Bidirectional Model Research Articles

Parsimonious airfoil Parameterisation: A deep learning framework with Bidirectional LSTM and Gaussian Mixture models

The choice of airfoil parameterisation method significantly influences the overall wing optimisation performance by affecting the flexibility and computational efficiency of the process. Ideally, one should be able to intuitively constrain airfoil shape and structural characteristics as input to the optimisation process. Current parameterisation techniques lack the flexibility to generate airfoils efficiently by specifying parsimonious shape and structural features. To address this limitation, a deep learning framework is proposed, enabling conditional airfoil generation from an airfoil’s shape and structural feature definition.Specifically, we demonstrate the application of Bidirectional Long Short Term Memory models and Bayesian Gaussian Mixture models to derive airfoil coordinates from a compact set of shape and structural characteristics that we define. The proposed framework is shown to achieve favorable airfoil performance optimisation due to improved exploration and exploitation of the design space, compared to traditional approaches. Overall, the proposed optimisation framework is able to realise a 9.04% performance improvement over an airfoil design optimised with traditional parameterisation techniques.

Position-aware Interactive Attention Network for multi-intent spoken language understanding

Spoken Language Understanding (SLU) is a crucial component of task-oriented dialog systems. In recent years, there has been increasing attention on multi-intent SLU due to its relevance to complex real-world applications. Most existing joint models only utilize multi-intent information to guide the slot filling, with only a small number of models achieving bidirectional interaction. Additionally, unlike traditional single-intent SLU, multi-intent SLU is scope-dependent, where each intent in a sentence has its specific dependency range. However, current bidirectional joint models often employ a pipeline approach to implement interaction between the two sub-tasks, which fails to fully leverage the semantic information between them and can lead to error propagation. Moreover, attention-based multi-intent joint models do not adequately model the positional relationships between words, resulting in suboptimal overall performance. In this paper, we propose a novel multi-intent SLU model called Position-aware Interactive Attention Network (PIAN), which consists of interactive attention and rotary position embedding. The aim of PIAN is to fully exploit the correlation between the two sub-tasks and capturing the positional dependencies between words. This facilitates mutual guidance and enhancement between the two sub-tasks. Additionally, to address the issue of uncoordinated slot problem generated by traditional slot filling decoders, we employ an MCRF slot filling decoder to constrain the slot labels. We evaluate our model on two public multi-intent SLU datasets, and the experimental results demonstrate that our model achieves state-of-the-art performance on key metrics. Code for this paper is publicly available at https://github.com/puhahahahahaha/SLU_with_Co_PRoE.

Bidirectional Reflectance Distribution Function (BRDF) measurements, modeling and applications in space-agile Tx/Rx common aperture system

The Tx/Rx common aperture system not only needs to suppress out-of-field stray light, but also needs to suppress the influence of optical surface backscatter on detectivity. Backscatter generated by optical surfaces is often the dominant source of stray radiation that causes T/R isolation problems, and the black surfaces chosen for baffles also play a significant role in stray light suppression. Four commonly used optomechanical structure surfaces scattering BRDF data are measured, emulated and used. CFRP is selected as the baffle material, for the processing performance, cost performance and thermal control difficulty. A comparison is made between theoretical and experimental results for clean and particle contaminated ultra-smooth optical surface (CL= 800) scattering BRDF data. With the help of Mie theory and measured BRDF, the study improved the fitting accuracy of the Harvey-Shack model for the contaminated micro roughness surface. The Harvey-Shack model modeling coefficients of the particle contaminated micro-roughness surface are b = 0.12, l = 0.08, and s=-2.2. The scattering radiation introduced by particulates reduces the T/R isolation of the space-agile Tx/Rx common aperture system by 3.74 dB. Measurement results of particulates reducing the T/R isolation by 4.05 dB, verified the accuracy of particle contaminated surface BRDF modeling.

