Relational Features Research Articles

AFTGAN: prediction of multi-type PPI based on attention free transformer and graph attention network.

Protein-protein interaction (PPI) networks and transcriptional regulatory networks are critical in regulating cells and their signaling. A thorough understanding of PPIs can provide more insights into cellular physiology at normal and disease states. Although numerous methods have been proposed to predict PPIs, it is still challenging for interaction prediction between unknown proteins. In this study, a novel neural network named AFTGAN was constructed to predict multi-type PPIs. Regarding feature input, ESM-1b embedding containing much biological information for proteins was added as a protein sequence feature besides amino acid co-occurrence similarity and one-hot coding. An ensemble network was also constructed based on a transformer encoder containing an AFT module (performing the weight operation on vital protein sequence feature information) and graph attention network (extracting the relational features of protein pairs) for the part of the network framework. The experimental results showed that the Micro-F1 of the AFTGAN based on three partitioning schemes (BFS, DFS and the random mode) on the SHS27K and SHS148K datasets was 0.685, 0.711 and 0.867, as well as 0.745, 0.819 and 0.920, respectively, all higher than that of other popular methods. In addition, the experimental comparisons confirmed the performance superiority of the proposed model for predicting PPIs of unknown proteins on the STRING dataset. The source code is publicly available at https://github.com/1075793472/AFTGAN. Supplementary data are available at Bioinformatics online.

Text-based person search via local-relational-global fine grained alignment

The core difficulty of text-based person search is how to achieve fine-grained alignment of visual and linguistic modal data, so as to bridge the gap of modal heterogeneity. Most existing works on this task focus on global and local features extraction and matching, ignoring the importance of relational information. This paper proposes a new text-based person search model, named CM-LRGNet, which extracts Cross-Modal Local-Relational-Global features in an end-to-end manner, and performs fine-grained cross-modal alignment on the above three feature levels. Concretely, we first split the convolutional feature maps to obtain local features of images, and adaptively extract textual local features. Then a relation encoding module is proposed to implicitly learn the relational information implied in the images and texts. Finally, a relation-aware graph attention network is designed to fuse the local and relational features to generate global representations for both images and text queries. Extensive experimental results on benchmark dataset (CUHK-PEDES) show that our approach can achieve state-of-the-art performance (64.18%, 82.97%, 89.85% in terms of Top-1, Top-5, and Top-10 accuracies), by learning and aligning local-relational-global representations from different modalities. Our code has been released in https://github.com/zhangweifeng1218/Text-based-Person-Search.

Boosting Scene Graph Generation with Visual Relation Saliency

The scene graph is a symbolic data structure that comprehensively describes the objects and visual relations in a visual scene, while ignoring the inherent perceptual saliency of each visual relation (i.e., relation saliency). However, humans often quickly allocate attention to important/salient visual relations in a scene. To align with such human perception of a scene, we explicitly model the perceptual saliency of visual relation in scene graph by upgrading each graph edge (i.e., visual relation) with an attribute of relation saliency. We present a new design, named as Saliency-guided Message Passing (SMP), that boosts the generation of such scene graph structure with the guidance from the visual relation saliency. Technically, an object interaction encoder is first utilized to strengthen object relation representations by jointly exploiting the appearance, semantic, and spatial relations in between. A branch is further leveraged to estimate the relation saliency of each visual relation by ordinal regression. Next, conditioned on the object and relation features (coupled with the estimated relation saliency), our SMP enhances scene graph generation by performing message passing over the objects and the most salient relations. Extensive experiments on VG-KR and VG150 datasets demonstrate the superiority of SMP for the scene graph generation. Moreover, we empirically validate the compelling generalizability of the learned scene graphs via SMP on downstream tasks like cross-model retrieval and image captioning.

