Marcin Korytkowski

Artificial Intelligence and Soft Computing

In recent years wind energy has been the fastest growing branch of the power generation industry. Maintenance of the wind turbine generates its the largest cost. A remote monitoring is a common method to reduce this cost. Growing number of monitored turbines requires an automatized way of support for diagnostic experts. Early fault detection and identification is still a very challenging task. A tool, which can alert an engineer about potentially dangerous cases, is required to work in real-time. The goal of this paper is to show an efficient system to online classification of operational states of the wind turbines and to detecting their early fault cases. The proposed system was designed as a hybrid of ART-2 and RBF networks. It had been proved before that the ART-type ANNs can successfully recognize operational states of a wind turbine during the diagnostic process. There are some difficulties, however, when classification is done in real-time. The disadvantages of using a classic ART-2 network are pointed and it is explained why the RBF unit of the hybrid system is needed to have a proper classification of turbine operational states.

Swarm and Evolutionary Computation

This paper presents a work inspired by the Pachycondyla apicalis ants behavior for the clustering problem. These ants have a simple but efficient prey search strategy: when they capture their prey, they return straight to their nest, drop off the prey and systematically return back to their original position. This behavior has already been applied to optimization, as the API meta-heuristic. API is a shortage of api-calis. Here, we combine API with the ability of ants to sort and cluster. We provide a comparison against Ant clustering Algorithm and K-Means using Machine Learning repository datasets. API introduces new concepts to ant-based models and gives us promising results.

Fast image classification by boosting fuzzy classifiers

This paper presents a novel approach to visual objects classification based on generating simple fuzzy classifiers using local image features to distinguish between one known class and other classes. Boosting meta-learning is used to find the most representative local features. The proposed approach is tested on a state-of-the-art image dataset and compared with the bag-of-features image representation model combined with the Support Vector Machine classification. The novel method gives better classification accuracy and the time of learning and testing process is more than 30% shorter.

From Ensemble of Fuzzy Classifiers to Single Fuzzy Rule Base Classifier

Neuro-fuzzy systems show very good performance and the knowledge comprised within their structure is easily interpretable. To further improve their accuracy they can be combined into ensembles. In the paper we combine specially modified Mamdani neuro-fuzzy systems into an AdaBoost ensemble. The proposed modification improves the interpretability of knowledge by allowing merging the subsystems rule bases into one knowledge base. Simulations on two benchmarks shows excellent performance of the modified neuro-fuzzy systems.

On Combining Backpropagation with Boosting

Boosting is a method for learning combined classifiers. In a boosting ensemble of classifiers trained by the backpropagation algorithm, the learning rate takes much smaller value comparing with the backpropagation applied alone. We propose a method which overcomes the above drawback and test it on neuro-fuzzy systems constituting a classifier ensemble using some well known benchmarks.

From Single Image to List of Objects Based on Edge and Blob Detection

In this paper we present a new method for obtaining a list of interest objects from a single image. Our object extraction method works on two well known algorithms: the Canny edge detection method and the quadrilaterals detection. Our approach allows to select only the significant elements of the image. In addition, this method allows to filter out unnecessary key points in a simple way (for example obtained by the SIFT algorithm) from the background image. The effectiveness of the method is confirmed by experimental research.

Modular type-2 neuro-fuzzy systems

In the paper we study a modular system which can be converted into a type-2 neuro-fuzzy system. The rule base of such system consists of triangular type-2 fuzzy sets. The modular structure is trained using the backpropagation method combined with the AdaBoost algorithm. By applying the type-2 neurofuzzy system, the modular structure is converted into a compressed form. This allows to overcome the training problem of type-2 neuro-fuzzy systems. An illustrative example is given to show the efficiency of our approach in the problems of classification.

Neuro-fuzzy Rough Classifier Ensemble

The paper proposes a new ensemble of neuro-fuzzy rough set classifiers. The ensemble uses fuzzy rules derived by the Adaboost metalearning. The rules are used in an ensemble of neuro-fuzzy rough set systems to gain the ability to work with incomplete data (in terms of missing features). This feature is not common among different machine learning methods like neural networks or fuzzy systems. The systems are combined into the larger ensemble to achieve better accuracy. Simulations on a well-known benchmark showed the ability of the proposed system to perform relatively well.

Adaboost ensemble of DCOG rough-neuro-fuzzy systems

Neural networks are able to perfectly fit to data and fuzzy logic systems use interpretable knowledge. These methods cannot handle data with missing or unknown features what can be achieved easily using rough set theory. In the paper we incorporate the rough set theory to ensembles of neuro-fuzzy systems to achieve better classification accuracy. The ensemble is created by the AdaBoost metalearning algorithm. Our approach results in accurate classification systems which can work when the number of available features is changing. Moreover, our rough-neuro-fuzzy systems use knowledge comprised in the form of fuzzy rules to perform classification. Simulations showed very clearly the accuracy of the system and the ability to work when the number of available features decreases.

Content-Based Image Indexing by Data Clustering and Inverse Document Frequency

In this paper we present an algorithm for creating and searching large image databases. Effective browsing and searching such collections of images based on their content is one of the most important challenges of computer science. In the presented algorithm, the process of inserting data to the database consists of several stages. In the first step interest points are generated from images by e.g. SIFT, SURF or PCA SIFT algorithms. The resulting huge number of key points is then reduced by data clustering, in our case by a novel, parameterless version of the mean shift algorithm. The reduction is achieved by subsequent operation on generated cluster centers. This algorithm has been adapted specifically for the presented method. Cluster centers are treated as terms and images as documents in the term frequency-inverse document frequency (TF-IDF) algorithm. TF-IDF algorithm allows to create an indexed image database and to fast retrieve desired images. The proposed approach is validated by numerical experiments on images with different content.