Contextual Correlates Research Articles

SCANet: Implementation of Selective Context Adaptation Network in Smart Farming Applications.

In the last decade, deep learning has enjoyed its spotlight as the game-changing addition to smart farming and precision agriculture. Such development has been predominantly observed in developed countries, while on the other hand, in developing countries most farmers especially ones with smallholder farms have not enjoyed such wide and deep adoption of this new technologies. In this paper we attempt to improve the image classification part of smart farming and precision agriculture. Agricultural commodities tend to possess certain textural details on their surfaces which we attempt to exploit. In this work, we propose a deep learning based approach called Selective Context Adaptation Network (SCANet). SCANet performs feature enhancement strategy by leveraging level-wise information and employing context selection mechanism. In exploiting contextual correlation feature of the crop images our proposed approach demonstrates the effectiveness of the context selection mechanism. Our proposed scheme achieves 88.72% accuracy and outperforms the existing approaches. Our model is evaluated on the cocoa bean dataset constructed from the real cocoa bean industry scene in Indonesia.

Charting everyday activities in later life: Study protocol of the mobility, activity, and social interactions study (MOASIS).

Prominent theories of aging emphasize the importance of resource allocation processes as a means to maintain functional ability, well-being and quality of life. Little is known about which activities and what activity patterns actually characterize the daily lives of healthy older adults in key domains of functioning, including the spatial, physical, social, and cognitive domains. This study aims to gain a comprehensive understanding of daily activities of community-dwelling older adults over an extended period of time and across a diverse range of activity domains, and to examine associations between daily activities, health and well-being at the within- and between-person levels. It also aims to examine contextual correlates of the relations between daily activities, health, and well-being. At its core, this ambulatory assessment (AA) study with a sample of 150 community-dwelling older adults aged 65 to 91 years measured spatial, physical, social, and cognitive activities across 30 days using a custom-built mobile sensor ("uTrail"), including GPS, accelerometer, and audio recording. In addition, during the first 15 days, self-reports of daily activities, psychological correlates, contexts, and cognitive performance in an ambulatory working memory task were assessed 7 times per day using smartphones. Surrounding the ambulatory assessment period, participants completed an initial baseline assessment including a telephone survey, web-based questionnaires, and a laboratory-based cognitive and physical testing session. They also participated in an intermediate laboratory session in the laboratory at half-time of the 30-day ambulatory assessment period, and finally returned to the laboratory for a posttest assessment. In sum, this is the first study which combines multi-domain activity sensing and self-report ambulatory assessment methods to observe daily life activities as indicators of functional ability in healthy older adults unfolding over an extended period (i.e., 1 month). It offers a unique opportunity to describe and understand the diverse individual real-life functional ability profiles characterizing later life.

Contextual and Behavioral Correlates of Coping Strategies Among an Ethnically Diverse Sample of Urban Adolescents in the Midwestern United States.

Coping is recognized as an important life skill. In the present cross-sectional analysis, early adolescents' relationships with their caregivers (support, conflict) and exposure to stressors (uncontrollable life events, violence) were examined as contextual correlates of both positive and negative coping strategies. Coping strategies were examined as mediators of associations between adolescents' family and community contexts and adjustment outcomes (externalizing symptoms, internalizing symptoms, academic investment). Participants were recruited from an urban Pre-K-8 school and Boys and Girls Club. Adolescents who reported greater support from caregivers reported greater engagement in all forms of positive coping (behavioral/problem-focused coping, cognitive/emotion-focused coping, and coping through seeking support); they also reported less engagement in coping through anger and helplessness. Adolescents who reported greater conflict with caregivers or violence exposure reported greater engagement in coping through avoidance, anger, and helplessness. Problem-focused coping, coping through anger, and coping through helplessness mediated associations between different contextual factors and outcomes.

Toward High-quality Face-Mask Occluded Restoration

Face-mask occluded restoration aims at restoring the masked region of a human face, which has attracted increasing attention in the context of the COVID-19 pandemic. One major challenge of this task is the large visual variance of masks in the real world. To solve it we first construct a large-scale Face-mask Occluded Restoration (FMOR) dataset, which contains 5,500 unmasked images and 5,500 face-mask occluded images with various illuminations, and involves 1,100 subjects of different races, face orientations, and mask types. Moreover, we propose a Face-Mask Occluded Detection and Restoration (FMODR) framework, which can detect face-mask regions with large visual variations and restore them to realistic human faces. In particular, our FMODR contains a self-adaptive contextual attention module specifically designed for this task, which is able to exploit the contextual information and correlations of adjacent pixels for achieving high realism of the restored faces, which are however often neglected in existing contextual attention models. Our framework achieves state-of-the-art results of face restoration on three datasets, including CelebA, AR, and our FMOR datasets. Moreover, experimental results on AR and FMOR datasets demonstrate that our framework can significantly improve masked face recognition and verification performance.

