Yazar "Temel, Turgay" seçeneğine göre listele

Listeleniyor 1 - 7 / 7
  • Küçük Resim Yok
    Öğe
    Finding number of clusters in single-step with similarity-based information-theoretic algorithm
    (Inst Engineering Technology-Iet, 2014) Temel, Turgay
    A single-step algorithm is presented to find the number of clusters in a dataset. An almost two-valued function called cluster-boundary indicator is introduced with the use of similarity-based information-theoretic sample entropy and probability descriptions. This function finds inter-cluster boundary samples for cluster availability in a single iteration. Experiments with synthetic and anonymous real datasets show that the new algorithm outperforms its major counterparts statistically in terms of time complexity and the number of clusters found successfully.
  • Küçük Resim Yok
    Öğe
    High-accuracy document classification with a new algorithm
    (Inst Engineering Technology-Iet, 2018) Temel, Turgay
    A new algorithm based on learning vector quantisation classifier is presented based on a modified proximity-measure, which enforces a predetermined correct classification level in training while using sliding-mode approach for stable variation in weight updates towards convergence. The proposed algorithm and some well-known counterparts are implemented by using Python libraries and compared in a task of text classification for document categorisation. Results reveal that the new classifier is a successful contender to those algorithms in terms of testing and training performances.
  • Küçük Resim
    Öğe
    A NEW CLASSIFICATION ALGORITHM: OPTIMALLY GENERALIZED LEARNING VECTOR QUANTIZATION (OGLVQ)
    (Acad Sciences Czech Republic, Inst Computer Science, 2017) Temel, Turgay
    We present a new Generalized Learning Vector Quantization classifier called Optimally Generalized Learning Vector Quantization based on a novel weight-update rule for learning labeled samples. The algorithm attains stable prototype/weight vector dynamics in terms of estimated current and previous weights and their updates. Resulting weight update term is then related to the proximity measure used by Generalized Learning Vector Quantization classifiers. New algorithm and some major counterparts are tested and compared for synthetic and publicly available datasets. For both the datasets studied, it is seen that the new classifier outperforms its counterparts in training and testing with accuracy above 80% its counterparts and in robustness against model parameter varition.
  • Küçük Resim
    Öğe
    A new high-performance current-mode fuzzy membership function circuit and its application
    (2015) Temel, Turgay
    A new current-mode fuzzy-membership function circuit is proposed. The circuit has a very compact and simple architecture based on the simultaneous use of winner- and loser-take-all topologies. The circuit has also the advantage of easily adjusting generated membership function characteristics such as center or mean and width with a straightforward application of respective input currents. It is shown that the new circuit outperforms previous counterparts in terms of speed, power consumption, layout area, and robustness to variations in design parameters and errors. As an application, a fuzzy-classifier is designed with a new membership circuit as a seven fuzzy level controller.
  • Küçük Resim
    Öğe
    A SINGLE-STEP CLUSTERING ALGORITHM BASED ON A NEW INFORMATION-THEORETIC SAMPLE ASSOCIATION METRIC DEFINITION
    (Acad Sciences Czech Republic, Inst Computer Science, 2017) Temel, Turgay
    A single-step information-theoretic algorithm that is able to identify possible clusters in dataset is presented. The proposed algorithm consists in representation of data scatter in terms of similarity-based data point entropy and probability descriptions. By using these quantities, an information-theoretic association metric called mutual ambiguity between data points is defined, which then is to be employed in determining particular data points called cluster identifiers. For forming individual clusters corresponding to cluster identifiers determined as such, a cluster relevance rule is defined. Since cluster identifiers and associative cluster member data points can be identified without recursive or iterative search, the algorithm is single-step. The algorithm is tested and justified with experiments by using synthetic and anonymous real datasets. Simulation results demonstrate that the proposed algorithm also exhibits more reliable performance in statistical sense compared to major algorithms.
  • Küçük Resim Yok
    Öğe
    Sliding-mode control approach for faster tracking
    (Inst Engineering Technology-Iet, 2012) Temel, Turgay; Ashrafiuon, H.
    A new sliding-mode control method is presented in which the nominal control is derived from a manifold consisting of the standard sliding manifold expression and its derivative. It is shown that the new method yields a faster reach time. Simulation results also show that it performs with smaller error and less control effort compared to standard counterparts in tracking an autonomous nonlinear affine system.
  • Küçük Resim Yok
    Öğe
    Sliding-mode speed controller for tracking of underactuated surface vessels with extended Kalman filter
    (Inst Engineering Technology-Iet, 2015) Temel, Turgay; Ashrafiuon, H.
    A sliding-mode speed controller for output tracking of underactuated surface vessels is presented based on a new manifold definition. The new sliding-mode control algorithm relies on a conventional sliding manifold and its time derivative. The states/outputs are estimated with the extended Kalman filter as an observer. The new controller is compared with its integral-type sliding-mode counterpart in terms of performance in tracking the outputs where the estimator is driven by respective control inputs and observed states involved with arbitrarily correlated process and measurement errors.