Recommendation Method for Contrastive Enhancement of Neighborhood Information

Knowledge graph can assist in improving recommendation performance and is widely applied in various person-alized recommendation domains.However,existing knowledge-aware recommendation methods face challenges such as weak user-item interaction supervisory signals and noise in the knowledge graph.To tackle these issues,this paper proposes a neighbor information contrast-enhanced recommendation method by adding subtle noise to construct contrast views and employing contrastive learning to strengthen supervisory signals and reduce knowledge noise.Specifically,first,this paper adopts heterogeneous propagation and knowledge-aware attention networks to obtain multi-order neighbor embedding of users and items,mining the high-order neighbor informa-tion of users and items.Next,in the neighbor information,this paper introduces weak noise following a uniform distribution to construct neighbor contrast views,effectively reducing the time overhead of view construction.This paper then performs contrastive learning between neighbor views to promote the uniformity of view information,adjusting the neighbor structure,and achieving the goal of reducing the knowledge noise in the knowledge graph.Finally,this paper introduces multi-task learning to mitigate the problem of weak supervisory signals.To validate the effectiveness of our method,experiments are conducted on theMovieLens-1M,MovieLens-20M,Book-Crossing,and Last-FM datasets.The results showthat compared to the best baselines,our method shows significant improvements in AUC and F1.

Contrastive Consistency and Attentive Complementarity for Deep Multi-View Subspace Clustering

Deep multi-view subspace clustering (DMVSC) based on self-expression has attracted increasing attention dueto its outstanding performance and nonlinear application. However, most existing methods neglect that viewprivatemeaningless information or noise may interfere with the learning of self-expression, which may lead to thedegeneration of clustering performance. In this paper, we propose a novel framework of Contrastive Consistencyand Attentive Complementarity (CCAC) for DMVsSC. CCAC aligns all the self-expressions of multiple viewsand fuses them based on their discrimination, so that it can effectively explore consistent and complementaryinformation for achieving precise clustering. Specifically, the view-specific self-expression is learned by a selfexpressionlayer embedded into the auto-encoder network for each view. To guarantee consistency across views andreduce the effect of view-private information or noise, we align all the view-specific self-expressions by contrastivelearning. The aligned self-expressions are assigned adaptive weights by channel attention mechanism according totheir discrimination. Then they are fused by convolution kernel to obtain consensus self-expression withmaximumcomplementarity ofmultiple views. Extensive experimental results on four benchmark datasets and one large-scaledataset of the CCAC method outperformother state-of-the-artmethods, demonstrating its clustering effectiveness.

Region-Aware Fashion Contrastive Learning for Unified Attribute Recognition and Composed Retrieval

Clothing attribute recognition has become an essential technology,which enables users to automatically identify the characteristics of clothes and search for clothing images with similar attributes.However,existing methods cannot recognize newly added attributes and may fail to capture region-level visual features.To address the aforementioned issues,a region-aware fashion contrastive language-image pre-training(RaF-CLIP)model was proposed.This model aligned cropped and segmented images with category and multiple fine-grained attribute texts,achieving the matching of fashion region and corresponding texts through contrastive learning.Clothing retrieval found suitable clothing based on the user-specified clothing categories and attributes,and to further improve the accuracy of retrieval,an attribute-guided composed network(AGCN)as an additional component on RaF-CLIP was introduced,specifically designed for composed image retrieval.This task aimed to modify the reference image based on textual expressions to retrieve the expected target.By adopting a transformer-based bidirectional attention and gating mechanism,it realized the fusion and selection of image features and attribute text features.Experimental results show that the proposed model achieves a mean precision of 0.6633 for attribute recognition tasks and a recall@10(recall@k is defined as the percentage of correct samples appearing in the top k retrieval results)of 39.18 for composed image retrieval task,satisfying user needs for freely searching for clothing through images and texts.

A Contrastive Analysis of English and Chinese Idioms About Color Words & Their Cultural Connotations

There are many idioms related to color words in English and Chinese.The use of color words in idioms adds beauty and vividness to the language.Due to the cultural differences,“color idioms”have gained different cultural connotations with the development of English and Chinese languages.It is of great significance to accurately understand and grasp the meanings and differences of color-related idioms in Chinese and English.This paper intends to analyze and expound the cultural connotations of English and Chinese idioms related to several widely used basic color words with the aim of helping English learners know and use the idioms about color words better.

