Recommender System Combining Popularity and Novelty Based on One-Mode Projection of Weighted Bipartite Network

Personalized recommendation algorithms,which are effective means to solve information overload,are popular topics in current research.In this paper,a recommender system combining popularity and novelty(RSCPN)based on one-mode projection of weighted bipartite network is proposed.The edge between a user and item is weighted with the item’s rating,and we consider the difference in the ratings of different users for an item to obtain a reasonable method of measuring the similarity between users.RSCPN can be used in the same model for popularity and novelty recommendation by setting different parameter values and analyzing how a change in parameters affects the popularity and novelty of the recommender system.We verify and compare the accuracy,diversity and novelty of the proposed model with those of other models,and results show that RSCPN is feasible.

Smart contract token-based privacy-preserving access control system for industrial Internet of Things

Due to mobile Internet technology's rapid popularization,the Industrial Internet of Things(IIoT)can be seen everywhere in our daily lives.While IIoT brings us much convenience,a series of security and scalability issues related to permission operations rise to the surface during device communications.Hence,at present,a reliable and dynamic access control management system for IIoT is in urgent need.Up till now,numerous access control architectures have been proposed for IIoT.However,owing to centralized models and heterogeneous devices,security and scalability requirements still cannot be met.In this paper,we offer a smart contract token-based solution for decentralized access control in IIoT systems.Specifically,there are three smart contracts in our system,including the Token Issue Contract(TIC),User Register Contract(URC),and Manage Contract(MC).These three contracts collaboratively supervise and manage various events in IIoT environments.We also utilize the lightweight and post-quantum encryption algorithm-Nth-degree Truncated Polynomial Ring Units(NTRU)to preserve user privacy during the registration process.Subsequently,to evaluate our proposed architecture's performance,we build a prototype platform that connects to the local blockchain.Finally,experiment results show that our scheme has achieved secure and dynamic access control for the IIoT system compared with related research.

Liver Tumor Segmentation Based on Multi-Scale and Self-Attention Mechanism

Liver cancer has the second highest incidence rate among all types of malignant tumors,and currently,its diagnosis heavily depends on doctors’manual labeling of CT scan images,a process that is time-consuming and susceptible to subjective errors.To address the aforementioned issues,we propose an automatic segmentation model for liver and tumors called Res2Swin Unet,which is based on the Unet architecture.The model combines Attention-Res2 and Swin Transformer modules for liver and tumor segmentation,respectively.Attention-Res2 merges multiple feature map parts with an Attention gate via skip connections,while Swin Transformer captures long-range dependencies and models the input globally.And the model uses deep supervision and a hybrid loss function for faster convergence.On the LiTS2017 dataset,it achieves better segmentation performance than other models,with an average Dice coefficient of 97.0%for liver segmentation and 81.2%for tumor segmentation.

Blending Basic Shapes By C-Type Splines and Subdivision Scheme

In this article,we adopt the C-type spline of degree 2 to model and blend basic shapes including conics and circle arcs.The C-type spline belongs to theωB-spline category of splines that are capable of blending polynomial,trigonometric and hyperbolic functions.Commonly used basic shapes can be exactly represented by these types of splines.We derive explicit formulas for the convenience of modeling the basic curves.The entire blending curve is C^1-continuous.In comparison with the existing best blending method by rational G^2 splines,which are rational splines of degree 3,the proposed method allows simpler representation and blending of the basic curves,and it can represent numerous basic shapes including the hyperbolic types.We also design a subdivision method to generate blending curves;this method is precise for the basic curves and approximate for the blending sections.The subdivision process is efficient for modeling and rendering.It has also proven to be C^1-continuous by the asymptotically equivalent theory and the continuity of stationary subdivision method.In addition,we extend the proposed methods to cases involving the modeling and blending of basic surfaces.We provide many examples that illustrate the merits of our methods.

