MVGL analyser for multi-classifier based spam filtering system

In the last decade, the rapid growth of the Internet and email, there has been a dramatic growth in spam. Spam is commonly defined as unsolicited email messages and protecting email from the infiltration of spam is an important research issue. Classifications algorithms have been successfully used to filter spam, but with a certain amount of false positive trade-offs, which is unacceptable to users sometimes. This paper presents an approach to overcome the burden of GL (grey list) analyzer as further refinements to our multi-classifier based classification model (Islam, M. and W. Zhou 2007). In this approach, we introduce a ldquomajority voting grey list (MVGL)rdquo analyzing technique which will analyze the generated GL emails by using the majority voting (MV) algorithm. We have presented two different variations of the MV system, one is simple MV (SMV) and other is the ranked MV (RMV). Our empirical evidence proofs the improvements of this approach compared to the existing GL analyzer of multi-classifier based spam filtering process.

Optimal actuator fault tolerance for static nonlinear systems based on minimum output velocity jump

Fault tolerance for a class of non linear systems is addressed based on the velocity of their output variables. This paper presents a mapping to minimize the possible jump of the velocity of the output, due to the actuator failure. The failure of the actuator is assumed as actuator lock. The mapping is derived and it provides the proper input commands for the healthy actuators of the system to tolerate the effect of the faulty actuator on the output of the system. The introduced mapping works as an optimal input reconfiguration for fault recovery, which provides a minimum velocity jump suitable for static nonlinear systems. The proposed mapping is validated through different case studies and a complementary simulation. In the case studies and the simulation, the mapping provides the commands to compensate the effect of different faults within the joints of a robotic manipulator. The new commands and the compare between the velocity of the output variables for the health and faulty system are presented.

An approach to filtering duplicate RFID data streams

In a system where distributed network of Radio Frequency Identification (RFID) readers are used to collaboratively collect data from tagged objects, a scheme that detects and eliminates redundant data streams is required. To address this problem, we propose an approach that is based on Bloom filter to detect duplicate readings and filter redundant RFID data streams. We have evaluated the performance of the proposed approach and compared it with existing approaches. The experimental results demonstrate that the proposed approach provides superior performance as compared to the baseline approaches.

Spamcooling : a parallel heterogeneous ensemble spam filtering system based on active learning techniques

Anti-spam technology is developing rapidly in recent years. With the emerging applications of machine learning in diverse fields, researchers as well as manufacturers around the world have attempted a large number of related algorithms to prevent spam. In this paper, we designed an effective anti-spam protection system, SpamCooling, based on the mechanism of active learning and parallel heterogeneous ensemble learning techniques. The system adopts a batch method to filter spam and can be easily incorporated with existing mail clients (MUA). It can actively obtain user feedbacks for providing users with personalized spam filtering experiences. The parallel heterogeneous ensemble method can help system achieve high spam detection rate as well as low ham misclassification rate.

Electronic commerce publications and the implications for research quality output in Australia

Australian universities and academics will soon see a major change in the way research is reported and funded. It is expected that by 2008, according to the most recent timetable (Bishop 2006), the Research Quality Framework (RQF) will be implemented. The result of the announcement has been an increased activity within universities focusing on the proposed criteria. The proposed RQF will seek to have research assessed according to quality and impact. Part of both quality and impact relates to where research is published. For academics it will be increasingly important to target high quality journals if the research is to be rated as high quality. The question this raises for Information Systems academics is where do we publish for maximum impact? The Information Systems (IS) field is diverse with researchers working in many areas and a publication outlet for one area may not be relevant for another. One area where many Australian IS researchers have focused their research interest is the field of electronic commerce (e-commerce). The research reported in this paper identified the publication outlets that would be regarded as amongst the highest quality for researchers wishing to publish e-commerce research. The authors analysed e-commerce research papers by Australian researchers published in the period 2000 to 2005. The results describe where Australian researchers are publishing in this field. The paper also provides guidance to those working in the e-commerce field on which journals and conferences to target to ensure their work rates highly in terms of the RQF.

Are shocks to real output permanent or transitory? Evidence from a panel of Pacific Island countries

Panel unit root and stationarity tests without structural breaks suggest that for eight Pacific island economies real gross domestic product (GDP) per capita contains a unit root. The panel stationarity test that accommodates structural change in the trend function, however, finds evidence in favour of regimewise trend stationarity. This result points to the importance of taking structural breaks into account. The finding implies that, for the period considered, the permanent secular component of output is dominated by transitory fluctuations accompanied by infrequent changes in the trend function. The only exceptions are Fiji and Kiribati, for which individual stationarity tests with multiple structural breaks suggest that real GDP per capita contains a unit root.