Log Anomaly Detection Based on Hierarchical Graph Neural Network and Label Contrastive Coding

System logs are essential for detecting anomalies,querying faults,and tracing attacks.Because of the time-consuming and labor-intensive nature of manual system troubleshooting and anomaly detection,it cannot meet the actual needs.The implementation of automated log anomaly detection is a topic that demands urgent research.However,the prior work on processing log data is mainly one-dimensional and cannot profoundly learn the complex associations in log data.Meanwhile,there is a lack of attention to the utilization of log labels and usually relies on a large number of labels for detection.This paper proposes a novel and practical detection model named LCC-HGLog,the core of which is the conversion of log anomaly detection into a graph classification problem.Semantic temporal graphs(STG)are constructed by extracting the raw logs’execution sequences and template semantics.Then a unique graph classifier is used to better comprehend each STG’s semantic,sequential,and structural features.The classification model is trained jointly by graph classification loss and label contrastive loss.While achieving discriminability at the class-level,it increases the fine-grained identification at the instance-level,thus achieving detection performance even with a small amount of labeled data.We have conducted numerous experiments on real log datasets,showing that the proposed model outperforms the baseline methods and obtains the best all-around performance.Moreover,the detection performance degrades to less than 1%when only 10%of the labeled data is used.With 200 labeled samples,we can achieve the same or better detection results than the baseline methods.

Contrastive Learning for Blind Super-Resolution via A Distortion-Specific Network

Previous deep learning-based super-resolution(SR)methods rely on the assumption that the degradation process is predefined(e.g.,bicubic downsampling).Thus,their performance would suffer from deterioration if the real degradation is not consistent with the assumption.To deal with real-world scenarios,existing blind SR methods are committed to estimating both the degradation and the super-resolved image with an extra loss or iterative scheme.However,degradation estimation that requires more computation would result in limited SR performance due to the accumulated estimation errors.In this paper,we propose a contrastive regularization built upon contrastive learning to exploit both the information of blurry images and clear images as negative and positive samples,respectively.Contrastive regularization ensures that the restored image is pulled closer to the clear image and pushed far away from the blurry image in the representation space.Furthermore,instead of estimating the degradation,we extract global statistical prior information to capture the character of the distortion.Considering the coupling between the degradation and the low-resolution image,we embed the global prior into the distortion-specific SR network to make our method adaptive to the changes of distortions.We term our distortion-specific network with contrastive regularization as CRDNet.The extensive experiments on synthetic and realworld scenes demonstrate that our lightweight CRDNet surpasses state-of-the-art blind super-resolution approaches.

A Memory-Guided Anomaly Detection Model with Contrastive Learning for Multivariate Time Series

Some reconstruction-based anomaly detection models in multivariate time series have brought impressive performance advancements but suffer from weak generalization ability and a lack of anomaly identification.These limitations can result in the misjudgment of models,leading to a degradation in overall detection performance.This paper proposes a novel transformer-like anomaly detection model adopting a contrastive learning module and a memory block(CLME)to overcome the above limitations.The contrastive learning module tailored for time series data can learn the contextual relationships to generate temporal fine-grained representations.The memory block can record normal patterns of these representations through the utilization of attention-based addressing and reintegration mechanisms.These two modules together effectively alleviate the problem of generalization.Furthermore,this paper introduces a fusion anomaly detection strategy that comprehensively takes into account the residual and feature spaces.Such a strategy can enlarge the discrepancies between normal and abnormal data,which is more conducive to anomaly identification.The proposed CLME model not only efficiently enhances the generalization performance but also improves the ability of anomaly detection.To validate the efficacy of the proposed approach,extensive experiments are conducted on well-established benchmark datasets,including SWaT,PSM,WADI,and MSL.The results demonstrate outstanding performance,with F1 scores of 90.58%,94.83%,91.58%,and 91.75%,respectively.These findings affirm the superiority of the CLME model over existing stateof-the-art anomaly detection methodologies in terms of its ability to detect anomalies within complex datasets accurately.