A Simulation Experiment of a Pipeline Based on Machine Learning for Neutral Hydrogen Intensity Mapping Surveys

We present a simulation experiment of a pipeline based on machine learning algorithms for neutral hydrogen(H I)intensity mapping(IM)surveys with different telescopes.The simulation is conducted on H I signals,foreground emission,thermal noise from instruments,strong radio frequency interference(s RFI),and mild RFI(m RFI).We apply the Mini-Batch K-Means algorithm to identify s RFI,and Adam algorithm to remove foregrounds and m RFI.Results show that there exists a threshold of the s RFI amplitudes above which the performance of our pipeline enhances greatly.In removing foregrounds and m RFI,the performance of our pipeline is shown to have little dependence on the apertures of telescopes.In addition,the results show that there are thresholds of the signal amplitudes from which the performance of our pipeline begins to change rapidly.We consider all these thresholds as the edges of the signal amplitude ranges in which our pipeline can function well.Our work,for the first time,explores the feasibility of applying machine learning algorithms in the pipeline of IM surveys,especially for large surveys with the next-generation telescopes.

Cooperative Rate Splitting Transmit Design for Full-Duplex-Enabled Multiple Multicast Communication Systems

This paper examines the performance of Full-Duplex Cooperative Rate Splitting(FD-CRS)with Simultaneous Wireless Information and Power Transfer(SWIPT)support in Multiple Input Single Output(MISO)networks.In a Rate Splitting Multiple Access(RSMA)multicast system with two local users and one remote user,the common data stream contains the needs of all users,and all users can decode the common data stream.Therefore,each user can receive some information that other users need,and local users with better channel conditions can use this information to further enhance the reception reliability and data rate of users with poor channel quality.Even using Cell-Center-Users(CCUs)as a cooperative relay to assist the transmission of common data can improve the average system speed.To maximize the minimum achievable rate,we optimize the beamforming vector of Base Station(BS),the common streamsplitting vector,the cooperative distributed beamvector and the strong user transmission power under the power budget constraints of BS and relay devices and the service quality requirements constraints of users.Since the whole problem is not convex,we cannot solve it directly.Therefore,we propose a low complexity algorithm based on Successive Convex Approximation(SCA)technology to find the optimal solution to the problemunder consideration.The simulation results show that FD C-RSMA has better gain andmore powerful than FD C-NOMA,HD C-RSMA,RSMA and NOMA.

A Pre-Selection-Based Ant Colony System for Integrated Resources Scheduling Problem at Marine Container Terminal

Marine container terminal(MCT)plays a key role in the marine intelligent transportation system and international logistics system.However,the efficiency of resource scheduling significantly influences the operation performance of MCT.To solve the practical resource scheduling problem(RSP)in MCT efficiently,this paper has contributions to both the problem model and the algorithm design.Firstly,in the problem model,different from most of the existing studies that only consider scheduling part of the resources in MCT,we propose a unified mathematical model for formulating an integrated RSP.The new integrated RSP model allocates and schedules multiple MCT resources simultaneously by taking the total cost minimization as the objective.Secondly,in the algorithm design,a pre-selection-based ant colony system(PACS)approach is proposed based on graphic structure solution representation and a pre-selection strategy.On the one hand,as the RSP can be formulated as the shortest path problem on the directed complete graph,the graphic structure is proposed to represent the solution encoding to consider multiple constraints and multiple factors of the RSP,which effectively avoids the generation of infeasible solutions.On the other hand,the pre-selection strategy aims to reduce the computational burden of PACS and to fast obtain a higher-quality solution.To evaluate the performance of the proposed novel PACS in solving the new integrated RSP model,a set of test cases with different sizes is conducted.Experimental results and comparisons show the effectiveness and efficiency of the PACS algorithm,which can significantly outperform other state-of-the-art algorithms.

