Browsing by Author "Mendez, J. R."

Now showing 1 - 5 of 5
  • Thumbnail Image
    ItemOpen Access
    Evolutionary Multi-objective Scheduling for Anti-Spam Filtering Throughput Optimization
    (Springer, 2017-06-02) Mendez, J. R.; Yevseyeva, Iryna; Basto-Fernandes, V.; Ruano-Ordás, D.
    This paper presents an evolutionary multi-objective optimization problem formulation for the anti-spam filtering problem, addressing both the classification quality criteria (False Positive and False Negative error rates) and email messages classification time (minimization). This approach is compared to single objective problem formulations found in the literature, and its advantages for decision support and flexible/adaptive anti-spam filtering configuration is demonstrated. A study is performed using the Wirebrush4SPAM framework anti-spam filtering and the SpamAssassin email dataset. The NSGA-II evolutionary multi-objective optimization algorithm was applied for the purpose of validating and demonstrating the adoption of this novel approach to the anti-spam filtering optimization problem, formulated from the multi-objective optimization perspective. The results obtained from the experiments demonstrated that this optimization strategy allows the decision maker (anti-spam filtering system administrator) to select among a set of optimal and flexible filter configuration alternatives with respect to classification quality and classification efficiency.
  • Thumbnail Image
    ItemOpen Access
    Improving the drug discovery process by using multiple classifier systems
    (Elsevier, 2019-03-01) Ruano-Ordas, D.; Yevseyeva, Iryna; Fernandes, Vitor Basto; Mendez, J. R.; Emmerich, Michael T. M.
    Machine learning methods have become an indispensable tool for utilizing large knowledge and data repositories in science and technology. In the context of the pharmaceutical domain, the amount of acquired knowledge about the design and synthesis of pharmaceutical agents and bioactive molecules (drugs) is enormous. The primary challenge for automatically discovering new drugs from molecular screening information is related to the high dimensionality of datasets, where a wide range of features is included for each candidate drug. Thus, the implementation of improved techniques to ensure an adequate manipulation and interpretation of data becomes mandatory. To mitigate this problem, our tool (called D2-MCS) can split homogeneously the dataset into several groups (the subset of features) and subsequently, determine the most suitable classifier for each group. Finally, the tool allows determining the biological activity of each molecule by a voting scheme. The application of the D2-MCS tool was tested on a standardized, high quality dataset gathered from ChEMBL and have shown outperformance of our tool when compare to well-known single classification models.
  • Thumbnail Image
    ItemOpen Access
    Optimizing anti-spam filters with evolutionary algorithms
    (Elsevier, 2013-01-18) Yevseyeva, Iryna; Basto-Fernandes, V.; Ruano-Ordás, D.; Mendez, J. R.
    This work is devoted to the problem of optimising scores for anti-spam filters, which is essential for the accuracy of any filter based anti-spam system, and is also one of the biggest challenges in this research area. In particular, this optimisation problem is considered from two different points of view: single and multiobjective problem formulations. Some of existing approaches within both formulations are surveyed, and their advantages and disadvantages are discussed. Two most popular evolutionary multiobjective algorithms and one single objective algorithm are adapted to optimisation of the anti-spam filters’ scores and compared on publicly available datasets widely used for benchmarking purposes. This comparison is discussed, and the recommendations for the developers and users of optimising anti-spam filters are provided.
  • Thumbnail Image
    ItemOpen Access
    Quadcriteria Optimization of Binary Classifiers: Error Rates, Coverage, and Complexity
    (Springer, 2017-11-11) Emmerich, M. T. M.; Basto-Fernandes, V.; Yevseyeva, Iryna; Ruano-Ordás, D.; Zhao, J.; Fdez-Riverola, F.; Mendez, J. R.
    This paper presents a 4-objective evolutionary multiobjective optimization study for optimizing the error rates (false positives, false negatives), reliability, and complexity of binary classifiers. The example taken is the email anti-spam filtering problem. The two major goals of the optimization is to minimize the error rates that is the false negative rate and the false positive rate. Our approach discusses three-way classification, that is the binary classifier can also not classify an instance in cases where there is not enough evidence to assign the instance to one of the two classes. In this case the instance is marked as suspicious but still presented to the user. The number of unclassified (suspicious) instances should be minimized, as long as this does not lead to errors. This will be termed the coverage objective. The set (ensemble) of rules needed for the anti-spam filter to operate in optimal conditions is addressed as a fourth objective. All objectives stated above are in general conflicting with each other and that is why we address the problem as a 4-objective (quadcriteria) optimization problem. We assess the performance of a set of state-of-the-art evolutionary multiobjective optimization algorithms. These are NSGA-II, SPEA2, and the hypervolume indicator-based SMS-EMOA. Focusing on the anti-spam filter optimization, statistical comparisons on algorithm performance are provided on several benchmarks and a range of performance indicators. Moreover, the resulting 4-D Pareto hyper-surface is discussed in the context of binary classifier optimization.
  • Thumbnail Image
    ItemOpen Access
    A spam filtering mult-iobjective optimization study covering parsimony maximization and three-way classification
    (Elsevier, 2016-07-04) Basto-Fernandes, V.; Yevseyeva, Iryna; Mendez, J. R.; Zhao, J.; Fdez-Riverola, F.; Emmerich, M. T. M.
    Classifier performance optimization in machine learning can be stated as a multi-objective optimization problem. In this context, recent works have shown the utility of simple evolutionary multi-objective algorithms (NSGA-II, SPEA2) to conveniently optimize the global performance of different anti-spam filters. The present work extends existing contributions in the spam filtering domain by using three novel indicator-based (SMS-EMOA, CH-EMOA) and decomposition-based (MOEA/D) evolutionary multi-objective algorithms. The proposed approaches are used to optimize the performance of a heterogeneous ensemble of classifiers into two different but complementary scenarios: parsimony maximization and e-mail classification under low confidence level. Experimental results using a publicly available standard corpus allowed us to identify interesting conclusions regarding both the utility of rule-based classification filters and the appropriateness of a three-way classification system in the spam filtering domain.