CEMA-LSTM: Enhancing Contextual Feature Correlation for Radar Extrapolation Using Fine-Grained Echo Datasets

Accurate precipitation nowcasting can provide great convenience to the public so they can conduct corresponding arrangements in advance to deal with the possible impact of upcoming heavy rain. Recent relevant research activities have shown their concerns on various deep learning models for radar echo extrapolation, where radar echo maps were used to predict their consequent moment, so as to recognize potential severe convective weather events. However, these approaches suffer from an inaccurate prediction of echo dynamics and unreliable depiction of echo aggregation or dissipation, due to the size limitation of convolution filter, lack of global feature, and less attention to features from previous states. To address the problems, this paper proposes a CEMA-LSTM recurrent unit, which is embedded with a Contextual Feature Correlation Enhancement Block (CEB) and a Multi-Attention Mechanism Block (MAB). The CEB enhances contextual feature correlation and supports its model to memorize significant features for near-future prediction; the MAB uses a position and channel attention mechanism to capture global features of radar echoes. Two practical radar echo datasets were used involving the FREM and CIKM 2017 datasets. Both quantification and visualization of comparative experimental results have demonstrated outperformance of the proposed CEMA-LSTM over recent models, e.g., PhyDNet, MIM and PredRNN++, etc. In particular, compared with the second-ranked model, its average POD, FAR and CSI have been improved by 3.87%, 1.65% and 1.79%, respectively on the FREM, and by 1.42%, 5.60% and 3.16%, respectively on the CIKM 2017.

Multiscale Adaptive Fusion Network for Hyperspectral Image Denoising

Removing the noise and improving the visual quality of hyperspectral images (HSIs) is challenging in academia and industry. Great efforts have been made to leverage local, global or spectral context information for HSI denoising. However, existing methods still have limitations in feature interaction exploitation among multiple scales and rich spectral structure preservation. In view of this, we propose a novel solution to investigate the HSI denoising using a Multi-scale Adaptive Fusion Network (MAFNet), which can learn the complex nonlinear mapping between clean and noisy HSI. Two key components contribute to improving the hyperspectral image denoising: A progressively multiscale information aggregation network and a co-attention fusion module. Specifically, we first generate a set of multiscale images and feed them into a coarse-fusion network to exploit the contextual texture correlation. Thereafter, a fine fusion network is followed to exchange the information across the parallel multiscale subnetworks. Furthermore, we design a co-attention fusion module to adaptively emphasize informative features from different scales, and thereby enhance the discriminative learning capability for denoising. Extensive experiments on synthetic and real HSI datasets demonstrate that the proposed MAFNet has achieved better denoising performance than other state-of-the-art techniques. Our codes are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/summitgao/MAFNet</uri> .

RADANet: Road Augmented Deformable Attention Network for Road Extraction From Complex High-Resolution Remote-Sensing Images

Extracting roads from complex high-resolution remote sensing images to update road networks has become a recent research focus. How to apply the contextual spatial correlation and topological structure of the roads properly to improve the extraction accuracy becomes a challenge in the increasingly complex road environment. In this article, inspired by the prior knowledge of the road shape and the progress in deformable convolution, we proposed a road augmented deformable attention network (RADANet) to learn the long-range dependencies for specific road pixels. We developed a road augmentation module (RAM) to capture the semantic shape information of the road from four strip convolutions. Deformable attention module (DAM) combines the sparse sampling capability of deformable convolution with the spatial self-attention mechanism. The integration of RAM enables DAM to extract road features more specifically. Furthermore, RAM is placed behind the fourth stage of encoder, and DAM is placed between last four stages of encoder and decoder in RADANet to extract multiscale road semantic information. Comprehensive experiments on representative public datasets (DeepGlobe and CHN6-CUG road datasets) demonstrate that our RADANet achieves advanced results compared with the state-of-the-art methods.