Cognition Guided Human-Object Relationship Detection

Human-object relationship detection reveals the fine-grained relationship between humans and objects, helping the comprehensive understanding of videos. Previous human-object relationship detection approaches are mainly developed with object features and relation features without exploring the specific information of humans. In this paper, we propose a novel Relation-Pose Transformer (RPT) for human-object relationship detection. Inspired by the coordination of eye-head-body movements in cognitive science, we employ the head pose to find those crucial objects that humans focus on and use the body pose with skeleton information to represent multiple actions. Then, we utilize the spatial encoder to capture spatial contextualized information of the relation pair, which integrates the relation features and pose features. Next, the temporal decoder aims to model the temporal dependency of the relationship. Finally, we adopt multiple classifiers to predict different types of relationships. Extensive experiments on the benchmark Action Genome validate the effectiveness of our proposed method and show the state-of-the-art performance compared with related methods.

Estetik Cerrahi Operasyonu Olan Bireylerin Temel Kişilik Özelliklerinin ve Sosyal Görünüş Kaygılarının İncelenmesi: Tanımlayıcı Bir Çalışma

Objective: This study was performed with the aim of determining the basic personality traits and social appearance anxiety of individuals undergoing esthetic surgery. Material and Methods: The research examined descriptive and relational features and included individuals undergoing esthetic surgical operations from May-September 2021. Sample selection was not performed in the research and the sample included 267 individuals abiding by the inclusion criteria. Data were collected with a descriptive information form, social appearance anxiety scale, and basic personality traits scale. The fit of data to normal distribution was assessed with the Shapiro-Wilk analysis. As data abided by normal distribution, the independent samples t-test and Pearson correlation analyses were used. In order to carry out the study, ethics committee permission from Nevşehir Hacı Bektaş Veli University and consent from the individuals within the scope of the study were obtained. Results: Individuals who were woman, single, employed and having esthetic operations for a nose were determined to have higher social appearance anxiety (p<0.05). There were significant differences identified between gender, marital status and income level with basic personality trait scale points (p<0.05). There were positive high-level correlations between the emotional inconsistency and negative values subdimensions of the basic personality traits scale with social appearance anxiety level (r: 0.606, 0.642; p<0.05), while there were negative moderate-level correlations between extraversion, openness to experience and adjustment subdimensions with social appearance anxiety levels (r: -0.491, -0.473, - 0.557; p<0.05). Conclusion: Individuals undergoing aesthetic surgery had high social appearance anxiety, there was difference between the subdimensions of basic personality traits according to their sociodemographic characteristics, and there was a relationship between basic personality traits and social appearance anxiety.

Process bottlenecks identification and its root cause analysis using fusion-based clustering and knowledge graph

Drawing the strengths of data science and machine learning, process mining has recently emerged as an effective research approach for process management and its decision support. Bottleneck identification and analysis is a key problem in process mining which is considered a critical component for process improvement. While previous studies focusing on bottlenecks have been reported, visible gaps remain. Most of these studies considered bottleneck identification from local perspectives by quantitative metrics, such as machine operation and resource requirement, which can not be applied to knowledge-intensive processes. Moreover, the root cause of such bottlenecks has not been given enough attention, which limits the impact of process optimisation. This paper proposes an approach that utilises fusion-based clustering and hyperbolic neural network-based knowledge graph embedding for bottleneck identification and root cause analysis. Firstly, a fusion-based clustering is proposed to identify bottlenecks automatically from a global perspective, where the execution frequency of each stage at different periods is calculated to reveal the abnormal stage. Secondly, a process knowledge graph representing tasks, organisations, workforce and relation features as hierarchical and logical patterns is established. Finally, a hyperbolic cluster-based community detection mechanism is researched, based on the process knowledge graph embedding trained by a hyperbolic neural network, to analyse the root cause from a process perspective. Experimental studies using real-world data collected from a multidisciplinary design project revealed the merits of the proposed approach. The execution of the proposed approach is not limited to event logs; it can automatically identify bottlenecks without local quantitative metrics and analyse the causes from a process perspective.

