Asim Karim

Global optimum design of cold-formed steel hat-shape beams

Abstract Using the computational neural network model developed recently by the authors, a comprehensive parametric study is performed for global optimization of cold-formed steel hat-shape beams based on the AISI Specifications. Design curves are presented for global optimum values of the thickness, the web-depth-to-thickness ratio, and the flange width-to-thickness ratio for unbraced beams having steel yield strengths of 250 and 345 N/mm 2 . The computational neural network model guarantees a local optimum solution. The global optimum is found by an exhaustive search that is guided by a heuristic approach to reduce the search effort. An extensive parametric study yielded insights into the behavior of cold-formed steel beams that are then used as rules to reduce the search space and guide the exhaustive search. The procedure for finding the global optimum design of cold-formed steel beams is presented in a few recursive steps. The optimum design curves presented in this article can be of great value to structural design engineers.

PSSF: A Novel Statistical Approach for Personalized Service-side Spam Filtering

The volume of spam e-mails has grown rapidly in the last two years resulting in increasing costs to users, network operators, and e-mail service providers (ESPs). E-mail users demand accurate spam filtering with minimum effort from their side. Since the distribution of spam and non-spam e-mails is often different for different users a single filter trained on a general corpus is not optimal for all users. The question asked by ESPs is: How do you build robust and scalable automatic personalized spam filters? We address this question by presenting PSSF, a novel statistical approach for personalized service-side spam filtering. PSSF builds a discriminative classifier from a statistical model of spam and non-spam e-mails. A classifier is first built on a general training corpus that is then adapted in one or more passes of soft labeling and classifier rebuilding over each users unlabeled e-mails. The statistical model captures the distribution of tokens in spam and non-spam e-mails. This model is robust in the sense that its size can be reduced significantly without degrading filtering performance. We evaluate PSSF on two datasets. The results demonstrate the superior performance and scalability of PSSF in comparison with other published results on the same datasets.

Decision Theory for Discrimination-Aware Classification

Social discrimination (e.g., against females) arising from data mining techniques is a growing concern worldwide. In recent years, several methods have been proposed for making classifiers learned over discriminatory data discrimination-aware. However, these methods suffer from two major shortcomings: (1) They require either modifying the discriminatory data or tweaking a specific classification algorithm and (2) They are not flexible w.r.t. discrimination control and multiple sensitive attribute handling. In this paper, we present two solutions for discrimination-aware classification that neither require data modification nor classifier tweaking. Our first and second solutions exploit, respectively, the reject option of probabilistic classifier(s) and the disagreement region of general classifier ensembles to reduce discrimination. We relate both solutions with decision theory for better understanding of the process. Our experiments using real-world datasets demonstrate that our solutions outperform existing state-of-the-art methods, especially at low discrimination which is a significant advantage. The superior performance coupled with flexible control over discrimination and easy applicability to multiple sensitive attributes makes our solutions an important step forward in practical discrimination-aware classification.

A Robust Discriminative Term Weighting Based Linear Discriminant Method for Text Classification

Text classification is widely used in applications ranging from e-mail filtering to review classification. Many of these applications demand that the classification method be efficient and robust, yet produce accurate categorizations by using the terms in the documents only. We present a supervised text classification method based on discriminative term weighting, discrimination information pooling, and linear discrimination. Terms in the documents are assigned weights according to the discrimination information they provide for one category over the others. These weights also serve to partition the terms into two sets. A linear opinion pool is adopted for combining the discrimination information provided by each set of terms yielding a two-dimensional feature space. Subsequently, a linear discriminant function is learned to categorize the documents in the feature space. We provide intuitive and empirical evidence of the robustness of our method with three term weighting strategies. Experimental results are presented for data sets from three different application areas. The results show that our methods accuracy is higher than other popular methods, especially when there is a distribution shift from training to testing sets. Moreover, our method is simple yet robust to different application domains and small training set sizes.

