Wyszukiwarka
Wyniki wyszukiwania dla: CLASS IMBALANCE PROBLEM (CIP)
-
Performance Analysis of Machine Learning Methods with Class Imbalance Problem in Android Malware Detection
PublikacjaDue to the exponential rise of mobile technology, a slew of new mobile security concerns has surfaced recently. To address the hazards connected with malware, many approaches have been developed. Signature-based detection is the most widely used approach for detecting Android malware. This approach has the disadvantage of being unable to identify unknown malware. As a result of this issue, machine learning (ML) for detecting malware...
-
Empirical analysis of tree-based classification models for customer churn prediction
PublikacjaCustomer churn is a vital and reoccurring problem facing most business industries, particularly the telecommunications industry. Considering the fierce competition among telecommunications firms and the high expenses of attracting and gaining new subscribers, keeping existing loyal subscribers becomes crucial. Early prediction of disgruntled subscribers can assist telecommunications firms in identifying the reasons for churn and...
-
Study of Multi-Class Classification Algorithms’ Performance on Highly Imbalanced Network Intrusion Datasets
PublikacjaThis paper is devoted to the problem of class imbalance in machine learning, focusing on the intrusion detection of rare classes in computer networks. The problem of class imbalance occurs when one class heavily outnumbers examples from the other classes. In this paper, we are particularly interested in classifiers, as pattern recognition and anomaly detection could be solved as a classification problem. As still a major part of...
-
Sampling-based novel heterogeneous multi-layer stacking ensemble method for telecom customer churn prediction
PublikacjaIn recent times, customer churn has become one of the most significant issues in business-oriented sectors with telecommunication being no exception. Maintaining current customers is particularly valuable due to the high degree of rivalry among telecommunication companies and the costs of acquiring new ones. The early prediction of churned customers may help telecommunication companies to identify the causes of churn and design...
-
Robustness in Compressed Neural Networks for Object Detection
PublikacjaModel compression techniques allow to significantly reduce the computational cost associated with data processing by deep neural networks with only a minor decrease in average accuracy. Simultaneously, reducing the model size may have a large effect on noisy cases or objects belonging to less frequent classes. It is a crucial problem from the perspective of the models' safety, especially for object detection in the autonomous driving...
-
Smart Approach for Glioma Segmentation in Magnetic Resonance Imaging using Modified Convolutional Network Architecture (U-NET)
PublikacjaSegmentation of a brain tumor from magnetic resonance multimodal images is a challenging task in the field of medical imaging. The vast diversity in potential target regions, appearance and multifarious intensity threshold levels of various tumor types are few of the major factors that affect segmentation results. An accurate diagnosis and its treatment demand strict delineation of the tumor affected tissues. Herein, we focus on...
-
Compact 4 × 4 butler matrix with non‐standard phase differences for IoT applications
PublikacjaButler matrices represent a popular class of feeding networks for antenna arrays. Large dimensions and the lack of flexibility in terms of achievable output phase difference make conventional Butler structures of limited use for modern communication devices. In this work, a compact planar 4 × 4 matrix with non-standard relative phase shifts of –30º, 150º, –120º, and 60º has been proposed. The structure is designed to operate at...
-
Intelligent Decision Forest Models for Customer Churn Prediction
PublikacjaCustomer churn is a critical issue impacting enterprises and organizations, particularly in the emerging and highly competitive telecommunications industry. It is important to researchers and industry analysts interested in projecting customer behavior to separate churn from non‐churn consumers. The fundamental incentive is a firm’s intent desire to keep current consumers, along with the exorbitant expense of gaining new ones....
-
Interpretable deep learning approach for classification of breast cancer - a comparative analysis of multiple instance learning models
PublikacjaBreast cancer is the most frequent female cancer. Its early diagnosis increases the chances of a complete cure for the patient. Suitably designed deep learning algorithms can be an excellent tool for quick screening analysis and support radiologists and oncologists in diagnosing breast cancer.The design of a deep learning-based system for automated breast cancer diagnosis is not easy due to the lack of annotated data, especially...
-
Marek Czachor prof. dr hab.
