다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

コンピュータ・サイエンス : ACM computing surveys

The Internet of Things (IoT) integrates billions of smart devices that can communicate with one another with minimal human intervention. IoT is one of the fastest developing fields in the history of computing, with an estimated 50 billion devices by the end of 2020. However, the crosscutting nature of IoT systems and the multidisciplinary components involved in the deployment of such systems have introduced new security challenges. Implementing security measures, such as encryption, authentication, access control, network and application security for IoT devices and their inherent vulnerabilities is ineffective. Therefore, existing security methods should be enhanced to effectively secure the IoT ecosystem. Machine learning and deep learning (ML/DL) have advanced considerably over the last few years, and machine intelligence has transitioned from laboratory novelty to practical machinery in several important applications. Consequently, ML/DL methods are important in transforming the security of IoT systems from merely facilitating secure communication between devices to security-based intelligence systems. The goal of this work is to provide a comprehensive survey of ML methods and recent advances in DL methods that can be used to develop enhanced security methods for IoT systems. IoT security threats that are related to inherent or newly introduced threats are presented, and various potential IoT system attack surfaces and the possible threats related to each surface are discussed. We then thoroughly review ML/DL methods for IoT security and present the opportunities, advantages and shortcomings of each method. We discuss the opportunities and challenges involved in applying ML/DL to IoT security. These opportunities and challenges can serve as potential future research directions.

A Survey of Machine and Deep Learning Methods for Internet of Things (IoT) Security

Smartphones have become pervasive due to the availability of office applications, Internet, games, vehicle guidance using location-based services apart from conventional services such as voice calls, SMSes, and multimedia services. Android devices have gained huge market share due to the open architecture of Android and the popularity of its application programming interface (APIs) in the developer community. Increased popularity of the Android devices and associated monetary benefits attracted the malware developers, resulting in big rise of the Android malware apps between 2010 and 2014. Academic researchers and commercial antimalware companies have realized that the conventional signature-based and static analysis methods are vulnerable. In particular, the prevalent stealth techniques, such as encryption, code transformation, and environment-aware approaches, are capable of generating variants of known malware. This has led to the use of behavior-, anomaly-, and dynamic-analysis-based methods. Since a single approach may be ineffective against the advanced techniques, multiple complementary approaches can be used in tandem for effective malware detection. The existing reviews extensively cover the smartphone OS security. However, we believe that the security of Android, with particular focus on malware growth, study of antianalysis techniques, and existing detection methodologies, needs an extensive coverage. In this survey, we discuss the Android security enforcement mechanisms, threats to the existing security enforcements and related issues, malware growth timeline between 2010 and 2014, and stealth techniques employed by the malware authors, in addition to the existing detection methods. This review gives an insight into the strengths and shortcomings of the known research methodologies and provides a platform, to the researchers and practitioners, toward proposing the next-generation Android security, analysis, and malware detection techniques.

/pdf/android-security-a-survey-of-issues-malware-penetration-and-2mhwbg9szu.pdf

Android Security: A Survey of Issues, Malware Penetration, and Defenses

Android Smartphone popularity has increased malware threats forcing security researchers and AntiVirus (AV) industry to carve out smart methods to defend Smartphone against malicious apps. Robust signature based solutions to mitigate threats become necessary to protect the Smartphone and confidential user data. In this paper we present AndroSimilar, a robust approach which generates signature by extracting statistically improbable features, to detect malicious Android apps. Proposed method is effective against code obfuscation and repackaging, widely used techniques to evade AV signature and to propagate unseen variants of known malware. AndroSimilar is a syntactic foot-printing mechanism that finds regions of statistical similarity with known malware to detect those unknown, zero day samples. Syntactic file similarity of whole file is considered instead of just opcodes for faster detection compared to known fuzzy hashing approaches. Results demonstrate robust detection of variants of known malware families. Proposed approach can be refined to deploy as Smartphone AV.

AndroSimilar: robust statistical feature signature for Android malware detection

The motivation behind Vehicular Ad Hoc Networks (VANETs) is to improve traffic safety and driving efficiency. VANET applications operate on the principle of periodic exchange of messages between nodes. However, a malicious node may transmit inaccurate messages to trigger inevitable situations. Each transmitted packet contains the status of sender like its identity, position and time of sending the packet in addition to safety message. A misbehaving node may tamper with any information present in the propagated packet. Fake messages may be created by attacker node itself or it may force another node to create fake messages. In this paper, we present a machine learning approach to classify multiple misbehaviors in VANET using concrete and behavioral features of each node that sends safety packets. A security framework is designed to differentiate a malicious node from legitimate node. We implement various types of misbehaviors in VANET by tampering information present in the propagated packet. These misbehaviors are classified based upon multifarious features like speed-deviation of node, received signal strength (RSS), number of packets delivered, dropped packets etc. Two types of classification accuracies are measured : Binary and Multi-Class. In Binary classification, all types of misbehaviors are considered to be in a single “misbehavior” class whereas, Multi-class classification is able to categorize misbehaviors into particular misbehaving classes. Features of packet sending nodes are extracted by performing experiments in NCTUns-5.0 simulator with different simulation scenario (varying the number of legitimate and misbehaving nodes). Proposed framework for classification of misbehavior is evaluated using WEKA. Our approach is efficient in classifying multiple misbehaviors present in VANET scenario. Experiment result shows that Random Forest and J-48 classifiers perform better compared to other classifiers.

Machine Learning Approach for Multiple Misbehavior Detection in VANET

In recent years, malware has emerged as a critical security threat. In addition, malware authors continue to embed numerous anti-detection features to evade the existing malware detection approaches. Against this advanced class of malicious programs, dynamic behavior-based malware detection approaches outperform the traditional signature-based approaches by neutralizing the effects of obfuscation and morphing techniques.

/pdf/employing-program-semantics-for-malware-detection-38kavyth29.pdf

Employing Program Semantics for Malware Detection

Malware detection and prevention methods are increasingly becoming necessary for computer systems connected to the Internet. The traditional signature based detection of malware fails for metamorphic malware which changes its code structurally while maintaining functionality at time of propagation. This category of malware is called metamorphic malware. In this paper we dynamically analyze the executables produced from various metamorphic generators through an emulator by tracing API calls. A signature is generated for an entire malware class (each class representing a family of viruses generated from one metamorphic generator) instead of for individual malware sample. We show that most of the metamorphic viruses of same family are detected by the same base signature. Once a base signature for a particular metamorphic generator is generated, all the metamorphic viruses created from that tool are easily detected by the proposed method. A Proximity Index between the various Metamorphic generators has been proposed to determine how similar two or more generators are.

Vijay Laxmi

Papers

Android Security: A Survey of Issues, Malware Penetration, and Defenses

AndroSimilar: robust statistical feature signature for Android malware detection

Machine Learning Approach for Multiple Misbehavior Detection in VANET

Employing Program Semantics for Malware Detection

MEDUSA: MEtamorphic malware dynamic analysis usingsignature from API