understanding of dependencies and interrelations within stock data sequences through Probabilistic Positional Attention. Furthermore, the Encoder incorporates Multi-scale Timing Feature Convolution, augmenting the model's capacity to discern multi-scale and significant features while adeptly managing lengthy input sequences. Additionally, the model's proficiency in addressing anomaly points in stock sequences is enhanced by substituting the optimizer with Ant Particle Swarm Optimization. To ascertain the model’s efficacy and applicability, we conducted an empirical study using stocks from four pivotal industries in China. The experimental outcomes demonstrate that PMANet is both feasible and versatile in its predictive capability, yielding forecasts closely aligned with actual values, thereby fulfilling application requirements more effectively.

Predicting significant wave height in the South China Sea using the SAC-ConvLSTM model

IntroductionThe precise forecasting of Significant wave height(SWH) is vital to ensure the safety and efficiency of aquatic activities such as ocean engineering, shipping, and fishing.MethodsThis paper proposes a deep learning model named SAC-ConvLSTM to perform 24-hour prediction with the SWH in the South China Sea. The long-term prediction capability of the model is enhanced by using the attention mechanism and context vectors. The prediction ability of the model is evaluated by mean absolute error (MAE), root mean square error (RMSE), mean square error (MSE), and Pearson correlation coefficient (PCC).ResultsThe experimental results show that the optimal input sequence length for the model is 12. Starting from 12 hours, the SAC-ConvLSTM model consistently outperforms other models in predictive performance. For the 24-hour prediction, this model achieves RMSE, MAE, and PCC values of 0.2117 m, 0.1083 m, and 0.9630, respectively. In addition, the introduction of wind can improve the accuracy of wave prediction. The SAC-ConvLSTM model also has good prediction performance compared to the ConvLSTM model during extreme weather, especially in coastal areas.DiscussionThis paper presents a 24-hour prediction of SWH in the South China Sea. Through comparative validation, the SAC-ConvLSTM model outperforms other models. The inclusion of wind data enhances the model's predictive capability. This model also performs well under extreme weather conditions. In physical oceanography, variables related to SWH include not only wind but also other factors such as mean wave period and sea surface air pressure. In the future, additional variables can be incorporated to further improve the model's predictive performance.

GraphPro: An interpretable graph neural network-based model for identifying promoters in multiple species

Promoters are DNA sequences that bind with RNA polymerase to initiate transcription, regulating this process through interactions with transcription factors. Accurate identification of promoters is crucial for understanding gene expression regulation mechanisms and developing therapeutic approaches for various diseases. However, experimental techniques for promoter identification are often expensive, time-consuming, and inefficient, necessitating the development of accurate and efficient computational models for this task. Enhancing the model's ability to recognize promoters across multiple species and improving its interpretability pose significant challenges. In this study, we introduce a novel interpretable model based on graph neural networks, named GraphPro, for multi-species promoter identification. Initially, we encode the sequences using k-tuple nucleotide frequency pattern, dinucleotide physicochemical properties, and dna2vec. Subsequently, we construct two feature extraction modules based on convolutional neural networks and graph neural networks. These modules aim to extract specific motifs from the promoters, learn their dependencies, and capture the underlying structural features of the promoters, providing a more comprehensive representation. Finally, a fully connected neural network predicts whether the input sequence is a promoter. We conducted extensive experiments on promoter datasets from eight species, including Human, Mouse, and Escherichia coli. The experimental results show that the average Sn, Sp, Acc and MCC values of GraphPro are 0.9123, 0.9482, 0.8840 and 0.7984, respectively. Compared with previous promoter identification methods, GraphPro not only achieves better recognition accuracy on multiple species, but also outperforms all previous methods in cross-species prediction ability. Furthermore, by visualizing GraphPro's decision process and analyzing the sequences matching the transcription factor binding motifs captured by the model, we validate its significant advantages in biological interpretability. The source code for GraphPro is available at https://github.com/liuliwei1980/GraphPro.

EditableNeRF: Editing Topologically Varying Neural Radiance Fields by Key Points.

Neural radiance fields (NeRF) achieve highly photo-realistic novel-view synthesis, but it's a challenging problem to edit the scenes modeled by NeRF-based methods, especially for dynamic scenes. We propose editable neural radiance fields that enable end-users to easily edit dynamic scenes and support topological changes. Input with an image sequence from a single camera, our network is trained automatically and models topologically varying dynamics using our picked-out surface key points. Then end-users can edit the scene by easily dragging the key points to desired new positions. To achieve this, we propose a scene analysis method to detect and initialize key points by considering the dynamics in the scene, and a weighted key points strategy to model topologically varying dynamics by joint key points and weights optimization. Our method supports intuitive multi-dimensional (up to 3D) editing and can generate novel scenes that are unseen in the input sequence. Experiments demonstrate that our method achieves high-quality editing on various dynamic scenes and outperforms the state-of-the-art.

Sentiment analysis of movie review classifications using deep learning approaches

Movie reviews reflect how the public feels about a movie they have watched. However, because many reviews are posted on various websites, it is practically impossible to read each one. Summarizing all movie reviews can help people make informed decisions without reading through all of them. Previous studies have used different machine learning and deep learning techniques for sentiment analysis (SA), but few have combined comprehensive hyperparameter tuning and novel datasets for better performance. This paper presents an SA approach using deep learning models with optimized hyperparameters and a novel Rotten Tomatoes (RT) dataset to help viewers make better movie choices. SA, or opinion mining, is a computational technique to extract and analyze opinions and emotions expressed in text. We explore deep learning models such as Long Short-Term Memory (LSTM), XLNet, Convolutional Neural Networks-LSTM (CNN-LSTM), and Bidirectional Encoder Representations from Transformers (BERT). These models are known for capturing complex language patterns and context from raw text data. XLNet, a pre-trained model, effectively understands context by considering all possible permutations of the input sequence, BERT excels at using bidirectional context to understand text, LSTM retains information about long-term patterns in sequential data, and CNN-LSTM combines local and global context for reliable feature extraction. The RT dataset was pre-processed with data cleaning, spelling correction, lemmatization, and handling of informal words to improve the results. Our experiments show that XLNet performed better than other models on the Rotten Tomatoes dataset. The study demonstrates that SA of movie reviews provides insights into emotions and attitudes, allowing us to estimate a movie’s performance based on its overall sentiment.

Early Prognostics of Remaining Useful Life in Lithium Ion Batteries Using Hybrid LSTM-Att-MLP Model with Fusing Aging Information

Predicting the remaining useful lifetime (RUL) stands as a crucial aspect of lithium-ion battery management systems, acting as a core component of their functioning. Accurately predicting the RUL is essential for ensuring safety, preventing failures, and averting catastrophic incidents, but it is challenging, due to capacity degradation and aging effects. To overcome this, a hybrid model termed LAM: LSTM with an attention mechanism and MLP, for early RUL prediction, leveraging fused aging information is proposed. LSTM adeptly captures the significance embedded within feature sequences, preserving essential long-term features while effectively filtering out less pertinent information. Embedding attention mechanism with LSTM, the model dynamically focuses on different parts of the input sequence by assigning varying importance levels to different aging information, enhancing the prediction performance. For capturing the dynamic and nonlinear degradation trend of batteries and predicting RUL by effectively learning intricate degradation patterns MLP is utilized. The proposed model’s efficacy is evaluated using a NASA dataset through leave-one-out evaluation, utilizing 50% of the training data from three batteries to predict the others, and with varying starting points. The results indicate that under conditions of limited historical samples, the LAM attains higher accuracy and achieves minimum Mean Squared Error of 3.9962 × 10−5.