Osoby -
Spiral Search Grasshopper Features Selection with VGG19-ResNet50 for Remote Sensing Object Detection
PublikacjaRemote sensing object detection plays a major role in satellite imaging and is required in various scenarios such as transportation, forestry, and the ocean. Deep learning techniques provide efficient performance in remote sensing object detection. The existing techniques have the limitations of data imbalance, overfitting, and lower efficiency in detecting small objects. This research proposes the spiral search grasshopper (SSG)...
-
Minimal number of periodic points for smooth self-maps of simply-connected manifolds
Dane BadawczeThe problem of finding the minimal number of periodic points in a given class of self-maps of a space is one of the central questions in periodic point theory. We consider a closed smooth connected and simply-connected manifold of dimension at least 4 and its self-map f. The topological invariant D_r[f] is equal to the minimal number of r-periodic points...
-
Minimal number of periodic points with the periods less or equal to r in the smooth homotopy class of simply-connected manifolds of dimension 4 and homology groups with the sum of ranks less or equal to10
Dane BadawczeAn important problem in periodic point theory is minimization of the number of periodic points with periods <= r in a given class of self-maps of a space. A closed smooth and simply-connected manifolds of dimension 4 and its self-maps f with periodic sequence of Lefschetz numbers are considered. The topological invariant Jr[f] is equal to the minimal...
-
Minimal number of periodic points with the periods less or equal to r in the smooth homotopy class of simply-connected manifolds of dimension 6 and homology groups with the sum of ranks less or equal to10
Dane BadawczeAn important problem in periodic point theory is minimization of the number of periodic points with periods <= r in a given class of self-maps of a space. A closed smooth and simply-connected manifolds of dimension 6 and its self-maps f with periodic sequence of Lefschetz numbers are considered. The topological invariant Jr[f] is equal to the minimal...
-
Minimal number of periodic points with the periods less or equal to r in the smooth homotopy class of simply-connected manifolds of dimension 5 and homology groups with the sum of ranks less or equal to10
Dane BadawczeAn important problem in periodic point theory is minimization of the number of periodic points with periods <= r in a given class of self-maps of a space. A closed smooth and simply-connected manifolds of dimension 5 and its self-maps f with periodic sequence of Lefschetz numbers are considered. The topological invariant Jr[f] is equal to the minimal...
-
Minimal number of periodic points with the periods less or equal to r in the smooth homotopy class of simply-connected manifolds of dimension 8 and homology groups with the sum of ranks less or equal to 10
Dane BadawczeAn important problem in periodic point theory is minimization of the number of periodic points with periods <= r in a given class of self-maps of a space. A closed smooth and simply-connected manifolds of dimension 8 and its self-maps f with periodic sequence of Lefschetz numbers are considered. The topological invariant Jr[f] is equal to the minimal...
-
Minimal number of periodic points with the periods less or equal to r in the smooth homotopy class of simply-connected manifolds of dimension 7 and homology groups with the sum of ranks less or equal to10
Dane BadawczeAn important problem in periodic point theory is minimization of the number of periodic points with periods <= r in a given class of self-maps of a space. A closed smooth and simply-connected manifolds of dimension 7 and its self-maps f with periodic sequence of Lefschetz numbers are considered. The topological invariant Jr[f] is equal to the minimal...
-
Modeling and Simulation for Exploring Power/Time Trade-off of Parallel Deep Neural Network Training
PublikacjaIn the paper we tackle bi-objective execution time and power consumption optimization problem concerning execution of parallel applications. We propose using a discrete-event simulation environment for exploring this power/time trade-off in the form of a Pareto front. The solution is verified by a case study based on a real deep neural network training application for automatic speech recognition. A simulation lasting over 2 hours...
-
Parasites of the digestive tract of sheep and goats from organic farms in Western Pomerania, Poland
PublikacjaThe aim of the study was to determine the prevalence and intensity of parasite infection of the digestive tract in sheep and goats from the West Pomerania region following anti-parasite treatment. Feces were freely collected from sheep and goats kept on organic farms and subjected to analysis by the Willis-Schlaf and McMaster’s flotation methods. The mean extensity of infection by gastrointestinal parasites...
-
Existence of solutions with an exponential growth for nonlinear differential-functional parabolic equations
PublikacjaWe consider the Cauchy problem for nonlinear parabolic equations with functional dependence.We prove Schauder-type existence results for unbounded solutions. We also prove existence of maximal solutions for a wide class of differential functional equations.