Translation of Discourse Marker Okay from the Perspective of Adaptation Theory

Films are mainly comprised of dialogues in which discourse markers are commonly used to structure the interaction. The use of discourse markers in fabricated text can increase the naturalness of language. In order to recreate the naturalness and orality of the fictional dialogues, due attention should be paid to the translation of discourse markers. This paper takes the discourse marker Okay as an example and gives an analysis of the translation of the discourse marker Okay from the four angels of Adaptation Theory, including contextual correlates of adaptability, structural objects of adaptability, the dynamics of adaptability and the salience of the adaptation processes. Based on the analysis of the translation of Okay, it can be seen that in the process of choice-making, the translator makes dynamic adaptation with different degrees of salience to both the context and the structure so as to make appropriate choices and achieve successful language communication.

On the Consecutive Interpreting of Chinese Governmental Discourse from the Perspective of Adaptation Theory: The Interpreting of Premier Li Keqiang Meets the Press in 2022 as an Example

The consecutive interpreting of governmental discourse shoulders the responsibility to represent the official stance of the country, and the seriousness of the discourse puts forward higher standards for interpreting output: interpreters should reasonably choose and adapt to various internal and external factors of both the source language and target language in order to assume the role of an accurate, authoritative, and immediate intermediary of information transfer. Adaptation theory just provides interpreters with standards for making choices and adaptations. Taking Verschueren’s adaptation theory as the theoretical framework, this study analyzes the consecutive interpreting of Premier Li Keqiang Meets the Press in 2022, a representative example of consecutive interpreting of Chinese governmental discourse, from the angles of contextual correlates of adaptability, structural objects of adaptability, and salience of the adaptation processes, aiming to explore how could interpreters adapt through language choices in Chinese governmental discourse, and to reveal the adaptation theory’s role as the standard and guidance for consecutive interpreting.

Salient Object Detection in Optical Remote Sensing Images Driven by Transformer.

Existing methods for Salient Object Detection in Optical Remote Sensing Images (ORSI-SOD) mainly adopt Convolutional Neural Networks (CNNs) as the backbone, such as VGG and ResNet. Since CNNs can only extract features within certain receptive fields, most ORSI-SOD methods generally follow the local-to-contextual paradigm. In this paper, we propose a novel Global Extraction Local Exploration Network (GeleNet) for ORSI-SOD following the global-to-local paradigm. Specifically, GeleNet first adopts a transformer backbone to generate four-level feature embeddings with global long-range dependencies. Then, GeleNet employs a Direction-aware Shuffle Weighted Spatial Attention Module (D-SWSAM) and its simplified version (SWSAM) to enhance local interactions, and a Knowledge Transfer Module (KTM) to further enhance cross-level contextual interactions. D-SWSAM comprehensively perceives the orientation information in the lowest-level features through directional convolutions to adapt to various orientations of salient objects in ORSIs, and effectively enhances the details of salient objects with an improved attention mechanism. SWSAM discards the direction-aware part of D-SWSAM to focus on localizing salient objects in the highest-level features. KTM models the contextual correlation knowledge of two middle-level features of different scales based on the self-attention mechanism, and transfers the knowledge to the raw features to generate more discriminative features. Finally, a saliency predictor is used to generate the saliency map based on the outputs of the above three modules. Extensive experiments on three public datasets demonstrate that the proposed GeleNet outperforms relevant state-of-the-art methods. The code and results of our method are available at https://github.com/MathLee/GeleNet.

Semantic and Temporal Contextual Correlation Learning for Weakly-Supervised Temporal Action Localization.

Weakly-supervised temporal action localization (WSTAL) aims to automatically identify and localize action instances in untrimmed videos with only video-level labels as supervision. In this task, there exist two challenges: (1) how to accurately discover the action categories in an untrimmed video (what to discover); (2) how to elaborately focus on the integral temporal interval of each action instance (where to focus). Empirically, to discover the action categories, discriminative semantic information should be extracted, while robust temporal contextual information is beneficial for complete action localization. However, most existing WSTAL methods ignore to explicitly and jointly model the semantic and temporal contextual correlation information for the above two challenges. In this article, a Semantic and Temporal Contextual Correlation Learning Network (STCL-Net) with the semantic (SCL) and temporal contextual correlation learning (TCL) modules is proposed, which achieves both accurate action discovery and complete action localization by modeling the semantic and temporal contextual correlation information for each snippet in the inter- and intra-video manners respectively. It is noteworthy that the two proposed modules are both designed in a unified dynamic correlation-embedding paradigm. Extensive experiments are performed on different benchmarks. On all the benchmarks, our proposed method exhibits superior or comparable performance in comparison to the existing state-of-the-art models, especially achieving gains as high as 7.2% in terms of the average mAP on THUMOS-14. In addition, comprehensive ablation studies also verify the effectiveness and robustness of each component in our model.