Solving Geometry Problems via Feature Learning and Contrastive Learning of Multimodal Data

This paper presents an end-to-end deep learning method to solve geometry problems via feature learning and contrastive learning of multimodal data.A key challenge in solving geometry problems using deep learning is to automatically adapt to the task of understanding single-modal and multimodal problems.Existing methods either focus on single-modal ormultimodal problems,and they cannot fit each other.A general geometry problem solver shouldobviouslybe able toprocess variousmodalproblems at the same time.Inthispaper,a shared feature-learning model of multimodal data is adopted to learn the unified feature representation of text and image,which can solve the heterogeneity issue between multimodal geometry problems.A contrastive learning model of multimodal data enhances the semantic relevance betweenmultimodal features and maps them into a unified semantic space,which can effectively adapt to both single-modal and multimodal downstream tasks.Based on the feature extraction and fusion of multimodal data,a proposed geometry problem solver uses relation extraction,theorem reasoning,and problem solving to present solutions in a readable way.Experimental results show the effectiveness of the method.

Contrastive Clustering for Unsupervised Recognition of Interference Signals

Interference signals recognition plays an important role in anti-jamming communication.With the development of deep learning,many supervised interference signals recognition algorithms based on deep learning have emerged recently and show better performance than traditional recognition algorithms.However,there is no unsupervised interference signals recognition algorithm at present.In this paper,an unsupervised interference signals recognition method called double phases and double dimensions contrastive clustering(DDCC)is proposed.Specifically,in the first phase,four data augmentation strategies for interference signals are used in data-augmentation-based(DA-based)contrastive learning.In the second phase,the original dataset’s k-nearest neighbor set(KNNset)is designed in double dimensions contrastive learning.In addition,a dynamic entropy parameter strategy is proposed.The simulation experiments of 9 types of interference signals show that random cropping is the best one of the four data augmentation strategies;the feature dimensional contrastive learning in the second phase can improve the clustering purity;the dynamic entropy parameter strategy can improve the stability of DDCC effectively.The unsupervised interference signals recognition results of DDCC and five other deep clustering algorithms show that the clustering performance of DDCC is superior to other algorithms.In particular,the clustering purity of our method is above 92%,SCAN’s is 81%,and the other three methods’are below 71%when jammingnoise-ratio(JNR)is−5 dB.In addition,our method is close to the supervised learning algorithm.

CoLM^(2)S:Contrastive self‐supervised learning on attributed multiplex graph network with multi‐scale information

Contrastive self‐supervised representation learning on attributed graph networks with Graph Neural Networks has attracted considerable research interest recently.However,there are still two challenges.First,most of the real‐word system are multiple relations,where entities are linked by different types of relations,and each relation is a view of the graph network.Second,the rich multi‐scale information(structure‐level and feature‐level)of the graph network can be seen as self‐supervised signals,which are not fully exploited.A novel contrastive self‐supervised representation learning framework on attributed multiplex graph networks with multi‐scale(named CoLM^(2)S)information is presented in this study.It mainly contains two components:intra‐relation contrast learning and interrelation contrastive learning.Specifically,the contrastive self‐supervised representation learning framework on attributed single‐layer graph networks with multi‐scale information(CoLMS)framework with the graph convolutional network as encoder to capture the intra‐relation information with multi‐scale structure‐level and feature‐level selfsupervised signals is introduced first.The structure‐level information includes the edge structure and sub‐graph structure,and the feature‐level information represents the output of different graph convolutional layer.Second,according to the consensus assumption among inter‐relations,the CoLM^(2)S framework is proposed to jointly learn various graph relations in attributed multiplex graph network to achieve global consensus node embedding.The proposed method can fully distil the graph information.Extensive experiments on unsupervised node clustering and graph visualisation tasks demonstrate the effectiveness of our methods,and it outperforms existing competitive baselines.