-
Graph classes generated by Mycielskians
PublikacjaIn this paper we use the classical notion of weak Mycielskian M'(G) of a graph G and the following sequence: M'_{0}(G) =G, M'_{1}(G)=M'(G), and M'_{n}(G)=M'(M'_{n−1}(G)), to show that if G is a complete graph oforder p, then the above sequence is a generator of the class of p-colorable graphs. Similarly, using Mycielskian M(G) we show that analogously defined sequence is a generator of the class consisting of graphs for which the...
-
Application of Barycentric Coordinates in Space Vector PWM Computations
PublikacjaThis paper proposes the use of barycentric coordinates in the development and implementationof space-vector pulse-width modulation (SVPWM) methods, especially for inverters with deformed space-vector diagrams. The proposed approach is capable of explicit calculation of vector duty cycles, independentof whether they assume ideal positions or are displaced due to the DC-link voltage imbalance. The use ofbarycentric coordinates also...
-
ANYTIME POLYNOMIAL HEURISTIC ALGORITHM FOR PARTITIONING GROUPS OF DATA WITH PRESERVING CLASS PROPORTIONS FOR CROSS-VALIDATION
PublikacjaThe article describes a problem of splitting data for k-fold cross-validation, where class proportions must be preserved, with additional constraint that data is divided into groups that cannot be split into different cross-validation sets. This problem often occurs in e.g. medical data processing, where data samples from one patient must be included in the same cross-validation set. As this problem is NP-complete, a heuristic...
-
Chromatic cost coloring of weighted bipartite graphs
PublikacjaGiven a graph G and a sequence of color costs C, the Cost Coloring optimization problem consists in finding a coloring of G with the smallest total cost with respect to C. We present an analysis of this problem with respect to weighted bipartite graphs. We specify for which finite sequences of color costs the problem is NP-hard and we present an exact polynomial algorithm for the other finite sequences. These results are then extended...
-
The Complexity of Zero-Visibility Cops and Robber
PublikacjaIn this work we deal with the computational complexity aspects of the zero-visibility Cops and Robber game. We provide an algorithm that computes the zero-visibility copnumber of a tree in linear time and show that the corresponding decision problem is NP-complete even for the class of starlike graphs.
-
Shared processor scheduling
PublikacjaWe study the shared processor scheduling problem with a single shared processor to maximize total weighted overlap, where an overlap for a job is the amount of time it is processed on its private and shared processor in parallel. A polynomial-time optimization algorithm has been given for the problem with equal weights in the literature. This paper extends that result by showing an (log)-time optimization algorithm for a class...
-
Minimum drag shape bodies moving in inviscid fluid - revisited
PublikacjaThis paper presents the classic approach to minimum drag shape body problem, moving at hypersonic speeds, leading to famous power law shapes with value of the exponent of 3/4. Two- and three-dimensional cases are considered. Furthermore, an exact pseudo solution is given and its uselessness is discussed. Two new solutions are introduced, namely an approximate solution due to form of the functional and solution by means of optimisation...
-
Diagnostyka Systemów 2022
Kursy OnlineThe aim of the course is to familiarize students with modern methods of systems diagnostics. As part of the project, each student solves one problem, prepares a report and finally a project presentation. Celem kursu jest zapoznanie studentów ze współczesnymi metodami diagnostyki systemów. W ramach projektu każdy student rozwiązuje samodzielnie jeden problem, przygotowuje sprawozdanie oraz prezentację projektu.
-
Diagnostyka Systemów 2021
Kursy OnlineThe aim of the course is to familiarize students with modern methods of systems diagnostics. As part of the project, each student solves one problem, prepares a report and finally a project presentation. Celem kursu jest zapoznanie studentów ze współczesnymi metodami diagnostyki systemów. W ramach projektu każdy student rozwiązuje samodzielnie jeden problem, przygotowuje sprawozdanie oraz prezentację projektu.
-
Prevalence Problem in the Set of Quadratic Stochastic Operators Acting on L1
PublikacjaThis paper is devoted to the study of the problem of prevalence in the class of quadratic stochastic operators acting on the L1 space for the uniform topology. We obtain that the set of norm quasi-mixing quadratic stochastic operators is a dense and open set in the topology induced by a very natural metric. This shows the typical long-term behaviour of iterates of quadratic stochastic operators.
