Software aging refers to system performance degradation and failure due Aging-Related Bugs (ARBs) in long-running software systems. ARB prediction helps identify ARBs prevent aging. However, early project stages often lack data for model training. To address this, cross-project (CPAP) is proposed, where a trained using labeled source projects predict the present target project. In CPAP, distribution discrepancy class imbalance pose challenges, impacting performance. this paper, hybrid CPAP...

Detecting the gene sequence of virus strains from patients and classifying them into specific are very important to provide effective treatment. However, there significant barriers sharing strains' data in plaintext privacy concerns patients. Homomorphic encryption is a form that allows users calculate encrypted without decrypting it. Achieving highly accurate viral strain prediction while safeguarding user challenge. We develop secure multi-label classification method using homomorphic...

In long-running systems, the phenomenon of performance degradation and failure rate increase caused by Aging-Related Bugs (ARBs) is known as software aging. Because low presence reproducing difficulty ARBs, collecting enough training data to predict ARBs in a project not easy. Thus, cross-project ARB prediction has been proposed. There are two main challenges prediction, namely distribution differences severe class imbalance. As for first challenge, existing approaches only focus on...

Weakly supervised visual recognition using inexact supervision is a critical yet challenging learning problem. It significantly reduces human labeling costs and traditionally relies on multi-instance pseudo-labeling. This paper introduces WeakSAM solves the weakly-supervised object detection (WSOD) segmentation by utilizing pre-learned world knowledge contained in vision foundation model, i.e., Segment Anything Model (SAM). addresses two limitations traditional WSOD retraining, pseudo ground...

<title>Abstract</title> The oracle bone script, one of the most well-known ancient writing systems, plays a key role in study Chinese characters. To speed up digitization documents through automatic recognition, we propose recognition model called QROB (Quick Response Oracle Bone). Due to unique characteristics characters, only small portion has been translated, leading an imbalance sample distribution across different address this issue, use Inverse Image Frequency De-biasing method and...

In recent years, approaches for scene text recognition based on the attention mechanism have achieved amazing success. However, majority of are coupled, and ways adhere to concept locating most pertinent image regions. this paper, we present a language model with contrastive that is detached from standard encoder-decoder architecture. First, preliminary results obtained framework; second, perform two steps prediction in calculation weight image, not only find relevant area, but also look...

Effort-Aware Software Defect Prediction (EADP) ranks software modules according to the defect density of modules, which allows testers find more defects while reviewing a certain amount code, and allocates testing resources effectively. However, recently proposed CBS+ EASC methods tend rank with LOC (Lines Code) first. Therefore, there are less inspected when inspecting top 20% via EASC. Although two achieve high Precision@20% value, Recall@20% PofB@20% (Proportion found Bugs LOC) values...

Existing distributed video coding (DVC) frameworks use manually designed and optimized modules when encoding decoding video. Each module can set the appropriate parameters as much possible to achieve its independent optimization. But there is no connection between modules, overall end-to-end optimization not realized. Inspired by application of neural networks coding, we try implement DVC using networks. This article proposes an framework, which combines perform Wyner-Ziv (WZ) frames, while...

The ancient Chinese characters appear in various historical documents and poetry. People tend to use optical character recognition tools understand these uncommon characters. current text interface is restricted a limited set, such as GB2312-80 GB18010-2005 standard. However, the newest HanYu Dictionary contains over 55K characters, much more than commonly-used set. This work proposes compact deep network (HYD-CNet) composed of depthwise separable convolutional blocks co-ordinate attention...

Recognizing variations of text occurrences in scene photos is still difficult the present day. In recent years, performance recognition models based on attention mechanism has vastly increased. However, these typically focus recognizing image regions or visual that are significant. this paper, we a unique paradigm for named gated co-attention. Using our suggested model, and semantic may be jointly reasoned. Given features extracted by convolutional network language first step involves...

On behalf of the Organizing Committee, it is our pleasure to welcome you Fast Abstract Track 33nd International Symposium on Software Reliability Engineering (ISSRE 2022) in Charlotte, North Carolina, USA.

On behalf of the SSSML 2022 Organizing Committee, it is our pleasure to welcome you Workshop on Safety and Security Software Using Machine Learning in Cyber-physics System (SSSML), which co-located with ISSRE Charlotte, North Carolina, USA. With advancement information technology big data AI era, novel software cyber physical system (CPS) emerges. While such using machine learning helps develop applications CPS, also brings challenges safety concerns, especially safety, reliability security....

Binary code similarity detection for cross-platform is widely used in plagiarism detection, malware and vulnerability search, aiming to detect whether two binary functions over different platforms are similar. Existing cross-architecture approaches mainly rely on the approximate matching calculation of complex high-dimensional features, such as graph, which inevitably slow unsuitable large-scale applications. To solve this problem, we propose a novel approach based index table called CBSDI,...