A General Dual-Branch Framework for Land Cover Mapping Models With Multispectral Data

Land cover mapping based on multi-spectral images can in principle be considered an application of semantic segmentation, but land cover mapping inputs include near-infrared (NIR) data in addition to RGB data. It has been experimentally found that LULC mapping performance based on RGB data alone is better than that based on RGB and NIR data when using some established single-branch encoder-decoder models. To address this issue, we propose a dual-branch encoder-decoder (DBED) framework that can be applied to existing encoder-decoder models for semantic segmentation. First, the multi-spectral input data is divided into two parts: RGB & NIR, and fed to the respective branch for encoding. The dual-branch structure facilitates cross-modal complementary information encoding without deteriorating the original RGB modality-specific feature extraction. Second, an attention module named multi-spectral attention module (MSAM) is proposed to mine the contextual correlation between the multi-spectral feature maps, leading to further performance boosting. We apply this framework to three mainstream semantic segmentation models and validate it on the Gaofen Image Dataset. Experimental results show that this structure brings performance improvements. The source code of DBED is publicly available: https://github.com/MalignusCN/DBED.

Multi‐scale homography estimation based on dual feature aggregation transformer

AbstractThe accuracy of registration in image stitching task directly affects the performance of subsequent stages. Traditional registration methods rely heavily on the quality of the features when calculating the homography matrix, resulting in alignment failures in low‐texture or low‐overlap scenes due to extracting insufficient features. On the other hand, existing DNN‐based methods for homography estimation are more robust in multiple scenes but previous work usually employs an overly simple convolutional network structure to directly regress the homography, ignoring the redundant information contained in the feature maps so that their prediction accuracy is inferior to the traditional methods in simple scenarios. To overcome the disadvantages of the two methods, a Multi‐scale structure is proposed to extract feature maps at three scales and design two modules to handle the matrix prediction respectively. The DFA‐T module analyzes semantic information on the high‐level features to accomplish coarse‐grained alignment while the Contextual Correlation module on the bottom level to accomplish more accurate alignment. Experiments demonstrate that this method provides more accurate alignment results than the existing state‐of‐the‐art DNN‐based methods and outperforms traditional algorithms with more stable results in some extreme scenarios.

Unsupervised Word Embedding with Ensemble Deep Learning for Twitter Rumor Identification

In social networks, rumor identification is a major problem. The structural data in a topic is applied to derive useful attributes for rumor identification. Most standard rumor identification methods concentrate on local structural attributes, ignoring the global structural attributes that exist between the source tweet and its responses. To tackle this issue, a Source-Replies relation Graph (SR-graph) has been built to develop an Ensemble Graph Convolutional neural Net (EGCN) with a Nodes Proportion Allocation Mechanism (NPAM) which identifies the rumor. But, the word vectors were trained by the standard word-embedding model which does not increase the accuracy for large Twitter databases. To solve this problem, an unsupervised word-embedding method is needed for large Twitter corpora. As a result, the Twitter word-embedded EGCN (T-EGCN) model is proposed in this article, which uses unsupervised learning-based word embedding to find rumors in huge Twitter databases. Initially, the latent contextual semantic correlation and co-occurrence statistical attributes among words in tweets are extracted. Then, to create a rumor attribute vector of tweets, these word embeddings are concatenated with the GloVe model's word attribute vectors, Twitter-specific attributes, and n-gram attributes. Further, the EGCN is trained by using this attribute vector to identify rumors in a huge Twitter database. Finally, the testing results exhibit that the T-EGCN achieves 87.56% accuracy, whereas the RNN, GCN, PGNN, EGCN, and BiLSTM-CNN attain 65.38%, 68.41%, 75.04%, 81.87%, and 86.12%, respectively for rumor identification.