Leveraging Vision-Language Pre-Trained Model and Contrastive Learning for Enhanced Multimodal Sentiment Analysis

Multimodal sentiment analysis is an essential area of research in artificial intelligence that combines multiple modes,such as text and image,to accurately assess sentiment.However,conventional approaches that rely on unimodal pre-trained models for feature extraction from each modality often overlook the intrinsic connections of semantic information between modalities.This limitation is attributed to their training on unimodal data,and necessitates the use of complex fusion mechanisms for sentiment analysis.In this study,we present a novel approach that combines a vision-language pre-trained model with a proposed multimodal contrastive learning method.Our approach harnesses the power of transfer learning by utilizing a vision-language pre-trained model to extract both visual and textual representations in a unified framework.We employ a Transformer architecture to integrate these representations,thereby enabling the capture of rich semantic infor-mation in image-text pairs.To further enhance the representation learning of these pairs,we introduce our proposed multimodal contrastive learning method,which leads to improved performance in sentiment analysis tasks.Our approach is evaluated through extensive experiments on two publicly accessible datasets,where we demonstrate its effectiveness.We achieve a significant improvement in sentiment analysis accuracy,indicating the supe-riority of our approach over existing techniques.These results highlight the potential of multimodal sentiment analysis and underscore the importance of considering the intrinsic semantic connections between modalities for accurate sentiment assessment.

Person Re-Identification with Model-Contrastive Federated Learning in Edge-Cloud Environment

Person re-identification(ReID)aims to recognize the same person in multiple images from different camera views.Training person ReID models are time-consuming and resource-intensive;thus,cloud computing is an appropriate model training solution.However,the required massive personal data for training contain private information with a significant risk of data leakage in cloud environments,leading to significant communication overheads.This paper proposes a federated person ReID method with model-contrastive learning(MOON)in an edge-cloud environment,named FRM.Specifically,based on federated partial averaging,MOON warmup is added to correct the local training of individual edge servers and improve the model’s effectiveness by calculating and back-propagating a model-contrastive loss,which represents the similarity between local and global models.In addition,we propose a lightweight person ReID network,named multi-branch combined depth space network(MB-CDNet),to reduce the computing resource usage of the edge device when training and testing the person ReID model.MB-CDNet is a multi-branch version of combined depth space network(CDNet).We add a part branch and a global branch on the basis of CDNet and introduce an attention pyramid to improve the performance of the model.The experimental results on open-access person ReID datasets demonstrate that FRM achieves better performance than existing baseline.

Multi-View Hybrid Contrastive Learning for Bundle Recommendation

Bundle recommendation aims to provide users with convenient one-stop solutions by recommending bundles of related items that cater to their diverse needs. However, previous research has neglected the interaction between bundle and item views and relied on simplistic methods for predicting user-bundle relationships. To address this limitation, we propose Hybrid Contrastive Learning for Bundle Recommendation (HCLBR). Our approach integrates unsupervised and supervised contrastive learning to enrich user and bundle representations, promoting diversity. By leveraging interconnected views of user-item and user-bundle nodes, HCLBR enhances representation learning for robust recommendations. Evaluation on four public datasets demonstrates the superior performance of HCLBR over state-of-the-art baselines. Our findings highlight the significance of leveraging contrastive learning and interconnected views in bundle recommendation, providing valuable insights for marketing strategies and recommendation system design.

Signal intensity changes of dentate nucleus on plain MR T1WI innasopharyngeal carcinoma patients after radiotherapy andmultiple injections of gadolinium-base contrast agent

Objective To observe changes of plain MR T1WI signal intensity of dentate nucleus in nasopharyngeal carcinoma patients after radiotherapy and multiple times of intravenous injection of gadolinium-based contrast agent(GBCA).Methods Fifty patients with pathologically confirmed nasopharyngeal carcinoma and received intensity-modulated radiotherapy were retrospectively enrolled as the nasopharyngeal carcinoma group,and 50 patients with other malignant tumors and without history of brain radiotherapy were retrospectively enrolled as the control group.All patients received yearly GBCA enhanced MR examinations for the nasopharynx or the head.T1WI signal intensities of the dentate nucleus and the pons on same plane were measured based on images in the year of confirmed diagnosis(recorded as the first year)and in the second to the fifth years.T1WI signal intensity ratio of year i(ranging from 1 to 5)was calculated with values of dentate nucleus divided by values of the pons(ΔSI i),while the percentage of relative changes of year j(ranging from 2 to 5)was calculated withΔSI j compared toΔSI 1(Rchange j).The values of these two parameters were compared,and the correlation ofΔSI and GBCA injection year-time was evaluated within each group.Results No significant difference of gender,age norΔSI 1 was found between groups(all P>0.05).The second to the fifth yearΔSI and Rchange in nasopharyngeal carcinoma group were all higher than those in control group(all P<0.05).Within both groups,ΔSI was positively correlated with GBCA injection year-time(both P<0.05).Conclusion Patients with nasopharyngeal carcinoma who underwent radiotherapy and multiple times of intravenous injection of GBCA tended to be found with gradually worsening GBCA deposition in dentate nucleus,for which radiotherapy might be a risk factor.