Inventory Prediction Using a Modified Multi-Dimensional Collaborative Wrapped Bi-Directional Long Short-Term Memory Model

Inventory prediction is concerned with the forecasting of future demand for products in order to optimize inventory levels and supply chain management. The challenges include demand volatility, data quality, multi-dimensional interactions, lead time variability, seasonal trends, and dynamic pricing. Nevertheless, these models suffer from numerous shortcomings, and in this research, we propose a new model, MMCW-BiLSTM (modified multi-dimensional collaboratively wrapped BiLSTM), for inventory prediction. The MMCW-BiLSTM model reflects a considerable leap in inventory forecasting by combining a number of components in order to consider intricate temporal dependencies and incorporate feature interactions. The MMCW-BiLSTM makes use of BiLSTM layers, collaborative attention mechanisms, and a multi-dimensional attention approach to learn from augmented datasets consisting of the original features and the extracted time series data. Moreover, adding a Taylor series transformation allows for a more precise description of the features in the model, thus improving the prediction precision. The results show that the models make the least mistakes when they use the AV demand forecasting dataset, with MAE values of 1.75, MAPE values of 2.89, MSE values of 6.76, and RMSE values of 2.6. Similarly, when utilizing the product demand dataset, the model also achieves the lowest error values for these metrics at 1.97, 3.91, 8.76, and 2.96. Likewise, when utilizing the dairy goods sales dataset, the model also achieves the lowest error values for these metrics at 2.54, 3.69, 10.39, and 3.22.

Enhance carbon emission prediction using bidirectional long short-term memory model based on text-based and data-driven multimodal information fusion

Accurate projections of carbon emissions are essential for climate change responses and environmental policy development. However, the complexity of carbon emission projections has increased due to various factors like technology and policy, posing challenges for accuracy. Many studies rely on time-series data alone, overlooking the combined impact of multiple information sources on carbon emissions. To overcome these bottlenecks, a hybrid model is proposed, integrating multimodal data, convolutional neural network inceptionv2 (CNN-inceptionv2), text convolutional neural network (TextCNN) optimized by attention mechanism, and bi-directional long and short-term memory (BiLSTM). Firstly, the time series feature set and text feature set are extracted using the CNN-inceptionv2 and the TextCNN, respectively. Subsequently, the feature sets of the two modal information are fused and outputted into a multimodal feature set, which are ultimately predicted using BiLSTM. The validity of the study model was also verified by different methods such as ablation experiments, single model and combined model. The coefficient of determination (R2) values were 0.986 for the validation set and 0.933 for the prediction set. In comparison to other models, the present model showed an average improvement of 16.6% in R2 values, fully indicating its superior predictive performance. The high point and interval predictions of carbon emissions further verified the superior prediction accuracy of the present model. The present model can accurately predict carbon emissions, providing data references for carbon emission decision makers. Additionally, it further enriches the methodological system of carbon emission prediction and aids in strategic planning for future carbon emission and energy management.

COMPARISON OF DEEP LEARNING TECHNIQUES FOR TEXTUAL SENTIMENT ANALYSIS

Examining the opinions of the people might provide us with valuable knowledge. Sentiment analysis is a method for analyzing textual data that helps find subjective information, such as opinions and feelings that individuals or groups have expressed. It improves our understanding of human language using deep learning and natural language techniques. Several deep learning models, including RNNs, LSTMs, GRUs, and their bidirectional variants, are compared in this work. Three publicly available datasets - the imdb_reviews, Twitter Sentiment Dataset, and Emotions dataset were used in the investigation. Accuracy performance is evaluated for six deep learning models. According to experimental studies, bidirectional structures outperform their unidirectional counterparts in most cases. Across several datasets, the bidirectional models continuously produced the best accuracy.

