As smartphones and mobile devices are rapidly becoming indispensable for many network users, malware has become a serious threat in the security privacy. Especially on popular Android platform, malicious apps hiding large number of normal apps, which makes detection more challenging. In this paper, we propose ML-based method that utilizes than 200 features extracted from both static analysis dynamic app detection. The comparison modeling results demonstrates deep learning technique is...

As smartphones and mobile devices are rapidly becoming indispensable for many network users, malware has become a serious threat in the security privacy. Especially on popular Android platform, malicious apps hiding large number of normal apps, which makes detection more challenging. In this paper, we propose ML-based method that utilizes than 200 features extracted from both static analysis dynamic app detection. The comparison modeling results demonstrates deep learning technique is...

This paper presents the open-source COREMU, a scalable and portable parallel emulation framework that decouples complexity of parallelizing full-system emulators from building mature sequential one. The key observation is CPU cores devices in current (and likely future) multiprocessors are loosely-coupled communicate through well-defined interfaces. Based on this observation, COREMU emulates multiple by creating instances existing emulators, uses thin library layer to handle inter-core...

Android smartphone users have been suffering from the security problems these years. There is a serious threat to network and privacy brought by mobile malware. In this paper, we use deep-learning-based method detect malware implement an automatic detection engine families of malicious applications. The results evaluation show that can 97% at 0.1% false positive rate (FPR) when detecting fine-grained families.

Recently, Android malware is spreading rapidly.Although static or dynamic analysis techniques for detecting can provide a comprehensive view, it still subjected to time consuming and high cost in deployment manual efforts.In this paper, we propose A3, an Automatic Analysis of malware, detect automatically.Unlike traditional reverse-engineering methods, A3 looks Command & Control (C&C) Server), monitors the sensitive API invoking, declares Intent-filter automatically.Then constructs...

Smartphones are rapidly becoming a necessity in our lives and Android is one of the most popular mobile operating systems. However, large number malicious applications hidden behind benign pose serious threat to platform. In this paper, we have proposed implemented FgDetector, an automatically malware detection tool, based on machine learning models. FgDetector can extract features from convert it into low-dimensional feature vector for training model detect whether application or not....

This paper presents the open-source COREMU, a scalable and portable parallel emulation framework that decouples complexity of parallelizing full-system emulators from building mature sequential one. The key observation is CPU cores devices in current (and likely future) multiprocessors are loosely-coupled communicate through well-defined interfaces. Based on this observation, COREMU emulates multiple by creating instances existing emulators, uses thin library layer to handle inter-core...

Android malware threats have recently become a real concern. The growing amount and diversity of these applications render conventional defenses largely ineffective. To fight against variants zero-day malware, this paper proposes DroidChain, detection method based on behavior chain model, which is composed typical processes apps. Using the method, we summarize four kinds models, including privacy leakage, SMS financial charge, installation privilege escalation. 1260 shows that accuracy reaches 81.8%.

The security issue of Android Smart Phones has been most concerned by the users these years. We propose a novel method to extract exhaustive features from applications and use deep-learning-based detect malicious applications. Then we implement an automatic detection engine, DeepDetector, Furthermore, model can identify fine-grained malware family at same time. conducted evaluation analysis with thousands benign public dataset. results show that DeepDetector 97% 0.1 % false positive rate...

Text classification tasks are indispensable in natural language processing. With the development of Internet technology, way people transmit information has changed from letters to Internet. increase amount information, manual data annotation is inefficient. After 2010, emergence deep learning methods brought text into an epoch-making stage. ReNN-> MLP-> RNN-> CNN-> Attention -> Transformer-> GNN and other gradually being developed known by everyone, which also shows that...

As many malicious apps have been distributed in Android markets, it is urgent to fix the vulnerabilities exploited by these and develop effective mitigation methods. In this paper, we identify a common vulnerability apps. This rooted an unprotected component, called "Activity". Utilizing vulnerability, app can stealthily cannily collect sensitive user data. To better understand issue, built "ActivityHijacker", that detect right moment hijack Activity component intercept user's password while...

Reposting is the basic and key behavior for information diffusion in online social networks. It would be beneficial to understand influence factors of reposting predict future status, which could practically applied breaking news detection, marketing, media researches so on. Existing analytics prediction approaches mainly focus on related original content publishers. However, diffuses by viral cascades instead single-source broadcast network, means some actually occurs propagators rather...

The recent proposal of learned index structures opens up a new perspective on how traditional range indexes can be optimized. However, the current assume data distribution is relatively static and access pattern uniform, while real-world scenarios consist skew query evolving data. In this paper, we demonstrate that missing consideration patterns dynamic notably hinders applicability indexes. To end, propose solutions for workloads (called Doraemon). improve latency queries, Doraemon augments...

The potential security problems of blockchain technology are constantly restricting the development process related industrial applications. cost deploying a system in real environment to conduct research on issues is relatively high, and analysis verification also destructive irreproducible. Therefore, based idea layered design, this paper proposes simulation platform. divided into four layers platform: consensus layer, network contract storage layer. In problem computing resource waste...

Nowadays, data is essential in several fields, such as science, finance, medicine, and transportation, which means its value continues to rise. Relational databases are vulnerable copyright threats when transmitted shared a carrier of data. The watermarking technique seen partial solution the problem securing ownership. However, most them currently restricted numerical attributes relational databases, limiting their versatility. Furthermore, they modify source large extent, failing keep...

Abstract In this study, peptide‐calcium chelate was screened from antler bone hydrolysate, and its bioactivity on MC3T3‐E1 cells chelating mechanism were investigated. vitro experiments showed that promoted the differentiation mineralization of cells. Subsequently, three novel calcium‐chelating peptides obtained hydrolysate using hydroxyapatite chromatography (HAC), Sephadex G‐25 gel filtration chromatography, reversed‐phase high‐performance liquid (RP‐HPLC). Meanwhile, work determined...

Inspired by hummingbirds and certain insects, flapping wing micro aerial vehicles (FWMAVs) exhibit potential energy efficiency maneuverability advantages. Among them, the bi-directional motor-driven tailless FWMAV with simple structure prevails in research, but it requires active pose control for hovering. In this paper, we employ deep reinforcement learning to train a low-level hovering strategy that directly maps drone’s state motor voltage outputs. To our knowledge, other FWMAVs both...

Tourism is an important method for the revitalization and utilization of agricultural cultural heritage, with farmers playing a pivotal role in development heritage tourism (AHT). The implementation AHT from lens farmer participation essential fostering sustainable growth heritage. Based on 257 questionnaires two villages using Anshan Nanguo pear cultivation systems, this study uses structural equation modeling (SEM) to explore mechanism influencing farmers’ willingness participate...