SRRV: A Novel Document Object Detector Based on Spatial-Related Relation and Vision

Document object detection is a challenging task due to layout complexity and object diversity. Most of existing methods mainly focus on vision information, neglecting representative inherent spatial-related relationship between document objects. To capture structural information and contextual dependencies among document objects, we propose a novel framework combining spatial-related relation and vision (SRRV) for document object detection. SRRV consists of three parts: vision feature extraction network, relation feature aggregation network and result refinement network. Vision feature extraction network enhances information propagation of hierarchical feature pyramid by adopting feature augmentation paths. Then, relation feature aggregation network combines graph construction module and graph learning module. Specifically, graph construction module calculates spatial awareness information from geometric attributes to encode relationship, while graph learning module stacks Graph Convolutional Network (GCN) layers to aggregate relation information propagated among objects at global scale. Both the vision and relation features are fed into result refinement network for feature fusion and relational reasoning. Experiments on the PubLayNet, POD and Article Regions demonstrate that spatial relation information significantly improve the performance with better accuracy and precise bounding box prediction.

Arbitrary-Shape Scene Text Detection via Visual-Relational Rectification and Contour Approximation

One trend in the latest bottom-up approaches for arbitrary-shape scene text detection is to determine the links between text segments using Graph Convolutional Networks (GCNs). However, the performance of these bottom-up methods is still inferior to that of state-of-the-art top-down methods even with the help of GCNs. We argue that a cause of this is that bottom-up methods fail to make proper use of visual-relational features, which results in accumulated false detection, as well as the error-prone route-finding used for grouping text segments. In this paper, we improve classic bottom-up text detection frameworks by fusing the visual-relational features of text with two effective false positive/negative suppression (FPNS) mechanisms and developing a new shape-approximation strategy. First, dense overlapping text segments depicting the ‘`characterness’' and ‘`streamline’' properties of text are constructed and used in weakly supervised node classification to filter the falsely detected text segments. Then, relational features and visual features of text segments are fused with a novel Location-Aware Transfer (LAT) module and Fuse Decoding (FD) module to jointly rectify the detected text segments. Finally, a novel multiple-text-map-aware contour-approximation strategy is developed based on the rectified text segments, instead of the error-prone route-finding process, to generate the final contour of the detected text. Experiments conducted on five benchmark datasets demonstrate that our method outperforms the state-of-the-art performance when embedded in a classic text detection framework, which revitalizes the strengths of bottom-up methods.

Spatio-Temporal Self-Attention Network for Video Saliency Prediction

3D convolutional neural networks have achieved promising results for video tasks in computer vision, including video saliency prediction that is explored in this paper. However, 3D convolution encodes visual representation merely on fixed local spacetime according to its kernel size, while human attention is always attracted by relational visual features at different time. To overcome this limitation, we propose a novel Spatio-Temporal Self-Attention 3D Network (STSANet) for video saliency prediction, in which multiple Spatio-Temporal Self-Attention (STSA) modules are employed at different levels of 3D convolutional backbone to directly capture long-range relations between spatio-temporal features of different time steps. Besides, we propose an Attentional Multi-Scale Fusion (AMSF) module to integrate multi-level features with the perception of context in semantic and spatio-temporal subspaces. Extensive experiments demonstrate the contributions of key components of our method, and the results on DHF1K, Hollywood-2, UCF, and DIEM benchmark datasets clearly prove the superiority of the proposed model compared with all state-of-the-art models.

Point-Based Learnable Query Generator for Human-Object Interaction Detection.

