A key issue in collaborative software development is communication among developers. One modality of a commit message, which developers describe the changes they make repository. As such, messages serve as an "audit trail" by can understand how source code project has changed-and why. Hence, quality affects effectiveness Commit are often poor lack time and motivation to craft good message. Several automatic approaches have been proposed generate messages. However, these based on uncurated...

With the development of automatic driving technology, traffic sign detection has become a very important task. However, it is challenging task because complex scene and small size target. In recent years, number convolutional neural network (CNN) based object methods have brought great progress to detection. Considering still high false rate, as well time overhead computational overhead, effect not satisfactory. Therefore, we employ lightweight model YOLO v5 (You Only Look Once) our work...

Ultra-high molecular weight polyethylene (UHMWPE) fiber-reinforced concrete (UHMWPE-FRC) is a hard–soft multiphase hybrid composite with exceptional toughness and impact resistance compared to conventional concrete. However, its cutting characteristics processing performance have not been sufficiently investigated, potentially causing accelerated saw blade wear, higher energy consumption, poor quality, thus increasing project costs duration. In order intelligently evaluate the of diamond...

Software defect prediction is one of the important ways to guarantee quality software systems. Combining various algorithms in machine learning predict defects has become a hot topic current study. The paper uses datasets MDP as experimental research objects and takes ensemble focus construct model. With experimenting five different types analyzing features procedures, this discusses best algorithm which Random Forest through comparison. Then we utilize SMOTE over-sampling Resample methods...

Abstract Emotion recognition based on Electroencephalogram (EEG) has been applied in various fields, including human–computer interaction and healthcare. However, for the popular Valence-Arousal-Dominance emotion model, researchers often classify dimensions into high low categories, which cannot reflect subtle changes emotion. Furthermore, there are issues with design of EEG features efficiency transformer. To address these issues, we have designed TPRO-NET, a neural network that takes...

Convolution neural network is a depth learning algorithm that can automatically extract features. A face recognition method based on convolution and Fisher criterion brought up to resolve the difficulty of poor property networks under small samples. First, discriminant metric function added in cost error enhance classification network. Then features by utilizing modified are extracted. Finally, advantage support vector machine sample, nonlinear high dimension used classify extracted The...

In this paper, a robust watermarking algorithm  against geometric distortions and common signal processing is proposed. This algorithm based on two method to solve the synchronization of original images test in wavelet domain watermarking. One method, which estimates rotation scaling embedding template circle middle frequency DFT, translation extracting an invariant centroid from restricted area inside image. process, watermark embedded adaptively low band DWT domain, according conceal...

Assisted tracing is the process where human analyst vets and makes decisions concerning automated method's output. Current research reveals fallibility in this process, shows that often incorrect lead to inaccurate final trace matrix. To help enhance performance, we leverage tagging assisted tracing. Specifically, implement as a front-end feature allows analysts freely mark what they feel worth externalizing during We then carry out an experiment investigate practices of 28 student vetting...

Abstract Accurate and informative hand-object collision feedback is of vital importance for hand manipulation in virtual reality (VR). However, to our best knowledge, the movement performance fully-occluded confined VR spaces under visual still investigation. In this paper, we firstly studied effects several popular on performance. To test effects, conducted a within-subject user study ( n =18) using target-reaching task box. Results indicated that users had with see-through visualization,...

Although algorithms based on rough set (RS) theory can extract useful decision rules with the effectiveness in dealing inexact, uncertain or vague information, deterministic mechanism for description of error is very simple and generated by RS are often unstable have low classification accuracy. Neural networks (NN) considered most powerful classifier their rates robustness to noise. But NN usually require long time train huge amount data large databases lack explanation facilities...

A new method for feature extraction of acoustic targets based on manifold learning is proposed. The overcomes the shortcomings original algorithm by minimising impact noise and makes full use classification information to extract intrinsic features targets, thus improving accuracy recognition. Results experiments suggest proposed efficient feasible.

Grant-free random access (RA) utilizing massive multiple-input multiple-output (MIMO) technology has attracted considerable attention in recent years due to its potential enhance spectral efficiency. This paper introduces an innovative and advanced approach for the joint detection of users estimation channels grant-free RA. The incorporates two distinct preamble structures: single orthogonal (SOP) concatenated (COP). proposed algorithms make full use inherent quasiorthogonal characteristic...

In order to maximize the lifetime of wireless sensor networks, while ensuring monitoring quality for specific applications, we develop sleep-wakeup algorithms networks. Firstly, three-dimensional formulate virtual barrier coverage problems. Secondly, design and implementation nodes scheduling algorithm, FWP, covering time a single k-virtual barriers OBP, network k are presented. Finally, through comprehensive simulation, effectiveness is validated. The relationships among density barriers,...

JPEG compression can significantly impair the performance of adversarial face examples, which previous attacks on recognition (FR) have not adequately addressed. Considering this challenge, we propose a novel attack FR that aims to improve resistance examples against compression. Specifically, during iterative process generating interpolate into smaller size. Then utilize these interpolated create in next iteration. Subsequently, restore their original size by interpolating. Throughout...