Efficient Rule Generation for Cost-Sensitive Misuse Detection Using Genetic Algorithms

This paper presents a genetic algorithm (GA) for generating efficient rules for cost-sensitive misuse detection in intrusion detection systems. The GA employs only the five most relevant features for each attack category for rule generation. Furthermore, it incorporates the different costs of misclassifying attacks in its fitness function to yield rules that are cost sensitive. The generated rules signal an attack as well as its category. The GA is implemented and evaluated on the KDDCup 99 dataset. Its detection performance is comparable to the winners of the KDDCup 99 competition. However, the rules generated by our GA are short and amenable for real time misuse detection

Automatic Personalized Spam Filtering through Significant Word Modeling

Many real life situations can be modeled as Prisoners dilemma. There are various strategies in the literature. However, few of which match the design objectives of an intelligent agent - being reactive and pro-active. In this paper, we incorporate risk attitude and reputation into infinitely repeated games. In this way, we find that the original game matrix can be transformed to a new matrix, which has a kind of cooperative equilibrium. We use the proposed concepts to analyze the Iterated Prisoners dilemma. Simulation also shows that agents, which consider risk attitude and reputation in the decision-making process, have improved performance and are reactive as well as pro-active.

CDIM: Document Clustering by Discrimination Information Maximization

Ideally, document clustering methods should produce clusters that are semantically relevant and readily understandable as collections of documents belonging to particular contexts or topics. However, existing popular document clustering methods often ignore term-document corpus-based semantics while relying upon generic measures of similarity. In this paper, we present CDIM, an algorithmic framework for partitional clustering of documents that maximizes the sum of the discrimination information provided by documents. CDIM exploits the semantic that term discrimination information provides better understanding of contextual topics than term-to-term relatedness to yield clusters that are describable by their highly discriminating terms. We evaluate the proposed clustering algorithm using well-known discrimination/semantic measures including Relative Risk (RR), Measurement of Discrimination Information (MDI), Domain Relevance (DR), and Domain Consensus (DC) on twelve data sets to prove that CDIM produces high-quality clusters comparable to the best methods. We also illustrate the understandability and efficiency of CDIM, suggesting its suitability for practical document clustering.

Controlling Attribute Effect in Linear Regression

In data mining we often have to learn from biased data, because, for instance, data comes from different batches or there was a gender or racial bias in the collection of social data. In some applications it may be necessary to explicitly control this bias in the models we learn from the data. This paper is the first to study learning linear regression models under constraints that control the biasing effect of a given attribute such as gender or batch number. We show how propensity modeling can be used for factoring out the part of the bias that can be justified by externally provided explanatory attributes. Then we analytically derive linear models that minimize squared error while controlling the bias by imposing constraints on the mean outcome or residuals of the models. Experiments with discrimination-aware crime prediction and batch effect normalization tasks show that the proposed techniques are successful in controlling attribute effects in linear regression models.

Metarule-guided association rule mining for program understanding

Software systems are expected to change over their lifetime in order to remain useful. Understanding a software system that has undergone changes is often difficult owing to the unavailability of up-to-date documentation. Under these circumstances, source code is the only reliable means of information regarding the system. In the paper, association rule mining is applied to the problem of software understanding i.e. given the source files of a software system, association rule mining is used to gain an insight into the software. To make association rule mining more effective, constraints are placed on the mining process in the form of metarules. Metarule-guided mining is carried out to find associations which can be used to identify recurring problems within software systems. Metarules are related to re-engineering patterns which present solutions to these problems. Association rule mining is applied to five legacy systems and results presented show how extracted association rules can be helpful in analysing the structure of a software system and modifications to improve the structure are suggested. A comparison of the results obtained for the five systems also reveals legacy system characteristics, which can lead to understanding the nature of open source legacy software and its evolution.

Helicobacter urease: Niche construction at the single molecule level

The urease of the human pathogen, Helicobacter pylori, is essential for pathogenesis. The ammonia produced by the enzyme neutralizes stomach acid; thereby modifying its environment. The dodecameric enzyme complex has high affinity for its substrate, urea. We compared urease sequences and derivative 3D homology model structures from all published Helicobacter genomes and an equal number of genomes belonging to strains of another enteric bacterium, Escherichia coli. We found that the enzyme’s architecture adapts to fit its niche. This finding, coupled to a survey of other physiological features responsible for the bacterium’s acid resistance, suggests how it copes with pH changes caused by disease onset and progression.