Optimizing gate control coordination signal for urban traffic network boundaries using multi-agent deep reinforcement learning

The macro-aggregated dynamic characteristics of urban network traffic flow encapsulated embodied in macroscopic fundamental diagram theory provide a concise perspective for network-level traffic control. However, accurately translating inter-regional macro-traffic flow control into fine-grained control at each intersection along the boundary remains challenging. To address this issue, a regional boundary intersections collaborative control method is proposed, leveraging the model-free data-driven advantages of multi-agent deep reinforcement learning. First, the proposed model develops a real-time communication module using a bi-directional long short-term memory model and an attention mechanism, enabling intersections to capture spatiotemporal features of the boundary traffic flow. Second, the feedback reward mechanism of the boundary intersection agent is meticulously constructed by considering the traffic state of the protected region, the boundary control pressure, and the traffic pressure of the intersection. The proposed model fosters cooperation at boundary intersections and delicately balances their macro-micro control contradictions based on mastering the boundary state information of the road network. Data-driven boundary gating control can fine-regulate the traffic flow distribution of the network from the perspective of macro-micro to promote its operation efficiency. The simulation experiments verify the effectiveness of the proposed multi-agent signal collaborative control method.

GRU-AE-wiener: A generative adversarial network assisted hybrid gated recurrent unit with Wiener model for bearing remaining useful life estimation

Bearings play a pivotal role in various mechanical systems, and their health directly impacts the reliability and safety of these systems. Consequently, extensive research has been dedicated to the estimation of Bearing Remaining Useful Life (RUL) through the lens of information fusion theory. The absence of comprehensive life-cycle degradation data for bearings, a common challenge within the information fusion domain, can hinder the accuracy and reliability of RUL prediction models. A novel hybrid data and model approach named GRU-AE-Wiener has been developed to address this limitation. This approach combines the power of Gated Recurrent Unit (GRU) and Wiener process models within the information fusion framework. Firstly, a Loop Generative Adversarial Network (Loop-GAN) is introduced to synthesize pseudo data to enhance the quality of synthetic data. Next, a bidirectional GRU model is structurally integrated with the Wiener process. In this design, the GRU model is configured in an Auto-Encoder-like structure, with the Wiener process serving as the hidden layer within this Auto-Encoder. Importantly, both the GRU and Wiener processes are jointly optimized with the assistance of Loop-GAN, emphasizing the collaborative nature of information fusion in this approach. The effectiveness of the proposed GRU-AE-Wiener is validated using the PHM 2012 dataset and XJTU-SY dataset. Experimental results underscore its superior RUL predictive performance compared to other deep learning models, highlighting the practical application of information fusion principles in bearing health assessment.

Boosting agricultural green development: Does socialized service matter?

Agricultural socialized service is gradually emerging as a new stimulus for enhancing the agricultural production environment. However, their precise impact on improving the agricultural ecological environment and promoting the green development of agriculture remains incompletely understood. Therefore, leveraging panel data spanning from 2003 to 2020 across 31 provinces in China, this study utilizes the bidirectional fixed effect model, moderating effect model, and spatial Durbin model to systematically assess the influence of agricultural socialized services on agricultural green development and its spatial ramifications. The findings show that (I) agricultural socialized services significantly contribute to promoting agricultural green development, particularly in regions with lower aging demographics. (II) The application of the spatial Durbin model reveals that this promotional effect does not exhibit significant spatial spillover effect. (III) The role of agricultural socialized services in fostering agricultural green development can be significantly enhanced by advancements in land transfer, agricultural technological innovations, and the improvement of rural human capital. In conclusion, the study provides a set of policy recommendations that include government financial support, facilitating land transfer, improving rural education and technical training, and promoting green production technologies to effectively promote agricultural green development.

Entrepreneurship Teaching : Design and Activities of 21st Century Learning for Economic Class under Mini-Project (Participatory-Phenomenological Design)

