247 research outputs found
A novel semantic smoothing kernel for text classification with class-based weighting
No full textAltınel, Berna (Dogus Author), Diri, Banu (Dogus Author), Ganiz, Murat Can (Dogus Author) -- #articleinpress#Altınel, Berna (Dogus Author), Diri, Banu (Dogus Author), Ganiz, Murat Can (Dogus Author)In this study, we propose a novel methodology to build a semantic smoothing kernel to use with Support Vector Machines (SVM) for text classification. The suggested approach is based on two key concepts; class-based term weighting and changing the orthogonality of vector space. A class-based term weighting methodology is used for transformation of documents from the original space to the feature space. This class-based weighting basically groups terms based on their importance for each class and consequently smooths the representation of documents. This is accomplished by changing the orthogonality of the Vector Space Model (VSM) with introducing class-based dependencies between terms. As a result, on the extreme case, two documents can be seen as similar even if they do not share any terms but their terms are similarly weighted for a particular class. The resulting semantic kernel can directly make use of class information in extracting semantic information between terms, therefore it can be considered as a supervised kernel. For our experimental evaluation, we analyze the performance of the suggested kernel with a large number of experiments on benchmark textual datasets and present results with respect to varying experimental conditions. To the best of our knowledge, this is the first study to use class-based term weighting in order to build a supervised semantic kernel for SVM. We compare our results with kernels that are commonly used in SVM such as linear kernel, polynomial kernel, Radial Basis Function (RBF) kernel and with several corpus-based semantic kernels. According to our experimental results the proposed method favorably improves classification accuracy over linear kernel and several corpus-based semantic kernels in terms of both accuracy and speed
VIBRATIONAL SPECTRA OF THE MLCl COMPLEX FROM THEORETICAL CALCULATIONS
No full textAuthor Institution: Department of Physics, Mustafa Kemal University, Hatay, Turkey, 31034 (email to B.C.: [email protected])The geometric and vibrational parameters (harmonic and anharmonic frequencies) of the MLCl [M= Mn, Fe, Co, Ni, Cu, Zn, Cd, Hg; L= Ethylenediamine (en)] donor-acceptor complexes have been studied by using HF and MPW1PW91+iop(3/76=00572004280)/gen methods. Binding, reorganization, atomization, HOMO-LUMO and ionization potential energies have also been calculated with the same method. SQM calculations have been performed by using anharmonic frequencies and experimental data. The obtained results were found to be in good agreement with the corresponding experimental findings
Reducing CO HC Emissions in Two Stroke Two Wheelers 11th International Research Expert Conference
No full textAJTEC2011-44028 OPTIMIZATION OF THE RELATIONSHIP BETWEEN THE TWO-PHASE FRICTION FACTOR AND REYNOLDS EQUIVALENT NUMBER MODEL BY MEANS OF GENETIC ALGORITHM
No full textABSTRACT The two-phase friction factor of R134a in a coppe
Investigating Hemolytic Activity of Candida Isolates with Two Different Methods
No full textThe hemolytic activity of Candida isolates with agar and microplate methods were investigated and compared efficiency of these methods to assess relationship between hyphal formation and hemolysis
Genetic and tabu search approaches for optimizing the hall call-car allocation problem in elevator group systems
No full textThe most common problem in vertical transportation using elevator group appears when a
passenger wants to travel from a floor to other different floor in a building. The passenger
makes a hall call by pressing a landing call button installed at the floor and located near the
cars of the elevator group. After that, the elevator controller receives the call and identifies
which one of the elevators in the group is most suitable to serve the person having issued
the call. In this paper, we have developed different elevator group controllers based on
genetic and tabu search algorithms. Even though genetic algorithm has been previously
considered in vertical transportation problems, the use of tabu search approaches is a
novelty in vertical transportation and has not been considered previously. Tests have been
carried out for high-rise buildings considering diverse sizes in the group of cars. Results
indicate that the waiting time and journey time of passengers were significantly improved
when dealing with such soft computing approaches. Also, a quickly evaluable solution
quality function in the algorithms allows suitable computational times for industry
implementation
Genetic and tabu search approaches for optimizing the hall call- Car allocation problem in elevator group systems
No full textApplication of Genetic Algorithms on Ship Control Systems 10th International Research Expert Conference
No full text- …