RISK FACTORS AND IMPACT OF FINANCIAL FRAUD ON OLDER ADULTS AND DISADVANTAGED COMMUNITIES

Abstract In 2021, Americans age 80 and older reported median fraud losses of $1,300 per incident, substantially higher than younger age groups (Federal Trade Commission, 2022). Using data from the Federal Trade Commission’s Sentinel Fraud Complaint Database and the Area Deprivation Index, this study assesses the demographic and contextual correlates of consumer fraud victimization in the United States. Models show a positive association between living in a disadvantaged community and reporting victimization (versus reporting no losses after being targeted). Black and Hispanic consumers are significantly more likely report victimization, although non-Hispanic White victims report higher average fraud losses. Results also show that older adults are significantly less likely to report fraud victimization relative to people age 30 and younger, but older victims lose hundreds of dollars more, on average. Applying a sentiment analysis to the consumer complaint narratives, we find that compared to emotionally neutral complaints, emotionally positive and emotionally negative complaints are positively associated with reporting victimization versus reporting no losses. Gift cards are the most common method of payment demanded by fraud criminals overall, but victims who pay using cash, wire transfer, or cryptocurrency lose significantly more money, on average. To limit the impact of fraud on the aging population, consumer protection agencies need to focus their efforts on educating older adults and members of socially and economically disadvantaged communities about scams, as well as enforce greater controls on cryptocurrency exchanges and retail gift card sales.

Contextual Interpretation: Correlation of KH Bisri Mustafa’s Interpretation and Abdullah Saeed’s Contextual Approach to Q.S. Yūsuf: 55

Abstract.Interpretive works in the archipelago are often influenced by the subjectivity of mufassir. This can be seen from the language and script used in writing the interpretation, as well as the local issues included in the interpretation, one of which is the interpretation of Al-Ibrīz by K.H. Bisri Musthafa. The authorship of subjectivity in interpretation is inevitable, it is argued by Abdullah Saeed in his contextual approach that the first step that an interpreter must pay attention to is to understand the subjectivity of the mufassir. In his interpretation of Q.S Yūsuf: 55, K.H. Bisri Musthafa allows a person to ask for a position, but he also presents a hadith that forbids asking for office. To produce a contemporary meaning and become a solution to current social problems, the author will also read the verse through the contextual approach of Abdullah Saeed. This article aims to see how the interpretation of Q.S Yūsuf: 55 in the interpretation of Al-Ibrīz and also in the contextual approach of Abdullah Saeed. Then the author also wants to show the correlation between the results of the interpretation of the two. The method used in this study is descriptive-analytical, the results of which will expose the contextual correlation of the interpretation of KH. Bisri Musthafa and the results of interpretation through Saeed's contextual approach to Q.S Yūsuf: 55. The results of this study prove that the moral ideal of the verse is that a Muslim is required to participate in advancing the life of the nation and state while volunteering as a leader is one of the wasīlaḥ in achieving that goal.&#x0D;

On using contextual correlation to detect multi-stage cyber attacks in smart grids

While the digitization of the distribution grids brings numerous benefits to grid operations, it also increases the risks imposed by serious cyber security threats such as coordinated, timed attacks. Addressing this new threat landscape requires an advanced security approach beyond established preventive IT security measures such as encryption, network segmentation, or access control. Here, detective capabilities and reactive countermeasures as part of incident response strategies promise to complement nicely the security-by-design approach by providing cyber security situational awareness. However, manually evaluating extensive cyber intelligence within a reasonable timeframe requires an unmanageable effort to process a large amount of cross-domain information. An automated procedure is needed to systematically process and correlate the various cyber intelligence to correctly assess the situation to reduce the manuel effort and support security operations. In this paper, we present an approach that leverages cyber intelligence from multiple sources to detect multi-stage cyber attacks that threaten the smart grid. We investigate the detection quality of the presented correlation approach and discuss the results to highlight the challenges in automated methods for contextual assessment and understanding of the cyber security situation.