This study aims to explain the process of teaching entrepreneurship education courses in economics classes under a detailed mini-project. This research used Participatory action research (PAR) under qualitative approach. The sample size was only taken purposefully from one semester's three economics classes. This research used observation sheets in the form of learning activities and interviews. All data collected were analyzed descriptively with the step of reducing, displaying and drawing conclusion. The research results showed the effectiveness of a bidirectional teaching model in entrepreneurship education. This model, which combines theoretical instruction with practical application, significantly enhances student engagement and learning outcomes. The bidirectional approach allows students to relate theoretical concepts to real-world examples, fostering a deeper understanding and active participation. An essential element of this teaching model is the inclusion of mini-projects. These projects enable students to apply theoretical knowledge in practical scenarios, such as developing business plans and analyzing recruitment strategies. The hands-on experience gained through mini-projects not only solidifies students' grasp of the material but also hones their problem-solving and critical-thinking skills. The study's findings underscore the superiority of the bidirectional teaching model over traditional methods. The experimental class, which utilized this approach, showed a 93% success rate in meeting various evaluation criteria, compared to just 30% in the control class. This stark contrast highlights the importance of integrating mini-projects and practical activities into the curriculum to enhance educational outcomes. The bidirectional teaching model, complemented by mini-projects, proves to be a highly effective approach in entrepreneurship education. It not only engages students more deeply but also equips them with practical skills and knowledge that are essential for their future careers.

Evaluation of fiscal policy with text mining under "dual carbon" target in China

The establishment of “dual carbon” goals exemplifies China's global commitment as a responsible sovereign and the methodical advancement toward these aims is illustrative of China's capable governance in countering climate transformation. The actualization of “dual carbon” targets is contingent upon the foundation of robust fiscal policies, and meticulous assessments of policy documentation are instrumental in reflecting the foreseen efficacies of these measures. The study employs text mining techniques to articulate evaluative benchmarks for fiscal policy scripts under the “dual carbon” framework and engages a bidirectional fixed-effects model to corroborate the linkage between fiscal policy implements and carbon emissions, alongside a holistic appraisal using the PMC-CRITIC index model. The research corroborates that fiscal instruments, such as tax reliefs and green fiscal transfers, significantly encourage the diminution of carbon emissions. The average PMC-CRITIC index of the policy specimens assessed is 0.62, indicating a level that is permissible yet indicating potential for further refinement. Of the reviewed policy samples, the preponderance adheres to satisfactory thresholds, with an ensuing tier demonstrating exceptional policy. The average evaluations of policy timeliness, nature, and evaluation are laudable, notwithstanding the necessity for ameliorations in the precision of policy objectives, the credibility of the policy objects, and the targeted applicability of policy tools. Accordingly, in the trajectory of impending policy development, there should be an amplification of the collaborative mechanisms both horizontally among sundry tiers of local governance and vertically across disparate bureaucratic strata, the adoption of a systemized approach to elucidate and resolve the fundamental dissonances in actualizing “dual carbon” objectives.

Green taxation, regional green development and innovation: Mechanisms of influence and policy optimization

As the severity of global climate change escalates, carbon peak and carbon neutrality have emerged as international imperatives. Green taxation, a crucial fiscal strategy promoting the transition to a green low-carbon economy, demands thorough investigation regarding its efficacy and operational mechanisms. Predominantly, existing research has concentrated on the macroeconomic implications of green taxes, while their influence on regional green growth and innovation remains underexplored. This paper delineates green taxation in both a broad and narrow sense, elucidating its essence and potential in stimulating regional green progress and innovation. Utilizing panel data from 30 provinces in China spanning 2004 to 2021, a bidirectional fixed-effects model was developed for empirical analysis. Moreover, an impact mechanism model was constructed to delve into how green taxation specifically affects regional green development and innovation. Findings reveal that green taxation significantly fosters regional green growth and markedly enhances green innovation. Nonetheless, due to the resource crowding-out effect, it adversely impacts the comprehensive innovation capabilities of regional enterprises. This insight offers a fresh perspective on the paradoxical effects of green taxation, underscoring the necessity for a balanced approach in policy design and implementation. Drawing on these conclusions, the paper presents tailored policy recommendations.