Rapid Prototype Development Approach for Genetic Programming

Genetic Programming (GP) is an important approach to deal with complex problem analysis and modeling, and has been applied in a wide range of areas. The development of GP involves various aspects, including design of genetic operators, evolutionary controls and implementations of heuristic strategy, evaluations and other mechanisms. When designing genetic operators, it is necessary to consider the possible limitations of encoding methods of individuals. And when selecting evolutionary control strategies, it is also necessary to balance search efficiency and diversity based on representation characteristics as well as the problem itself. More importantly, all of these matters, among others, have to be implemented through tedious coding work. Therefore, GP development is both complex and time-consuming. To overcome some of these difficulties that hinder the enhancement of GP development efficiency, we explore the feasibility of mutual assistance among GP variants, and then propose a rapid GP prototyping development method based on πGrammatical Evolution (πGE). It is demonstrated through regression analysis experiments that not only is this method beneficial for the GP developers to get rid of some tedious implementations, but also enables them to concentrate on the essence of the referred problem, such as individual representation, decoding means and evaluation. Additionally, it provides new insights into the roles of individual delineations in phenotypes and semantic research of individuals.

Performance Analysis of Intelligent Reflecting Surface Assisted Wireless Communication System

In this paper,we investigate the end-to-end performance of intelligent reflecting surface(IRS)-assisted wireless communication systems.We consider a system in which an IRS is deployed on a uniform planar array(UPA)configuration,including a large number of reflecting elements,where the transmitters and receivers are only equipped with a single antenna.Our objective is to analytically obtain the achievable ergodic rate,outage probability,and bit error rate(BER)of the system.Furthermore,to maximize the system’s signal-to-noise ratio(SNR),we design the phase shift of each reflecting element and derive the optimal reflection phase of the IRS based on the channel state information(CSI).We also derive the exact expression of the SNR probability density function(p.d.f.)and show that it follows a non-central Chi-square distribution.Using the p.d.f.,we then derive the theoretical results of the achievable rate,outage probability,and BER.The accuracy of the obtained theoretical results is also verified through numerical simulation.Itwas shown that the achievable rate,outage probability,and BER could be improved by increasing the number of reflecting elements and choosing an appropriate SNR regime.Furthermore,we also find that the IRS-assisted communication system achieves better performance than the existing end-to-end wireless communication.

Dynamic performance and energy efficiency of reflective and insulative composite coating on building exterior wall

Reflective and insulative composite coatings are a new energy-saving material with high solar reflectance and extremely low thermal conductivity for buildings.The optimization and impact of high solar reflectance and low thermal conductivity on the insulating capacity of walls remain uncertain.This work investigates the dynamic thermal performance and energy efficiency of a reflective and insulative composite coating in regions with hot summer and warm winter.A simplified thermal resistance-heat capacitance model of an exterior building wall is established to predict thermal performance.The dynamic temperature and heat flow of the wall are predicted to reduce heat loss through the interior surface of the wall and compared to the conventional coating.The specific impact of the thermal conductivity and solar reflectance of the coating on the heat loss is further investigated to minimize heat loss of the wall.This research shows that the composite coating shows better performance on adjusting outdoor climate change than the other coating.Compared with cement,it reduces the maximum temperature of the exterior surface of the wall by 7.45°C,and the heat loss through the interior surface of the wall by 38%.The heat loss is reduced with the increase of solar reflectance and the reduction of thermal conductivity.The results can provide a useful reference and guidance for the application of reflective and insulative composite coating on building exterior wall to promote their energy-saving use on building envelopes.

A Position-Aware Transformer for Image Captioning

Image captioning aims to generate a corresponding description of an image.In recent years,neural encoder-decodermodels have been the dominant approaches,in which the Convolutional Neural Network(CNN)and Long Short TermMemory(LSTM)are used to translate an image into a natural language description.Among these approaches,the visual attention mechanisms are widely used to enable deeper image understanding through fine-grained analysis and even multiple steps of reasoning.However,most conventional visual attention mechanisms are based on high-level image features,ignoring the effects of other image features,and giving insufficient consideration to the relative positions between image features.In this work,we propose a Position-Aware Transformer model with image-feature attention and position-aware attention mechanisms for the above problems.The image-feature attention firstly extracts multi-level features by using Feature Pyramid Network(FPN),then utilizes the scaled-dot-product to fuse these features,which enables our model to detect objects of different scales in the image more effectivelywithout increasing parameters.In the position-aware attentionmechanism,the relative positions between image features are obtained at first,afterwards the relative positions are incorporated into the original image features to generate captions more accurately.Experiments are carried out on the MSCOCO dataset and our approach achieves competitive BLEU-4,METEOR,ROUGE-L,CIDEr scores compared with some state-of-the-art approaches,demonstrating the effectiveness of our approach.

