Tuberculosis disease diagnosis by using adaptive neuro fuzzy inference system and rough sets

A correct diagnosis of tuberculosis disease can be only stated by applying a medical test to patient’s phlegm. The result of this test is obtained after a time period of about 45 days. The purpose of this study is to develop a data mining solution that makes diagnosis of tuberculosis as accurate as possible and helps deciding whether it is reasonable to start tuberculosis treatment on suspected patients without waiting for the exact medical test results. We proposed the use of Sugeno-type “adaptive-network-based fuzzy inference system” (ANFIS) to predict the existence of mycobacterium tuberculosis. Data set collected from 503 different patient records which are obtained from a private health clinic (consent of physicians and patients). Patient record has 30 different attributes which covers demographical and medical test data. ANFIS model was generated by using 250 records. Also, rough set method was implemented by using the same data set. The ANFIS model classifies the instances with correctness of 97 %, whereas rough set algorithm does the same classification with correctness of 92 %. This study has a contribution on forecasting patients before the medical tests.     

Forecasting model of Shanghai and CRB commodity indexes

This paper examines the long-run relationship between the Shanghai index and CRB commodity index. We run our vector error correction model (VECM) for two sub-samples as pre-crisis period and post-crisis period. In pre-crisis period, there is strong bidirectional causality link between the Shanghai and CRB. In post-crisis period, there is no causality between the indices. In the second part of the article, we employ Fuzzy System Modeling (FSM) to increase the performances of root mean-square error, R² and Adjusted R². We show the results of our analysis for both Shanghai and CRB indexes. We have demonstrated the results for a good number of our investigations ANFIS, GENFIS, Classical LSE and three versions of support vector regression. For both Shanghai and CRB indexes, our FSMIFF with LSE obtains better results than all other models we have investigated and thus are more suitable for forecasting stable and unstable stock market behavior.     

Ferromagnetic Behavior of Fe+ Implanted Si(100) Semiconductor

Fe ions have been implanted into Si (100) single crystals using ion implantation technique. The Fe ions have been accelerated to 45 keV with a dose of 5×10¹⁷ ion/cm² at room temperature. The ions have been sent to the substrate??s surface at normal incidence. The temperature dependence of magnetization measurement was explored at the temperature range of 10?C300 K. The implanted Si substrate was studied with Ferromagnetic Resonance (FMR) technique and Vibrating Sample Magnetometer (VSM). The FMR spectra were recorded by applying external magnetic field in different experimental geometries. FMR spectra were analyzed and the magnetic properties, which are the g-factor, effective magnetization and uniaxial anisotropy parameter, were estimated by simulation of the experimental data. The sample showed two-fold magnetic anisotropic symmetry. By fitting the Si-2p region obtained through XPS measurements it is observed that Fe and Fe compounds are present in the material.     

Controlling phosphonic acid substitution degree on proton conducting polyphosphazenes

This paper is to describe the development of a synthetic strategy for the preparation of phosphonic acid functionalized poly(aryloxyphosphazene) membranes with different substitution degree of phosphonic acid. Synthesized polymers have been characterized by standard spectroscopic techniques; FT-IR, ¹H, ³¹P and ¹⁹F NMR and element analysis. Proton conductivities of phosphonic acid substituted polymers have been investigated with impedance spectroscopy at different temperatures. Furthermore, the correlations of the proton conductivity and ion exchange capacity with the substitution degree of phosphonic acid have also been investigated. Thermal properties and water uptake properties of the polymers are also investigated. It is found that the proton conductivity and initial decomposition temperature of the polymers increases up to an appropriate amount of phosphonic acid substitution degree, which can be a useful PEM candidate for fuel cells.     