Transformer-based and interaction point-based methods have demonstrated promising performance and potential in human-object interaction detection. However, due to differences in structure and properties, direct integration of these two types of models is not feasible. Recent Transformer-based methods divide the decoder into two branches: an instance decoder for human-object pair detection and a classification decoder for interaction recognition. While the attention mechanism within the Transformer enhances the connection between localization and classification, this paper focuses on further improving HOI detection performance by increasing the intrinsic correlation between instance and action features. To address these challenges, this paper proposes a novel Transformer-based HOI Detection framework. In the proposed method, the decoder contains three parts: learnable query generator, instance decoder, and interaction classifier. The learnable query generator aims to build an effective query to guide the instance decoder and interaction classifier to learn more accurate instance and interaction features. These features are then applied to update the query generator for the next layer. Especially, inspired by the interaction point-based HOI and object detection methods, this paper introduces the prior bounding boxes, keypoints detection and spatial relation feature to build the novel learnable query generator. Finally, the proposed method is verified on HICO-DET and V-COCO datasets. The experimental results show that the proposed method has the better performance compared with the state-of-the-art methods.

Masked Relation Learning for DeepFake Detection

DeepFake detection aims to differentiate falsified faces from real ones. Most approaches formulate it as a binary classification problem by solely mining the local artifacts and inconsistencies of face forgery, which neglect the relation across local regions. Although several recent works explore local relation learning for DeepFake detection, they overlook the propagation of relational information and lead to limited performance gains. To address these issues, this paper provides a new perspective by formulating DeepFake detection as a graph classification problem, in which each facial region corresponds to a vertex. But relational information with large redundancy hinders the expressiveness of graphs. Inspired by the success of masked modeling, we propose Masked Relation Learning which decreases the redundancy to learn informative relational features. Specifically, a spatiotemporal attention module is exploited to learn the attention features of multiple facial regions. A relation learning module masks partial correlations between regions to reduce redundancy and then propagates the relational information across regions to capture the irregularity from a global view of the graph. We empirically discover that a moderate masking rate (e.g., 50%) brings the best performance gain. Experiments verify the effectiveness of Masked Relation Learning and demonstrate that our approach outperforms the state of the art by 2% AUC on the cross-dataset DeepFake video detection. Code will be available at https://github.com/zimyang/MaskRelation.

A Multidimensional Feature-Driven Ensemble Model for Accurate Classification of Complex Power Quality Disturbance

As the proportion of power electronics-related facilities in modern power systems increases, the types of power quality disturbances (PQDs) tend to become more complex. Traditional methods struggle to accurately perform the classification task of complex PQDs under artificial empirical guidance. This paper proposes a multidimensional feature-driven ensemble model for the accurate classification of complex PQDs, which can complete the self-learning function from data. Unlike existing deep learning-based methods, this model considers both spatial features in the time-frequency domain and temporal relational features. Based on fully convolutional networks (FCN) and bidirectional gated recurrent unit (BiGRU), sub-modules suitable for multidimensional features mining are constructed separately. Meanwhile, the squeeze-and-excitation network (SENet) is introduced to complete the computation of the channel attention mechanism for each convolutional layer, which effectively improves the training efficiency and classification accuracy of the model. The proposed model has been thoroughly tested to validate its effectiveness on a synthetic dataset consisting of 71 different types of PQDs under varying signal-to-noise ratios (SNRs). Additionally, the model has been proven to be robust in the face of external factors such as DC offset, frequency variations, and phase jumps. To further demonstrate its reliability, the proposed model has been tested on real PQDs generated from an AC power source. Results from both simulation and experimentation have conclusively shown that the proposed method is superior to existing deep learning-based methods for the classification of complex PQDs.

Multiple prescription pattern recognition model based on Siamese network.

Prescription data is an important focus and breakthrough in the study of clinical treatment rules, and the complex multidimensional relationships between Traditional Chinese medicine (TCM) prescription data increase the difficulty of extracting knowledge from clinical data. This paper proposes a complex prescription recognition algorithm (MTCMC) based on the classification and matching of TCM prescriptions with classical prescriptions to identify the classical prescriptions contained in the prescriptions and provide a reference for mining TCM knowledge. The MTCMC algorithm first calculates the importance level of each drug in the complex prescriptions and determines the core prescription combinations of patients through the Analytic Hierarchy Process (AHP) combined with drug dosage. Secondly, a drug attribute tagging strategy was used to quantify the functional features of each drug in the core prescriptions; finally, a Bidirectional Long Short-Term Memory Network (BiLSTM) was used to extract the relational features of the core prescriptions, and a vector representation similarity matrix was constructed in combination with the Siamese network framework to calculate the similarity between the core prescriptions and the classical prescriptions. The experimental results show that the accuracy and F1 score of the prescription matching dataset constructed based on this paper reach 94.45% and 94.34% respectively, which is a significant improvement compared with the models of existing methods.