N-gram MalGAN:Evading machine learning detection via feature n-gram

In recent years,many adversarial malware examples with different feature strategies,especially GAN and its variants,have been introduced to handle the security threats,e.g.,evading the detection of machine learning detectors.However,these solutions still suffer from problems of complicated deployment or long running time.In this paper,we propose an n-gram MalGAN method to solve these problems.We borrow the idea of n-gram from the Natural Language Processing(NLP)area to expand feature sources for adversarial malware examples in MalGAN.Generally,the n-gram MalGAN obtains the feature vector directly from the hexadecimal bytecodes of the executable file.It can be implemented easily and conveniently with a simple program language(e.g.,C++),with no need for any prior knowledge of the executable file or any professional feature extraction tools.These features are functionally independent and thus can be added to the non-functional area of the malicious program to maintain its original executability.In this way,the n-gram could make the adversarial attack easier and more convenient.Experimental results show that the evasion rate of the n-gram MalGAN is at least 88.58%to attack different machine learning algorithms under an appropriate group rate,growing to even 100%for the Random Forest algorithm.

Fairness-Aware Harvested Energy Efficiency Algorithm for IRS-Aided Intelligent Sensor Networks with SWIPT

In this paper,a novel fairness-aware harvested energy efficiency-based green transmission scheme for wireless information and power transfer(SWIPT)aided sensor networks is developed for active beamforming of multiantenna transmitter and passive beamforming at intelligent reflecting surfaces(IRS).By optimizing the active beamformer assignment at the transmitter in conjunction with the passive beamformer assignment at the IRS,we aimtomaximize the minimumharvested energy efficiency among all the energy receivers(ER)where information receivers(IR)are bound to the signal-interference-noise-ratio(SINR)and the maximum transmitted power of the transmitter.To handle the non-convex problem,both semi-definite relaxation(SDR)and block coordinate descent technologies are exploited.Then,the original problem is transformed into two convex sub-problems which can be solved via semidefinite programming.Numerical simulation results demonstrate that the IRS and energy beamformer settings in this paper provide greater system gain than the traditional experimental setting,thereby improving the fairness-aware harvested energy efficiency of the ER.

A Data Consistency Insurance Method for Smart Contract

As one of the major threats to the current DeFi(Decentralized Finance)ecosystem,reentrant attack induces data inconsistency of the victim smart contract,enabling attackers to steal on-chain assets from DeFi projects,which could terribly do harm to the confidence of the blockchain investors.However,protecting DeFi projects from the reentrant attack is very difficult,since generating a call loop within the highly automatic DeFi ecosystem could be very practicable.Existing researchers mainly focus on the detection of the reentrant vulnerabilities in the code testing,and no method could promise the non-existent of reentrant vulnerabilities.In this paper,we introduce the database lock mechanism to isolate the correlated smart contract states from other operations in the same contract,so that we can prevent the attackers from abusing the inconsistent smart contract state.Compared to the existing resolutions of front-running,code audit,andmodifier,our method guarantees protection resultswith better flexibility.And we further evaluate our method on a number of de facto reentrant attacks observed from Etherscan.The results prove that our method could efficiently prevent the reentrant attack with less running cost.