Res-BiLSTMs model based on multi-task attention for real-time measurement of the free calcium oxide content

Abstract The content of free calcium oxide (f-CaO) is the primary economic index to evaluate the quality of cement. A residual bidirectional long short-term memory network model (Res-BiLSTMs) based on a multi-task attention mechanism was proposed for the characteristics of cement clinker production, used for online monitoring f-CaO content. The model utilizes the Bi-LSTM as the foundational component and combines the residual network to construct the Res-BiLSTMs coding structure, which aims to summarize the multi-level characteristic information of the input sequence. Additionally, a multi-task attention mechanism is introduced, combining the attention mechanism with semi-supervision to extract control coupling and data coupling among devices and variables. The results demonstrate that the addition of the multi-task attention mechanism led to a reduction in model errors by 0.0175 and 0.022, respectively, and an improvement in the degree of fit by 14.61%. The effectiveness of the multi-task attention mechanism for quality monitoring is confirmed. Compared to traditional LSTM, this model exhibited a reduction in errors by 0.0469 and 0.019, respectively, an increase in the correlation coefficient by 45.37%, and outperformed all other models in the comparison. The model’s measurement performance under limited labeled samples is also validated.

Application of Unet-SE-Bisru Algorithm for Music Signal Processing in Music Source Separation

At present, the use of time-domain deep learning for end-to-end neural network models has the problem of long training time and poor performance in music source separation. To address this issue, a U-network squeezing excitation bidirectional simple recursive unit model was proposed based on the deep extractor model. Replace Unet SE Bisru with Unet SE Bisru in the following text. This model improves the bidirectional long short-term memory network into a bidirectional simple recurrent unit, and then introduces attention mechanisms in the generalized encoding and decoding layers. The squeezing excitation block is used to selectively extract features based on the type of audio to be separated. Finally, group normalization is added after onedimensional convolution, And its effectiveness was verified. The experimental results show that the signal noise distortion ratio in the improved model is 5.68 decibels compared to the bidirectional simple recursive unit value, which is higher than the 5.55 decibels of bidirectional long short-term memory. After adding the squeezing excitation module, the overall increase is about 0.1-0.5 decibels. In addition, in the model comparison, the three indicators of the improved model with the same number of channels were 5.68 decibels, 5.91 decibels, and 11.28 decibels, respectively, higher than the benchmark model. Compared with other music source classification models, the improved model has better comprehensive separation performance. Although some indicators are lower than the comparison model, the signal noise distortion ratio of drum and bass is 6.11 decibels and 6.36 decibels, which is better than the comparison model. Overall, the improved model has high performance in music source separation for music signal processing and can be effectively applied in practical music source separation.

Cross-Social-Network User Identification Based on Bidirectional GCN and MNF-UI Models

Due to the distinct functionalities of various social network platforms, users often register accounts on different platforms, posing significant challenges for unified user management. However, current multi-social-network user identification algorithms heavily rely on user attributes and cannot perform user identification across multiple social networks. To address these issues, this paper proposes two identity recognition models. The first model is a cross-social-network user identification model based on bidirectional GCN. It calculates user intimacy using the Jaccard similarity coefficient and constructs an adjacency matrix to accurately represent user relationships in the social network. It then extracts cross-social-network user information to accomplish user identification tasks. The second model is the multi-network feature user identification (MNF-UI) model, which introduces the concept of network feature vectors. It effectively maps the structural features of different social networks and performs user identification based on the common features of seed nodes in the cross-network environment. Experimental results demonstrate that the bidirectional GCN model significantly outperforms baseline algorithms in cross-social-network user identification tasks. The MNF-UI (multi-network feature user identification) model can operate in situations with two or more networks with inconsistent structures, resulting in improved identification accuracy. These two user identification algorithms provide technical and theoretical support for in-depth research on social network information integration and network security maintenance.

Risky or hedging centre? A novel approach for bidirectional ripple effects of urban house prices

ABSTRACT This study introduces a novel bidirectional ripple effect model to examine the static and dynamic ripple effects in the housing markets of urban cities in China. By investigating the influence of idiosyncratic volatility on market co-movements, this study effectively identifies risky and hedging centres, along with the durations of the ripple effects. The results show that Beijing, Shanghai, Guangzhou, Hangzhou, Nanjing, Bengbu, Luoyang, and Wuhan are more likely to be identified as risky centres, whereas Ningbo and Wenzhou stand out as hedging centres. The results also show that most cities exhibit greater density and longer durations of risky ripple effects than hedging ripple effects, implying a switch between risky and hedging centres. Overall, the proposed method provides a comprehensive understanding of the ripple effects in housing markets and benefits policymakers and investors.