-
Regularized Local Basis Function Approach to Identification of Nonstationary Processes
PublikacjaThe problem of identification of nonstationary stochastic processes (systems or signals) is considered and a new class of identification algorithms, combining the basis functions approach with local estimation technique, is described. Unlike the classical basis function estimation schemes, the proposed regularized local basis function estimators are not used to obtain interval approximations of the parameter trajectory, but provide...
-
On Computational Aspects of Greedy Partitioning of Graphs
PublikacjaIn this paper we consider a problem of graph P-coloring consisting in partitioning the vertex set of a graph such that each of the resulting sets induces a graph in a given additive, hereditary class of graphs P. We focus on partitions generated by the greedy algorithm. In particular, we show that given a graph G and an integer k deciding if the greedy algorithm outputs a P-coloring with a least k colors is NP-complete for an infinite...
-
New approach to noncausal identification of nonstationary stochastic systems subject to both smooth and abrupt parameter changes
PublikacjaIn this paper we consider the problem of finiteintervalparameter smoothing for a class of nonstationary linearstochastic systems subject to both smooth and abrupt parameterchanges. The proposed parallel estimation scheme combines theestimates yielded by several exponentially weighted basis functionalgorithms. The resulting smoother automatically adjustsits smoothing bandwidth to the type and rate of nonstationarityof the identified...
-
The complexity of zero-visibility cops and robber
PublikacjaWe consider the zero-visibility cops & robber game restricted to trees. We produce a characterisation of trees of copnumber k and We consider the computational complexity of the zero-visibility Cops and Robber game. We present a heavily modified version of an already-existing algorithm that computes the zero-visibility copnumber of a tree in linear time and we show that the corresponding decision problem is NP-complete on a nontrivial...
-
Computational aspects of greedy partitioning of graphs
PublikacjaIn this paper we consider a variant of graph partitioning consisting in partitioning the vertex set of a graph into the minimum number of sets such that each of them induces a graph in hereditary class of graphs P (the problem is also known as P-coloring). We focus on the computational complexity of several problems related to greedy partitioning. In particular, we show that given a graph G and an integer k deciding if the greedy...
-
Classification of submandibular salivary stones based on ultrastructural studies
PublikacjaIntroduction: Sialolithiasis remains a clinical problem with unclear etiopathogenesis, lack of prevention methods, and only surgical treatment. Materials and methods: An ultrastructure examination of submandibular sialoliths obtained from patients with chronic sialolithiasis was conducted using a scanning electron microscope and X-ray photoelectron...
-
Projektowanie układów VLSI
Kursy OnlineThe lecture tackles the problem of VLSI circuits design. In particular, emphasis is put on the physical design stage and more in-depth discussion of its sub-components. The lecture also focus on detailed explanation of selected numerical algorithms that are utilized in the course of physical design. Specialization: Microelectronic Systems
-
New Approach to Noncasual Identification of Nonstationary Stochastic FIR Systems Subject to Both Smooth and Abrupt Parameter Changes
PublikacjaIn this technical note, we consider the problem of finite-interval parameter smoothing for a class of nonstationary linear stochastic systems subject to both smooth and abrupt parameter changes. The proposed parallel estimation scheme combines the estimates yielded by several exponentially weighted basis function algorithms. The resulting smoother automatically adjusts its smoothing bandwidth to the type and rate of nonstationarity...
-
Clonal selection algorithm for vehicle routing
PublikacjaOver the years several successful computing techniques have been inspired by biological mechanisms. Studies of the mechanisms that allow the immune systems of vertebratesto adapt and learn have resulted in a class of algorithms called artificial immune systems. Clonal selection is a process that allows lymphocytes to launch a quick response to known pathogens and to adapt to new, previously unencountered ones. This paper presents...
-
A novel genetic approach to provide differentiated levels of service resilience in IP-MPLS/WDM networks
PublikacjaThis paper introduces a novel class-based method of survivable routing for connection-oriented IP-MPLS/WDM networks, called MLS-GEN-H. The algorithm is designed to provide differentiated levels of service survivability in order to respond to varying requirements of end-users. It divides the complex problem of survivable routing in IP-MPLS/WDM networks into two subproblems, one for each network layer, which enables finding the...