A Conditionally Anonymous Linkable Ring Signature for Blockchain Privacy Protection

In recent years,the issue of preserving the privacy of parties involved in blockchain transactions has garnered significant attention.To ensure privacy protection for both sides of the transaction,many researchers are using ring signature technology instead of the original signature technology.However,in practice,identifying the signer of an illegal blockchain transaction once it has been placed on the chain necessitates a signature technique that offers conditional anonymity.Some illegals can conduct illegal transactions and evade the lawusing ring signatures,which offer perfect anonymity.This paper firstly constructs a conditionally anonymous linkable ring signature using the Diffie-Hellman key exchange protocol and the Elliptic Curve Discrete Logarithm,which offers a non-interactive process for finding the signer of a ring signature in a specific case.Secondly,this paper’s proposed scheme is proven correct and secure under Elliptic Curve Discrete Logarithm Assumptions.Lastly,compared to previous constructions,the scheme presented in this paper provides a non-interactive,efficient,and secure confirmation process.In addition,this paper presents the implementation of the proposed scheme on a personal computer,where the confirmation process takes only 2,16,and 24ms for ring sizes of 4,24 and 48,respectively,and the confirmation process can be combined with a smart contract on the blockchain with a tested millisecond level of running efficiency.In conclusion,the proposed scheme offers a solution to the challenge of identifying the signer of an illegal blockchain transaction,making it an essential contribution to the field.

Attentive Neighborhood Feature Augmentation for Semi-supervised Learning

Recent state-of-the-art semi-supervised learning(SSL)methods usually use data augmentations as core components.Such methods,however,are limited to simple transformations such as the augmentations under the instance’s naive representations or the augmentations under the instance’s semantic representations.To tackle this problem,we offer a unique insight into data augmentations and propose a novel data-augmentation-based semi-supervised learning method,called Attentive Neighborhood Feature Aug-mentation(ANFA).The motivation of our method lies in the observation that the relationship between the given feature and its neighborhood may contribute to constructing more reliable transformations for the data,and further facilitating the classifier to distinguish the ambiguous features from the low-dense regions.Specially,we first project the labeled and unlabeled data points into an embedding space and then construct a neighbor graph that serves as a similarity measure based on the similar representations in the embedding space.Then,we employ an attention mechanism to transform the target features into augmented ones based on the neighbor graph.Finally,we formulate a novel semi-supervised loss by encouraging the predictions of the interpolations of augmented features to be consistent with the corresponding interpolations of the predictions of the target features.We carried out exper-iments on SVHN and CIFAR-10 benchmark datasets and the experimental results demonstrate that our method outperforms the state-of-the-art methods when the number of labeled examples is limited.

Real-Time Spammers Detection Based on Metadata Features with Machine Learning

Spammer detection is to identify and block malicious activities performing users.Such users should be identified and terminated from social media to keep the social media process organic and to maintain the integrity of online social spaces.Previous research aimed to find spammers based on hybrid approaches of graph mining,posted content,and metadata,using small and manually labeled datasets.However,such hybrid approaches are unscalable,not robust,particular dataset dependent,and require numerous parameters,complex graphs,and natural language processing(NLP)resources to make decisions,which makes spammer detection impractical for real-time detection.For example,graph mining requires neighbors’information,posted content-based approaches require multiple tweets from user profiles,then NLP resources to make decisions that are not applicable in a real-time environment.To fill the gap,firstly,we propose a REal-time Metadata based Spammer detection(REMS)model based on only metadata features to identify spammers,which takes the least number of parameters and provides adequate results.REMS is a scalable and robust model that uses only 19 metadata features of Twitter users to induce 73.81%F1-Score classification accuracy using a balanced training dataset(50%spam and 50%genuine users).The 19 features are 8 original and 11 derived features from the original features of Twitter users,identified with extensive experiments and analysis.Secondly,we present the largest and most diverse dataset of published research,comprising 211 K spam users and 1 million genuine users.The diversity of the dataset can be measured as it comprises users who posted 2.1 million Tweets on seven topics(100 hashtags)from 6 different geographical locations.The REMS’s superior classification performance with multiple machine and deep learning methods indicates that only metadata features have the potential to identify spammers rather than focusing on volatile posted content and complex graph structures.Dataset and REMS’s codes are available on GitHub(www.github.com/mhadnanali/REMS).