Probabilistic deep learning and transfer learning for robust cryptocurrency price prediction

Forecasting the price of Bitcoin (BTC) with precision is a complex endeavor, given the market’s inherent uncertainty and volatility, influenced by a diverse range of parameters. This research is driven by the central goal of introducing a specialized deep learning model tailored to predict digital currency prices, with a specific emphasis on BTC. To address this challenge, a pioneering strategy has been established, leveraging probabilistic gated recurrent units (P-GRU). This approach integrates probabilistic attributes into the model, facilitating the generation of probability distributions for projected values. The effectiveness of this method is assessed using one year of BTC price history, sampled at a five-minute interval. In parallel, a comparative analysis is conducted against alternative models, including GRU, long short-term memory (LSTM), and variants thereof (time-distributed, bidirectional, and simple models). In pursuit of optimizing model efficacy, a bespoke callback mechanism is deployed. This callback, driven by R2-score tracking, captures optimal model weights based on validation data. Moreover, a transfer learning paradigm is adopted to broaden the study’s horizons. A pre-trained model on BTC data is harnessed to predict prices for six other prominent cryptocurrencies: Ethereum, Litecoin, Tron, Polkadot, Cardano, and Stellar. Consequently, a distinct model is tailored for each cryptocurrency. The outcomes of this investigation conclusively underscore the superior performance of the proposed methodology. In the midst of a volatile and uncertain market landscape, the proposed approach outshines its counterparts, showcasing an enhanced ability for cryptocurrency price forecasting.

Bridging multi-level vehicle behavioural patterns with long-term pavement condition: a bidirectional modelling approach for progressively mixed autonomous traffic flow

The increasing of AVs’ market penetration rate (MPR) changes the traffic behavioural pattern and ultimately influences pavement. Pavement condition, in turn, influences microscopic behaviours. The above bidirection influencing mechanism has not been investigated fully yet. This research fill this gap by modelling the linkage between multi-level microscopic behaviour and pavement sustainability. Three levels of behaviour are considered: travel behaviour, microscopic traffic behaviour, and driving behaviour. The latter two are modelled using stochastic lateral movement model, while the former one is captured by a mixed autonomous traffic assignment model. Microscopic behaviours’ influence on pavement is described by a data-driven model, while the reverse counterpart is described by connecting pavement condition with link wandering and disutility. Our analysis results reveal that at link level, AVs can cause extra 50% drop in free flow speed due to pavement deterioration. Neglecting pavement evolution in strategic assignments leads to a 0.2% increase in network cost.

Medium-Term AQI Prediction in Selected Areas of Bangladesh Based on Bidirectional GRU Network Model

Medium-Term AQI Prediction in Selected Areas of Bangladesh Based on Bidirectional GRU Network Model

Matching analysis of new energy vehicle charging demand and charging infrastructure power supply capacity: A case study of China's capital Beijing

The random charging behavior of new energy vehicles (NEVs) will bring new challenges to the matching between electric vehicle charging facilities (EVCF) and NEVs. In order to explore whether the power supply capacity of urban EVCF can meet the charging requirements of NEVs, using the progression of NEVs in Beijing as a basis, initially, the Monte Carlo simulation (MCS) approach simulates the power demand trajectory for NEVs in the region. Subsequently, to forecast the ownership trends of NEVs across three distinct scenarios from 2021 to 2030, the study employs Grey correlation analysis combined with the bidirectional long short-term memory model (GRA-BiLSTM), facilitating the determination of NEVs’ charging needs. Second, the charging supply of EVCF for the next 10 years is derived by analyzing different development scenarios with three growth rates of EVCF and three combinations of fast and slow pile ratios. The findings indicate a discrepancy between the rate of increase in ownership of NEVs and the rate of increase in charging infrastructure in Beijing between 2021 and 2030. Even under a scenario of high growth in NEV ownership, the balance between supply and demand for charging capacity is not achieved, resulting in suboptimal utilization of charging infrastructure.