The Negative Selection Algorithm is an immunology-inspired algorithm for anomaly detection application. This algorithm has been implemented with different pattern representations and various matching rules and success...
详细信息
ISBN:
(纸本)1595933158
The Negative Selection Algorithm is an immunology-inspired algorithm for anomaly detection application. This algorithm has been implemented with different pattern representations and various matching rules and successfully applied to a broad range of problems. Recent research shows serious problems with this algorithm in terms of both efficiency and effectiveness. In this paper we evaluated the performance of the algorithm constraint-based representation. We argue that the algorithm and problem representations should be considered separately, and that best performance of the algorithm may be obtained by choosing a proper representation.
Next generation mobile networks (NGMN) are expected to integrate several heterogenous wireless technologies in order to provide high system capacity and cost effective global service coverage. In this paper we propose...
详细信息
ISBN:
(纸本)1595933069
Next generation mobile networks (NGMN) are expected to integrate several heterogenous wireless technologies in order to provide high system capacity and cost effective global service coverage. In this paper we propose an efficient predictive admission control policy for heterogenous wireless bandwidth allocation. We predict well chosen traffic parameters using neural networks and we estimate blocking probabilities using generally distributed traffic models. Furthermore, we use a Tabu search algorithm to find the optimal guard band for a multi-layer heterogenous NGMN. The objective of our multi-layer predictive admission control policy (MLPAC) is to minimize global blocking probability while guaranteeing a hard constraint on handoff dropping probability. It extends the overflow scheme used in two-layer hierarchical cellular systems (HCS) to multiple heterogenous access technologies in NGMN. Presented results show that our MLPAC approach is more efficient in allocating the scarce heterogenous wireless bandwidth to a higher number of accepted connections while maintaining minimal guard bands for horizontal and vertical handoff protection. Copyright 2006 ACM.
The two main challenges associated with mining data streams are concept drifting and data noise. Current algorithms mainly depend on the robust of the base classifier or learning ensembles, and have no active mechanis...
详细信息
ISBN:
(纸本)3540459162
The two main challenges associated with mining data streams are concept drifting and data noise. Current algorithms mainly depend on the robust of the base classifier or learning ensembles, and have no active mechanisms to deal noisy. However, noise still can induce the drastic drops in accuracy. In this paper, we present a clustering-based method to filter out hard instances and noise instances from data streams. We also propose a trigger to detect concept drifting and build RobustBoosting, an ensemble classifier, by boosting the hard instances. We evaluated RobustBoosting algorithm and AdaptiveBoosting algorithm [1] on the synthetic and real-life data sets. The experiment results show that the proposed method has substantial advantage over AdaptiveBoosting algorithm in prediction accuracy, and that it can converge to target concepts efficiently with hi h accuracy on datasets with noise level as high as 40%.
softwareengineering researchers are increasingly relying on the empirical approach to advance the state of the art. The level of empirical rigor and evidence required to guide softwareengineering research, however, ...
详细信息
ISBN:
(纸本)1595933751
softwareengineering researchers are increasingly relying on the empirical approach to advance the state of the art. The level of empirical rigor and evidence required to guide softwareengineering research, however, can vary drastically depending on many factors. In this session we identify some of these factors through a discussion of the state of the art in performing empirical studies in softwareengineering, and we show how we can utilize the notion of empirical maturity to set and adjust the empirical expectations for softwareengineering research efforts. Regarding the state of the art in performing empirical studies, we will offer perspectives on two classes of study: those concerned with humans utilizing a technology, e.g., a person applying a methodology, a technique, or a tool, where human skills and the ability to interact with the technology are some of the primes issues, and those concerned with the application of the technology to an artifact, e.g., a technique or tool applied to a design or a program. In the first case, the emphasis is typically on issues like feasibility, usefulness, and then on effectiveness. The technology tends to be less well specified and based more on the experience and skills of the technology applier. In the second case, the emphasis is typically on the efficiency and effectiveness of the technology. The technology tends to be well defined and the assumption is that the individual skill and experience plays a less important role. We will discuss the set of factors that influence the design, implementation, and validity of these studies. Regarding empirical maturity and its implications on the SE community's expectations, we will provide examples of the large spectrum of studies with different maturity levels that can be performed to successfully support softwareengineering research. We will then identify and analyze the following aspects that are likely to impact a study's maturity level: technology (well-specified vs. unde
Deploying probes easily and dynamically is very important to the success of end to end measurement systems. A P2P-like dynamic probe deployment mechanism for large scale end to end network measurement is proposed in t...
详细信息
This paper proposes behavioral footprinting, a new dimension of worm profiling based on worm infection sessions. A worm's infection session contains a number of steps (e.g., for probing, exploitation, and replicat...
详细信息
ISBN:
(纸本)1595934472
This paper proposes behavioral footprinting, a new dimension of worm profiling based on worm infection sessions. A worm's infection session contains a number of steps (e.g., for probing, exploitation, and replication) that are exhibited in certain order in every successful worm infection. Behavioral footprinting complements content-based signature by enriching a worm's profile, which will be used in worm identification, an important task in post worm attack investigation and recovery. We propose an algorithm to extract a worm's behavioral footprint from the worm's traffic traces. Our evaluation with a number of real worms and their variants confirms the existence of worms' behavioral footprints and demonstrates their effectiveness in worm identification. Copyright 2006 ACM.
Automatic verification techniques, which analyze all processes at once, typically do not scale well for large, complex concurrent software systems because of the theoretic barrier - PSPACE complexity in worst case. In...
详细信息
ISBN:
(纸本)1595931082
Automatic verification techniques, which analyze all processes at once, typically do not scale well for large, complex concurrent software systems because of the theoretic barrier - PSPACE complexity in worst case. In this paper, we present our tool named ARCATS (Architecture Refactoring and Compositional Analysis Tool Suite). ARCATS consists a set of tools to combat state explosion in a divide-and-conquer, hierarchical manner, which is also known as compositional analysis. We build these tools to seek out best combinations to scale the verification to larger software systems. Copyright 2006 ACM.
Recently, P2P (Peer to Peer) systems, direct file sharing systems among the peers, are one of the most attractive file sharing system. P2P architectures have high scalability and high performance due to the fact that ...
详细信息
ISBN:
(纸本)8955191294
Recently, P2P (Peer to Peer) systems, direct file sharing systems among the peers, are one of the most attractive file sharing system. P2P architectures have high scalability and high performance due to the fact that its architectures which have characteristics of distributed file processing. However, P2P Architecture is infamous for distribution channel of illegal contents. So we must apply the DRM (Digital Rights Management) system to the P2P architecture, and we should keep advantages of P2P even if after DRM system applied. In this paper, we propose new type of DRM applied P2P system architecture that keeps existing P2P system's advantages.
Large teams of practitioners (developers, testers, etc.) usually work in parallel on the same code base. A major concern when working in parallel is the introduction of integration bugs in the latest shared code. Thes...
详细信息
This paper describes the design and implementations of the first application service based on embedded Linux;Mobile Teller which uses WiBro network for data communications and CDMA network for voice communications. Wh...
详细信息
ISBN:
(纸本)8955191294
This paper describes the design and implementations of the first application service based on embedded Linux;Mobile Teller which uses WiBro network for data communications and CDMA network for voice communications. When a sender on a dual-mode terminal types texts, the texts are transmitted to a TTS server located in the Internet through WiBro network. Subsequently, the TTS server changes the texts into voices and transmits the voice data to the dual-mode terminal. At last the dual-mode terminal sends the voice to the receiver through the CDMA network. In case of noisy environment, Mobile Teller makes user to voice communication.
暂无评论