Energy consumption dynamic prediction for HVAC systems based on feature clustering deconstruction and model training adaptation

The prediction of building energy consumption offers essential technical support for intelligent operation and maintenance of buildings,promoting energy conservation and low-carbon control.This paper focused on the energy consumption of heating,ventilation and air conditioning(HVAC)systems operating under various modes across different seasons.We constructed multi-attribute and high-dimensional clustering vectors that encompass indoor and outdoor environmental parameters,along with historical energy consumption data.To enhance the K-means algorithm,we employed statistical feature extraction and dimensional normalization(SFEDN)to facilitate data clustering and deconstruction.This method,combined with the gated recurrent unit(GRU)prediction model employing adaptive training based on the Particle Swarm Optimization algorithm,was evaluated for robustness and stability through k-fold cross-validation.Within the clustering-based modeling framework,optimal submodels were configured based on the statistical features of historical 24-hour data to achieve dynamic prediction using multiple models.The dynamic prediction models with SFEDN cluster showed a 11.9%reduction in root mean square error(RMSE)compared to static prediction,achieving a coefficient of determination(R2)of 0.890 and a mean absolute percentage error(MAPE)reduction of 19.9%.When compared to dynamic prediction based on single-attribute of HVAC systems energy consumption clustering modeling,RMSE decreased by 12.6%,R2 increased by 4.0%,and MAPE decreased by 26.3%.The dynamic prediction performance demonstrated that the SFEDN clustering method surpasses conventional clustering method,and multi-attribute clustering modeling outperforms single-attribute modeling.

Using amino acid features to identify the pathogenicity of influenza B virus

Background:Influenza B virus can cause epidemics with high pathogenicity, so it poses a serious threat to public health. A feature representation algorithm is proposed in this paper to identify the pathogenicity phenotype of influenza B virus.Methods:The dataset included all 11 influenza virus proteins encoded in eight genome segments of 1724 strains. Two types of features were hierarchically used to build the prediction model. Amino acid features were directly delivered from 67 feature descriptors and input into the random forest classifier to output informative features about the class label and probabilistic prediction. The sequential forward search strategy was used to optimize the informative features. The final features for each strain had low dimensions and included knowledge from different perspectives, which were used to build the machine learning model for pathogenicity identification.Results:The 40 signature positions were achieved by entropy screening. Mutations at position 135 of the hemagglutinin protein had the highest entropy value (1.06). After the informative features were directly generated from the 67 random forest models, the dimensions for class and probabilistic features were optimized as 4 and 3, respectively. The optimal class features had a maximum accuracy of 94.2% and a maximum Matthews correlation coefficient of 88.4%, while the optimal probabilistic features had a maximum accuracy of 94.1% and a maximum Matthews correlation coefficient of 88.2%. The optimized features outperformed the original informative features and amino acid features from individual descriptors. The sequential forward search strategy had better performance than the classical ensemble method.Conclusions:The optimized informative features had the best performance and were used to build a predictive model so as to identify the phenotype of influenza B virus with high pathogenicity and provide early risk warning for disease control.

Deep Broad Learning for Emotion Classification in Textual Conversations

Emotion classification in textual conversations focuses on classifying the emotion of each utterance from textual conversations.It is becoming one of the most important tasks for natural language processing in recent years.However,it is a challenging task for machines to conduct emotion classification in textual conversations because emotions rely heavily on textual context.To address the challenge,we propose a method to classify emotion in textual conversations,by integrating the advantages of deep learning and broad learning,namely DBL.It aims to provide a more effective solution to capture local contextual information(i.e.,utterance-level)in an utterance,as well as global contextual information(i.e.,speaker-level)in a conversation,based on Convolutional Neural Network(CNN),Bidirectional Long Short-Term Memory(Bi-LSTM),and broad learning.Extensive experiments have been conducted on three public textual conversation datasets,which show that the context in both utterance-level and speaker-level is consistently beneficial to the performance of emotion classification.In addition,the results show that our proposed method outperforms the baseline methods on most of the testing datasets in weighted-average F1.