Identifying Cancer Disease Using Softmax-Feed Forward Recurrent Neural Classification

In today’s growing modern world environment, as human food activities are changing, it is affecting human health, thus leading to diseases like cancer. Cancer is a complex disease with many subtypes that affect human health without premature treatment and cause death. So the analysis of early diagnosis and prognosis of cancer studies can improve clinical management by analyzing various features of observation, which has become necessary to classify the type in cancer research. The research needs importance to organize the risk of the cancer patients based on data analysis to predict the result of premature treatment. This paper introduces a Maximal Region-Based Candidate Feature Selection (MRCFS) for early risk diagnosing using Soft-Max Feed Forward Neural Classification (SMF2NC) to solve the above problem. The predictive model is based on a different relational feature learning model, which is possessed to candidate selection to reduce the dimensionality. The redundant features are processed marginal weight rates for observing similar features’ variants and the absolute value. Softmax neural hidden layers are trained using the Sigmoid Activation Function (SAF) to create the logical condition for feed-forward layers. Further, the maximal features are introduced to invite a deep neural network constructed on the Feed Forward Recurrent Neural Network (FFRNN). The classifier produces higher classification accuracy than the previous methods and observes the cancer detection, which is recommended for early diagnosis.

Mental Filing Systems: A User's Guide

How seriously should we take the idea that the mind employs mental files? Goodman and Gray (2022) argue that mental filing—a thinker rationally treating her cognitive states as being about the same thing—can be explained without files. Instead, they argue that the standard commitments of mental file theory, as represented by Recanati’s indexical model, are better seen in terms of a relational representational feature of object representations, which in turn is based on the epistemic links a thinker bears to objects. This paper argues that this revision is misguided. Neither the representational property nor any basic role for epistemic links are needed for an adequate explanatory theory that makes use of the image of a mental filing system. A better alternative to the indexical model does posit files, albeit as causal-functional entities. This makes additional representational features redundant, and shows that epistemic links play a secondary role.

Who Deserves European Solidarity? How Recipient Characteristics Shaped Public Support for International Medical and Financial Aid during COVID-19

AbstractInternational solidarity is indispensable for coping with global crises; however, solidarity is frequently constrained by public opinion. Past research has examined who, on the donor side, is willing to support European and international aid. However, we know less about who, on the recipient side, is perceived to deserve solidarity. The article argues that potential donors consider situational circumstances and those relational features that link them to the recipients. Using factorial survey experiments, we analyse public support for international medical and financial aid in Germany during the COVID-19 pandemic. Our results show that recipient countries' situational need and control, as well as political community criteria, namely, group membership, adherence to shared values and reciprocity, played a crucial role in explaining public support for aid. Important policy implications result: on the donor side, fault-attribution frames matter; on the recipient side, honouring community norms is key to receiving aid.

R2-DDI: relation-aware feature refinement for drug-drug interaction prediction.

Precisely predicting the drug-drug interaction (DDI) is an important application and host research topic in drug discovery, especially for avoiding the adverse effect when using drug combination treatment for patients. Nowadays, machine learning and deep learning methods have achieved great success in DDI prediction. However, we notice that most of the works ignore the importance of the relation type when building the DDI prediction models. In this work, we propose a novel R$^2$-DDI framework, which introduces a relation-aware feature refinement module for drug representation learning. The relation feature is integrated into drug representation and refined in the framework. With the refinement features, we also incorporate the consistency training method to regularize the multi-branch predictions for better generalization. Through extensive experiments and studies, we demonstrate our R$^2$-DDI approach can significantly improve the DDI prediction performance over multiple real-world datasets and settings, and our method shows better generalization ability with the help of the feature refinement design.