Optical design of a novel near-infrared phase contrast imaging(NI-PCI)diagnostic on the HL-2A tokamak

The optical design of near-infrared phase contrast imaging(NI-PCI)diagnosis on HL-2A is introduced in this paper.This scheme benefits from the great progress of near-infrared laser technology and is a broadening of traditional phase contrast technology.This diagnostic can work as a keen tool to measure plasma wavenumber spectra by inferring string-integrated plasma density fluctuations.Design of both the front optical path which is the path before the laser transmitting into the tokamak plasma and the rear optics which is the path after the laser passing through the plasma is detailed.The 1550 nm laser is chosen as the probe beam and highprecision optical components are designed to fit the laser beam,in which a phase plate with a 194-nm-deep silver groove is the key.Compared with the conventional 10.6μm laser-based PCI system on HL-2A,NI-PCI significantly overcomes the unwanted phase scintillation effect and promotes the measurement capability of high-wavenumber turbulence with an increased maximal measurable wavenumber from 15 cm^(-1)to 32.6 cm^(-1).

Near-Infrared Fluorescence Imaging Contrast Agents for Clinical Research: Limitations and Alternatives

Introduction: Near-infrared fluorescence imaging is a technique that will establish itself in the short term at the international level because it is recognized for its potential to improve the performance of surgical interventions, its moderate investment and operating costs and its portability. Although the technology is now mature, there is currently the problem of the availability of contrast agents to be injected IV. The aim of this methodology article is to propose an alternative solution to the need for contrast agents for clinical research, particularly in oncology. Methodology: They consist of coupling a fluorescent marker in the form of an NHS derivative, such as IR DYE manufactured in compliance with GMP, with therapeutic monoclonal antibodies having marketing authorization for molecular imaging. For a given antibody, the marking procedure must be the subject of a validation file on the final preparation filtered on a sterilizing membrane at 0.22 μm. Once the procedure has been validated, it would be unnecessary to repeat the tests before each clinical research examination. A check of the marking by thin-layer chromatography (TLC) and place it in a sample bank at +4˚C for 1 month of each injected formulation would be sufficient for additional tests if necessary. Conclusion: Molecular near-infrared fluorescence imaging is experiencing development, the process of which could be accelerated by greater availability of clinical contrast agents. Alternative solutions are therefore necessary to promote clinical research in this area. These methods must be shared to make it easier for researchers.

Attribution of Biases of Interhemispheric Temperature Contrast in CMIP6 Models

One of the basic characteristics of Earth's modern climate is that the Northern Hemisphere(NH) is climatologically warmer than the Southern Hemisphere(SH). Here, model performances of this basic state are examined using simulation results from 26 CMIP6 models. Results show that the CMIP6 models underestimate the contrast in interhemispheric surface temperatures on average(0.8 K for CMIP6 mean versus 1.4 K for reanalysis data mean), and that there is a large intermodel spread, ranging from -0.7 K to 2.3 K. A box model energy budget analysis shows that the contrast in interhemispheric shortwave absorption at the top of the atmosphere, the contrast in interhemispheric greenhouse trapping, and the crossequatorial northward ocean heat transport, are all underestimated in the multimodel mean. By examining the intermodel spread, we find intermodel biases can be tracked back to biases in midlatitude shortwave cloud forcing in AGCMs. Models with a weaker interhemispheric temperature contrast underestimate the shortwave cloud reflection in the SH but overestimate the shortwave cloud reflection in the NH, which are respectively due to underestimation of the cloud fraction over the SH extratropical ocean and overestimation of the cloud liquid water content over the NH extratropical continents.Models that underestimate the interhemispheric temperature contrast exhibit larger double ITCZ biases, characterized by excessive precipitation in the SH tropics. Although this intermodel spread does not account for the multimodel ensemble mean biases, it highlights that improving cloud simulation in AGCMs is essential for simulating the climate realistically in coupled models.