Positive and negative effects of phoretic mites on the reproductive output of an invasive bark beetle

When multiple species coexist upon a single host, their combined effect on the host can be unpredictable. We explored the effect of phoretic mites on the reproductive output of the five-spined bark beetle, Ips grandicollis. Using correlative approaches and experimental manipulation of mite numbers we examined how mite load affected the number, size and condition of bark beetle offspring produced. We found that mites have both negative and positive consequences on different aspects of bark beetle reproduction. Females from which mites were removed were more fecund and produced larger offspring than females with mites, implying a cost of mite loads. However, when mites were present on females, those bearing the highest mite loads produced offspring that were larger and in better condition, indicating a beneficial effect of mites. These data suggest that phoretic interactions between mites and bark beetles differ over the course of the host’s lifespan, with either the mites interacting in different ways with different life stages of the host (parasitic on adult, mutualistic with larvae), and/or the beetles being host to different mite assemblages over their lifetime.

Gender differences in publication output: towards an unbiased metric of research performance

We examined the publication records of a cohort of 168 life scientists in the field of ecology and evolutionary biology to assess gender differences in research performance. Clear discrepancies in publication rate between men and women appear very early in their careers and this has consequences for the subsequent citation of their work. We show that a recently proposed index designed to rank scientists fairly is in fact strongly biased against female researchers, and advocate a modified index to assess men and women on a more equitable basis.

Mobile agent tracking with single frequency continuous wave radar : a linear robust filtering based approach

3-alkylpyrrole to the fabric surface. Direct applications of a conductive paint to the textile surface eliminate the exposure of the substrate to damaging oxidizing agents which allow the coating of more sensitive and delicate substrates. All textiles produced are tested for abrasion resistance and conductivity. For alkyl polypyrrole coated fabrics, the optimum carbon chain lengths are between n=10 and n=14, which result in optimum values of conductivity and solubility. The darkness of the tone is inversely related to the surface resistivity of the resulting conductive fabric. Therefore, deep black coatings have low resistivity whereas light gray coatings on a white fabric surface have higher surface resistivity. Longer alkyl chains result in higher surface resistivity in fabrics. The conductive coating of poly(3-decanylpyrrole) on the textile surface has a better abrasion resistance compared to that of an unsubstituted polypyrrole coating.

Ordinal Boltzmann machines for collaborative filtering

Collaborative filtering is an effective recommendation technique wherein the preference of an individual can potentially be predicted based on preferences of other members. Early algorithms often relied on the strong locality in the preference data, that is, it is enough to predict preference of a user on a particular item based on a small subset of other users with similar tastes or of other items with similar properties. More recently, dimensionality reduction techniques have proved to be equally competitive, and these are based on the co-occurrence patterns rather than locality. This paper explores and extends a probabilistic model known as Boltzmann Machine for collaborative filtering tasks. It seamlessly integrates both the similarity and cooccurrence in a principled manner. In particular, we study parameterisation options to deal with the ordinal nature of the preferences, and propose a joint modelling of both the user-based and item-based processes. Experiments on moderate and large-scale movie recommendation show that our framework rivals existing well-known methods.

Probabilistic models over ordered partitions with applications in document ranking and collaborative filtering

Ranking is an important task for handling a large amount of content. Ideally, training data for supervised ranking would include a complete rank of documents (or other objects such as images or videos) for a particular query. However, this is only possible for small sets of documents. In practice, one often resorts to document rating, in that a subset of documents is assigned with a small number indicating the degree of relevance. This poses a general problem of modelling and learning rank data with ties. In this paper, we propose a probabilistic generative model, that models the process as permutations over partitions. This results in super-exponential combinatorial state space with unknown numbers of partitions and unknown ordering among them. We approach the problem from the discrete choice theory, where subsets are chosen in a stagewise manner, reducing the state space per each stage significantly. Further, we show that with suitable parameterisation, we can still learn the models in linear time. We evaluate the proposed models on two application areas: (i) document ranking with the data from the recently held Yahoo! challenge, and (ii) collaborative filtering with movie data. The results demonstrate that the models are competitive against well-known rivals.