NeuralGait

Brain health attracts more recent attention as the population ages. Smartphone-based gait sensing and analysis can help identify the risks of brain diseases in daily life for prevention. Existing gait analysis approaches mainly hand-craft temporal gait features or developing CNN-based feature extractors, but they are either prone to lose some inconspicuous pathological information or are only dedicated to a single brain disease screening. We discover that the relationship between gait segments can be used as a principle and generic indicator to quantify multiple pathological patterns. In this paper, we propose NeuralGait, a pervasive smartphone-cloud system that passively captures and analyzes principle gait segments relationship for brain health assessment. On the smartphone end, inertial gait data are collected while putting the smartphone in the pants pocket. We then craft local temporal-frequent gait domain features and develop a self-attention-based gait segment relationship encoder. Afterward, the domain features and relation features are fed to a scalable RiskNet in the cloud for brain health assessment. We also design a pathological hot update protocol to efficiently add new brain diseases in the RiskNet. NeuralGait is practical as it provides brain health assessment with no burden in daily life. In the experiment, we recruit 988 healthy people and 417 patients with a single or combination of PD, TBI, and stroke, and evaluate the brain health assessment using a set of specifically designed metrics including global accuracy, exact accuracy, sensitivity, and false alarm rate. We also demonstrate the generalization (e.g., analysis of feature effectiveness and model efficiency) and inclusiveness of NeuralGait.

A Prototype Network Enhanced Relation Semantic Representation for Few-shot Relation Extraction

Few-shot relation extraction is one of the current research focuses. The key to this research is to fully extract the relation semantic information through very little training data. Intuitively, raising the relation semantics awareness in sentences can improve the efficiency of the model to extract relation features to alleviate the overfitting problem in few-shot learning. Therefore, we propose an enhanced relation semantic feature model based on prototype network to extract relations from few-shot texts. Firstly, we design a multi-level embedding encoder with position information and Transformer, which uses local information in the text to enhance the relation semantics representation. Secondly, the encoded relation features are fed into the novel prototype network, which designs a method that utilizes query prototype-level attention to guide the extraction of supporting prototypes, thereby enhancing the prototypes representation to better classify the relations in query sentences. Finally, through experimental comparison and discussion, we prove and analyze the effectiveness of the proposed multi-level embedding encoder, and prototype-level attention can enhance the stability of the model. Furthermore, our model has substantial improvements over baseline methods.

From Dasein to the Posthuman, Or toward a Rhizomatic-Coexistence-in-an-Open-World

Abstract This article presents a brief mapping of Martin Heidegger’s theory of Dasein within the critical posthumanism framework. Adopting a cartographic analysis, it maps some similarities and differences between Heidegger’s fundamental ontology and critical posthumanism. It relies especially on Braidotti’s posthumanist arguments and theories to explain both the relational and transversal features of Dasein (in the posthumanist framework) and the weak anthropocentrist ones (in Heidegger’s fundamental ontology). In this paradigm, the Heideggerian concept of Dasein is situated both beyond and between classical humanism and weak anthropocentrism, given that Heidegger is deconstructing the metaphysical dialectic of binary oppositions (between human and animal), while remaining anchored in some type of weak anthropocentrism. Relying on a posthumanist and post-anthropocentrist methodology, the posthuman convergence twists the existential inquiry of Dasein and opens up a vitalist–materialist and immanent approach to the posthuman. This attracts the conceptual deterritorialization of Dasein (from its existential paradigm) and its reterritorialization in an open materialist process ontology of becoming, toward the posthuman, which is situated on the basis of an immanent, rhizomatic, transversal, and symbiotic movement given by our multiple belongingness to this material world, which relies on relational assemblages with human/non-human others, the environment, the Earth, and so on.