Laser speckle contrast imaging to predict the effect of temporary spinal cord stimulation in postherpetic neuralgia patients: A prospective observational study

Temporary spinal cord stimulation(tSCS)can effectively reduce the pain and severity of postherpetic neuralgia(PHN).However,there are no effective and objective methods for predicting the effects of tSCS on PHN.Laser speckle contrast imaging(LSCI)is frequently used in neurology to evaluate the effectiveness of treatment.To assess the accuracy of LSCI in predicting the impact of tSCS on PHN,14 adult patients receiving tSCS treatments for spinal nerve-innervated(C6-T2)PHN participated in this observational study.Visual analog scale(VAS)assessments and LSCI bloodflow images of the-ngers were recorded after the tSCS procedure.The results showed that the VAS scores of all patients decreased signi-cantly.Moreover,the bloodflow index(BFI)values were signi-cantly higher than they were before the procedure.Increased bloodflow and pain alleviation were positively correlated.The-ndings indicated that spinal nerve PHN(C6-T2)was signi-cantly reduced by tSCS.Pain alleviation by tSCS was positively correlated with increased bloodflow in the hand.The effect of tSCS on PHN may thus be predicted using an independent and consistent indicator such as LSCI.

Gastrointestinal contrast-enhanced ultrasonography for diagnosis and treatment of peptic ulcer in children

BACKGROUND The detection rate of peptic ulcer in children is improving,with development of diagnostic procedures.Gastroscopy is the gold standard for the diagnosis of peptic ulcer,but it is an invasive procedure.Gastrointestinal contrast-enhanced ultrasonography(CEUS)has the advantages of being painless,noninvasive,nonradioactive,easy to use,and safe.AIM To investigate the clinical value of CEUS for diagnosis and treatment of peptic ulcer in children.METHODS We investigated 43 children with digestive tract symptoms in our hospital from January 2021 to June 2022.All children were examined by routine ultrasound,gastrointestinal CEUS,and gastroscopy.The pathological results of gastroscopy were taken as the gold standard.Routine ultrasonography was performed before gastrointestinal CEUS.Conventional ultrasound showed the thickness of the gastroduodenal wall,gastric peristalsis,and the adjacent organs and tissues around the abdominal cavity.Gastrointestinal CEUS recorded the thickness of the gastroduodenal wall;the size,location and shape of the ulcer;gastric peristalsis;and adjacent organs and tissues around the abdominal cavity.The results of routine ultrasound and gastrointestinal ultrasound were compared with those of gastroscopy to evaluate the diagnostic results and coincidence rate of routine ultrasound and gastrointestinal CEUS.All children received informed consent from their guardians for CEUS.This study was reviewed and approved by the hospital medical ethics committee.RESULTS Among the 43 children,17(15 male,2 female)were diagnosed with peptic ulcer by gastroscopy.There were 26 children with nonpeptic ulcer.There were eight cases of peptic ulcer and 35 of nonpeptic ulcer diagnosed by conventional ultrasound.The diagnostic coincidence rate of peptic ulcer in children diagnosed by conventional ultrasound was 79.1%(34/43),which was significantly different from that of gastroscopy(P=0.033).It indicates that the coincidence rate of gastrointestinal contrast-enhanced ultrasound and gastroscope is low.Fifteen cases of peptic ulcer and 28 of nonpeptic ulcer were diagnosed by CEUS.The diagnostic coincidence rate of peptic ulcer in children was 95.3%(41/43).There was no significant difference between CEUS and gastroscopy(P=0.655).It indicates that the coincidence rate of gastrointestinal contrast-enhanced ultrasound and gastroscope is high.CONCLUSION Gastrointestinal CEUS has a high coincidence rate in the diagnosis of peptic ulcer in children,and can be used as a preliminary examination method.

A Second-Order Image Denoising Model for Contrast Preservation

In this work,we propose a second-order model for image denoising by employing a novel potential function recently developed in Zhu(J Sci Comput 88:46,2021)for the design of a regularization term.Due to this new second-order derivative based regularizer,the model is able to alleviate the staircase effect and preserve image contrast.The augmented Lagrangian method(ALM)is utilized to minimize the associated functional and convergence analysis is established for the proposed algorithm.Numerical experiments are presented to demonstrate the features of the proposed model.