Image inpainting via spatial projections

Image inpainting is now-a-days sought after due to its wide variety of applications in the reconstruction of the corrupted image, occlusion removal, reflection removal, etc. Existing image inpainting approaches utilize different types of attention mechanisms to inpaint the image and produce visibly admirable results. These methods are more concerned at weighing the feature maps of the hole region with some weight from the non-hole region. But, due to the lack of spatial contextual correlation in the attention maps, the inpainted image may suffer from the inconsistencies among hole and non-hole regions. Transformer-based inpainting methods give significant results by capturing the relationship between the patches with a compromise of high computational complexity. In this context, we propose a novel spatial projection layer (SPL) without any attention mechanism to project the spatial contextual information in the hole region from non-hole regions for producing a spatially plausible inpainted image. The SPL is proposed mainly to focus on the non-hole spatial information in the high-level feature maps for filling the hole regions efficiently. Also, while training the network, we propose the use of edge loss with a Canny edge operator for image inpainting to focus on the relevant edges instead of noise contents. Analysis with the extensive experiments, ablation, and user study on the proposed architecture demonstrates the superiority over existing state-of-the-art methods for image inpainting. The code is available at: https://github.com/shrutiphutke/spatial_projection_inpainting.

HIV and suicide risk across adolescence and young adulthood: an examination of socio-demographic, contextual and psychosocial risk factors for attempted suicide in a longitudinal cohort of ageing adolescents affected by HIV living in the New York City Area.

IntroductionAs children become adolescents and young adults (AYA), their risk for attempting suicide increases dramatically, with chronic health conditions an important risk factor. This study examined correlates of suicidality across development in AYA living with perinatally acquired HIV (AYALPHIV) and those perinatally HIV‐exposed but uninfected (AYAPHEU).MethodsData come from an ongoing longitudinal New York City‐based study (N = 339) with AYALPHIV and AYAPHEU interviewed every 12–18 months from 2003 to 2019 (mean enrolment age = 12 years; current mean age = 27 years). The Diagnostic Interview Schedule for Children (adolescent or young adult version) assessed psychiatric disorders and first‐reported suicide attempt. Generalized estimating equations were used to examine associations between first‐reported suicide attempt and socio‐demographic, contextual and psychosocial correlates measured concurrently across six timepoints.ResultsAt enrolment, 51% of participants were female, 72% heterosexual, 60% Black and 50% Latinx. Attempted suicide was significantly higher among AYALPHIV (27%, CI 21–33%) compared to AYAPHEU (16%, CI 10–22%), with an OR of 1.74 (CI 1.04–2.92) in a model adjusting for age. For AYALPHIV, anxiety (OR 2.66, CI 1.46–4.85), mood (OR 3.62, CI 1.49–8.81) and behaviour disorders (OR 5.05, CI 2.15–11.87) and past‐year arrest (OR 3.05, CI 1.26–7.4), negative life events (OR 1.27, CI 1.11–1.46), city stress (OR 2.28, CI 1.46–3.57), pregnancy (OR 2.28, CI 1.08–4.81) and HIV stigma (OR 2.46, CI 1.27–4.75) were associated with increased odds of attempted suicide, while identifying as heterosexual (OR 0.27, CI 0.14–0.52), higher personal (OR 0.45, CI 0.26–0.80) and family self‐concept (OR 0.36, CI 0.22–0.57) were protective. Interactions by HIV status and age were found: substance use was more strongly associated with attempted suicide among AYAPHEU than AYALPHIV, while negative life events and higher religiosity were more strongly associated with increased odds of attempted suicide among AYA ≥ 19 versus ≤ 18 years.ConclusionsAYALPHIV compared to AYAPHEU faced unique risks for attempted suicide as they age into adulthood, with the highest risk among AYALPHIV experiencing HIV stigma or pregnancy and the highest risk among AYAPHEU with substance use. Assessing for suicide risk and correlates with attention to ageing can inform preventive interventions tailored to meet AYALPHIV and AYAPHEU needs.

Toward Understanding Most of the Context in Document-Level Neural Machine Translation

Considerable research has been conducted to obtain translations that reflect contextual information in documents and simultaneous interpretations. Most of the existing studies use concatenation data which merge previous and current sentences for training translation models. Although this corpus improves the performance of the model, ignoring the contextual correlation between the sentences can disturb translation performance. In this study, we introduce a simple and effective method to capture the contextual correlation of the sentence at the document level of the current sentence, thereby learning an effective contextual representation. In addition, the proposed model structure is applied to a separate residual connection network to minimize the loss of the beneficial influence of incorporating the context. The experimental results show that our methods improve the translation performance in comparison with the state-of-the-art baseline of the Transformer in various translation tasks and two benchmark machine translation tasks.