Chronic Obstructive Pulmonary Disease (COPD) is a prevalent chronic pulmonary condition that affects hundreds of millions people all over the world. Many COPD patients got readmitted to hospital within 30 days after discharge due various reasons. Such readmission can usually be avoided if additional attention paid with high risk and appropriate actions are taken. This makes early prediction an important problem. The goal this paper conduct systematic study on developing different types...

Recently several different deep learning architectures have been proposed that take a string of characters as the raw input signal and automatically derive features for text classification. Few studies are available compare effectiveness these approaches character based classification with each other. In this paper we perform such an empirical comparison important cybersecurity problem DGA detection: classifying domain names either benign vs. produced by malware (i.e., Domain Generation...

Networks, such as social networks, biochemical and protein-protein interaction networks are ubiquitous in the real world. Network representation learning aims to embed nodes a network low-dimensional, dense, real-valued vectors, facilitate downstream analysis. The existing embedding methods commonly endeavor capture structure information network, but lack of consideration subsequent tasks synergies between these tasks, which equal importance for desirable representations. To address this...

Abstract Data mining is a process to extract unknown, hidden, and potentially useful information from data. But the problem of data island makes it arduous for people collect analyze scattered data, there also privacy security issue when A collaboratively decentralized approach called federated learning unites multiple participants generate shareable global optimal model keeps privacy‐sensitive on local devices, which may bring great hope us solving problems protection. Though has been...

Recently, multi-task learning (MTL) has been extensively studied for various face processing tasks including detection, landmarks localization, pose estimation and gender recognition, which endeavors to train a better model by exploiting the synergy among related tasks. However, raw dataset used training often contains sensitive private information, can be maliciously recovered carefully analyzing outputs. To address this problem, we propose novel privacy-preserving approach, utilizes...

Real-world networks are composed of diverse interacting and evolving entities, while most existing researches simply characterize them as particular static networks, without consideration the evolution trend in dynamic networks. Recently, significant progresses tracking properties have been made, which exploit changes entities links network to devise embedding techniques. Compared widely proposed methods, endeavors encode nodes low-dimensional dense representations that effectively preserve...

For an explanation of a deep learning model to be effective, it must provide both insight into and suggest corresponding action in order achieve some objective. Too often, the litany proposed explainable methods stop at first step, providing practitioners with model, but no way act on it. In this paper, we propose contextual decomposition penalization (CDEP), method which enables leverage existing increase predictive accuracy models. particular, when shown that has incorrectly assigned...

Salmonella enterica serovar Typhimurium (S. Typhimurium), a foodborne pathogen that poses significant public health risks to humans and animals, presents formidable challenge due its antibiotic resistance. This study explores the potential of Lactobacillus acidophilus (L. 1.3251) probiotics as an alternative strategy combat resistance associated with S. infection. In this investigation, twenty-four BALB/c mice were assigned four groups: non-infected, non-treated group (CNG); infected, (CPG);...

Fermented soybean meal (FSM), which has lower anti-nutritional factors and higher active enzyme, probiotic oligosaccharide contents than its unfermented form, been reported to improve the feeding value of meal, hence, growth performance piglets. However, whether FSM can affect bacterial metabolites in large intestine piglets remains unknown. This study supplemented wet-FSM (WFSM) or dry-FSM (DFSM) (5% dry matter basis) diet investigated effects on carbon nitrogen metabolism piglets'...

Fermentation using appropriate microorganisms can remove anti-nutritional factors such as protease inhibitor, thus improving the feeding value of soybean meal, a major feed ingredient in pig diet. Two vitro experiments were conducted to obtain conditions for fermentation meal. Experiment 1 consisted eight treatments from combinations two levels three (Bacillus subtilis KC 101, Saccharomyces cerevisiae JM 102 and Bacillus lactis RG 103). broth was mixed with meal 18 L bucket incubated under...

Multi-task learning (MTL) is a paradigm which can improve generalization performance by transferring knowledge among multiple tasks. Traditional collaborative filtering recommendation methods suffer from cold start, sparsity and scalability problems. The latest research has shown that applying side information of graph not only solve the problems above, but also accuracy recommendation. However, existing multi-task for enhanced expose obvious issues disclosing private training samples. In...

In the era of big data, data has gradually become an important productivity driving social progress. Accelerating development and sharing resources is inherent requirement government transformation. Named entity recognition a wide range applications in fields information extraction retrieval, so its research great significance. Due to complexity Chinese structure lack mature domestic corpora, on named faces enormous challenges. Based analysis actual characteristics entities text, this paper...

Machine learning lies at the heart of new possibilities for scientific discovery, knowledge generation, and artificial intelligence. Its potential benefits to these fields requires going beyond predictive accuracy focusing on interpretability. In particular, many problems require interpretations in a domain-specific interpretable feature space (e.g. frequency domain) whereas attributions raw features pixel space) may be unintelligible or even misleading. To address this challenge, we propose...

Smart contracts are commonly used to build finance-related decentralized applications. If a smart contract vulnerability is exploited by an attacker, the owner may suffer financial losses. We focus on particular class of vulnerabilities that require specific sequence multiple transactions trigger, which we call multi-transaction vulnerabilities. Due combinatorial explosion problem caused huge number possible transaction sequences, efficiency and scalability for existing security analyzers...