Determination of the Intragrain Critical Current Density and Thermogravimetric Analysis on YBCO/Ag Superconductors Irradiated with 60Co Gamma Rays

We investigated the flux pinning behavior induced by gamma irradiation in Y₁Ba₂Cu₃O_(7–x) silver-added samples. The superconductors were prepared through solid-state route and added with silver in amounts of 0-, 6.5-, and 20-wt%, following which the samples were irradiated by high-energy gamma irradiation () at doses of 0, 500, and 1500 kGy at the dosage rate of 8.2 kGy/h. We performed magnetization loops to study the flux pinning of vortex in YBCO/Ag superconductors. We found that silver addition and irradiation on superconductors may increase the width in magnetization loops, which is related to the enhancement in the critical current density, J _c . We established that an interaction between silver addition and irradiation that results in higher values in J _c for YBCO/Ag irradiated samples already exists. The J _c in samples containing 20-wt% of silver decreased with irradiation because a higher content of silver produces small crystals and secondary phases growing. On comparing the TGA analysis for both YBCO and YBCO/Ag samples, we found higher-weight oxygen losses in YBCO silver samples, which suggests that silver increases the oxygen saturation content. The higher oxygen content found on irradiated specimens it is indicative of oxygen losses from CuO chains or CuO₂ planes reached by -irradiation.On leave from     

Selective Multiuser Detection for Cellular CDMA Systems

Multiuser detection-oriented CDMA systems have been anticipated to significantly improve system capacity in third-generation W-CDMA-based systems. However, they are greatly limited by the computational complexity of multiuser receivers. In this work, we propose a new, computationally efficient approach to multiuser detection (MUD), consisting in MUD of the subset of preselected users, and conventional detection of the rest of users, called selective multiuser detection (SMD). It allows for full exploitation of available processing power at the receiver by use of MUD and provides remedy for computational complexity of MUD techniques when the number of active users increases beyond the processing capability. We propose and examine three different criteria for selection of users to be processed by the multiuser receiver and analyze the capacity for the single-cell and the multicell CDMA cellular system. The capacity improvement with respect to the conventional CDMA detector combines the gain from MUD and reduction of other-cell interference. We apply the analysis to two SMD schemes using decorrelator and successive interference canceller (SIC) as the multiuser receiver. The results indicate that the SMD is a promising alternative for MUD-oriented CDMA systems with large numbers of active users.     

Security‐level classification for confidential documents by using adaptive neuro‐fuzzy inference systems

The security‐level detection of a confidential document is a vital task for organizations to protect their confidential information. Diverse classification rules and techniques are being applied by human experts. Increasing number of confidential information in organizations is making difficult to classify all the documents carefully with human effort. The recommended frameworks in this study classify the internal documents of TUBITAK UEKAE (National Research Institute of Electronics and Cryptology of Turkey) by using classification algorithms naïve Bayes, support vector machines (SVMs) and adaptive neuro‐fuzzy inference systems (ANFISs). A hybrid approach involving support vector classifiers and adaptive neuro‐fuzzy classifiers exposes the most successful accuracy rates of expert system classification. This study also states preprocessing tasks required for document classification with natural language processing. To represent term–document relations, a recommended metric TF‐IDF was chosen to construct a weight matrix. Agglutinative nature of Turkish documents is handled by Turkish stemming algorithms. At the end of the article, some experimental results and success metrics are projected with accuracy rates and receiver operating characteristic (ROC) curves.     

Wear behaviour of organic asbestos based and bronze based powder metal brake linings

Bronze based brake linings, were produced by powder metallurgy technique and their wear behaviour was investigated and compared to that of asbestos ones. Bronze powders were compacted under 350, 500 and 600 MPa pressures and sintered at 810 °C in ammonia atmosphere for 75 min. For the same friction distance, it was determined that temperature increase in the bronze based brake linings was lower than that of asbestos based ones. However, higher wear rate was observed in the bronze based brake linings. Moreover, thermal conductivity was decreased with high porosity level with low densities. Although, friction coefficient remained the same during breaking, an increase in wear resistance was observed.     

Artificial immune algorithm for IIR filter design

Over the recent years, several studies have been carried out by the researchers to describe a general, flexible and powerful design method based on modern heuristic optimisation algorithms for infinite impulse response (IIR) digital filters since these algorithms have the ability of finding global optimal solution in a nonlinear search space. One of the modern heuristic algorithms is the artificial immune algorithm which implements a learning technique inspired by human immune system. However, the immune system has not attracted the same kind of interest from researchers as other heuristic algorithms. In this work, an artificial immune algorithm is described and applied to the design of IIR filters, and its performance is compared to that of genetic and touring ant colony optimisation algorithms.