-
Generalized Savitzky–Golay filters for identification of nonstationary systems
PublikacjaThe problem of identification of nonstationary systems using noncausal estimation schemes is consid-ered and a new class of identification algorithms, combining the basis functions approach with localestimationtechnique,isdescribed.Unliketheclassicalbasisfunctionestimationschemes,theproposedlocal basis function estimators are not used to obtain interval approximations of the parametertrajectory, but provide a sequence of point...
-
Influence analysis of selected parameters on the ASON/GMPLS control plane performance
PublikacjaThe paper regards the problem of ASON/GMPLS performance. The authors present influence analysis of selected parameters on the ASON/GMPLS control plane performance represented in mean Connection Set-up Time E(CST) and mean Connection Release Time E(CRT). The selected parameters are: offered traffic, request intensity and proportion of requests class. The influence analysis is performed with simulation method by using OMNeT++ discrete-event...
-
Robustified estimators of radar elevation angle using a specular multipath model
PublikacjaWe consider the problem of estimating the elevation angle in the presence of multipath. The proposed method belongs to the class of maximum likelihood-like estimators and employs a modified specular reflection model that accounts for the uncertainty of the steering vector by assuming that they are subject to unknown deterministic perturbations with bounded norms. The analysis, performed using convex optimization methods, allows...
-
The complexity of bicriteria tree-depth
PublikacjaThe tree-depth problem can be seen as finding an elimination tree of minimum height for a given input graph G. We introduce a bicriteria generalization in which additionally the width of the elimination tree needs to be bounded by some input integer b. We are interested in the case when G is the line graph of a tree, proving that the problem is NP-hard and obtaining a polynomial-time additive 2b-approximation algorithm. This particular...
-
Cooperation of mono- and bi-articular muscles: human lower limb
PublikacjaObjectives: The aim of this study was to create and analyze a Pareto-optimal problem that would describe cooperation between mono- and bi-articulate lower limb muscles in sagittal plane. Methods: Equations describing the problem were derived and analyzed, additional constrains were introduced and experimental verification based on gait video analysis was performed. Results: Uncertainty of Pareto-optimal solution is shown for the...
-
Eqiuitable coloring of corona products of cubic graphs is harder than ordinary coloring
PublikacjaA graph is equitably k-colorable if its vertices can be partitioned into k independent sets in such a way that the number of vertices in any two sets differ by at most one. The smallest k for which such a coloring exists is known as the equitable chromatic number of G. In this paper the problem of determinig the equitable coloring number for coronas of cubic graphs is studied. Although the problem of ordinary coloring of coronas...
-
Multiclass AdaBoost Classifier Parameter Adaptation for Pattern Recognition
PublikacjaThe article presents the problem of parameter value selection of the multiclass ``one against all'' approach of an AdaBoost algorithm in tasks of object recognition based on two-dimensional graphical images. AdaBoost classifier with Haar features is still used in mobile devices due to the processing speed in contrast to other methods like deep learning or SVM but its main drawback is the need to assembly the results of binary...
-
THIRD-ORDER EXPONENTIAL INTEGRATOR FOR LINEAR KLEIN–GORDON EQUATIONS WITH TIME AND SPACE-DEPENDANT MASS
PublikacjaAllowing for space- and time-dependance of mass in Klein–Gordon equations re- solves the problem of negative probability density and of violation of Lorenz covariance of interaction in quantum mechanics. Moreover it extends their applicability to the domain of quantum cosmology, where the variation in mass may be accompanied by high oscillations....
-
THIRD-ORDER EXPONENTIAL INTEGRATOR FOR LINEAR KLEIN–GORDON EQUATIONS WITH TIME AND SPACE-DEPENDANT MASS
PublikacjaAllowing for space- and time-dependance of mass in Klein–Gordon equations re- solves the problem of negative probability density and of violation of Lorenz covariance of interaction in quantum mechanics. Moreover it extends their applicability to the domain of quantum cosmology, where the variation in mass may be accompanied by high oscillations....
-
Fundamentals of Physics-Based Surrogate Modeling
PublikacjaChapter 1 was focused on data-driven (or approximation-based) modeling methods. The second major class of surrogates are physics-based models outlined in this chapter. Although they are not as popular, their importance is growing because of the challenges related to construction and handling of approximation surrogates for many real-world problems. The high cost of evaluating computational models, nonlinearity of system responses,...