Cellular neural network with trapezoidal activation function

This paper presents a cellular neural network (CNN) scheme employing a new non‐linear activation function, called trapezoidal activation function (TAF). The new CNN structure can classify linearly non‐separable data points and realize Boolean operations (including eXclusive OR) by using only a single‐layer CNN. In order to simplify the stability analysis, a feedback matrix W is defined as a function of the feedback template A and 2D equations are converted to 1D equations. The stability conditions of CNN with TAF are investigated and a sufficient condition for the existence of a unique equilibrium and global asymptotic stability is derived. By processing several examples of synthetic images, the analytically derived stability condition is also confirmed. Copyright © 2005 John Wiley & Sons, Ltd.     

Validation of the Adult Suicidal Ideation Questionnaire and the Reasons for Living Inventory in an adult psychiatric inpatient sample.

This study investigated the factor structure of the Adult Suicidal Ideation Questionnaire (ASIQ) and the Linehan Reasons for Living Inventory (LRFI) in a sample of 205 adult psychiatric inpatients. Confirmatory factor analyses provided moderate support for the construct validity of each instrument. Coefficient alphas for the ASIQ (.98) and LRFL (.93) were high. In addition, a range of different clinical cutoff points was derived for each instrument. Both instruments were also better than chance in differentiating between the suicide attempter and psychiatric control groups. High ASIQ and low LRFL scores were significantly associated with scores on selected Minnesota Multiphasic Personality Inventory-2 Content scales. The analyses also indicated that only the ASIQ added to the symptoms of hopelessness and negative affect in differentiating between the suicide attempter and psychiatric control groups. Results suggest that both instruments may be useful screening tests for suicidal behavior in psychiatric long-term care inpatient samples. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Potential Anomaly Separation and Archeological Site Localization Using Genetically Trained Multi‐level Cellular Neural Networks

In this paper, a supervised algorithm for the evaluation of geophysical sites using a multi‐level cellular neural network (ML‐CNN) is introduced, developed, and applied to real data. ML‐CNN is a stochastic image processing technique based on template optimization using neighborhood relationships of the pixels. The separation/enhancement and border detection performance of the proposed method is evaluated by various interesting real applications. A genetic algorithm is used in the optimization of CNN templates. The first application is concerned with the separation of potential field data of the Dumluca chromite region, which is one of the rich reserves of Turkey; in this context, the classical approach to the gravity anomaly separation method is one of the main problems in geophysics. The other application is the border detection of archeological ruins of the Hittite Empire in Turkey. The Hittite civilization sites located at the Sivas‐Altinyayla region of Turkey are among the most important archeological sites in history, one reason among others being that written documentation was first produced by this civilization.     

The lower and upper approximations in a quotient hypermodule with respect to fuzzy sets

This paper provides a continuation of ideas presented by Davvaz and Mahdavipour [B. Davvaz, M. Mahdavipour, Roughness in modules, Inform. Sci. 176 (2006) 3658-3674]. The notion of hypermodule is a generalization of the notion of module. In this paper, we consider the quotient hypermodule M/A and interpret the lower and upper approximations as subsets of the quotient hypermodule M/A. Then, we introduce the concept of quotient rough sub-hypermodule. Also, using the concept of fuzzy sets, we introduce and discuss the concept of fuzzy rough hypermodules and then we obtain the relation between fuzzy rough sub-hypermodules and level rough sets. This relation is characterized as a necessary and sufficient condition.     

Suicide risk assessment in a college student population.

A group of 211 students at a midwestern university completed the Suicidal Behaviors Questionnaire (M. M. Linehan & S. L. Nielsen, 1981), Adult Suicidal Ideation Questionnaire (W. M. Reynolds, 1991a), Multi-Attitude Suicide Tendency Scale (I. Orbach et al., 1991), Beck Helplessness Scale (A. T. Beck, A. Weissman, D. Lester, & L. Trexler, 1974), and the Reasons for Living Inventory (M. M. Linehan, L. J. Goodstein, S. L. Nielsen, & J. A. Chiles, 1983) to determine if this group of commonly used self-report measures can distinguish between individuals with high and low levels of suicidal ideation and history of self-harmful behaviors. Exploratory principal-axis factor analysis resulted in an interpretable 2-factor solution accounting for 36.2% of the variance in suicidality. Support for convergent validity of the chosen measures was also found. It appears that rapid, accurate assessment of university student suicide risk is possible. Implications for reduction of suicide risk in this segment of the population are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Performance of joint multilevel/AES‐LDPCC‐CPFSK schemes over wireless sensor networks

In wireless sensor networks, data encryption and channel coding are considered together for ensuring secure and robust communication. In order to achieve this purpose, we introduce a new joint scheme, namely ‘Multilevel/Advanced Encryption Standard‐Low Density Parity Check Coded‐Continuous Phase Frequency Shift Keying (ML/AES‐LDPCC‐CPFSK)’. AES algorithm is the most powerful and widely used symmetric key cryptography in providing secure data transmission. LDPC codes have very large Euclidean distance and use iterative decoding algorithms. In this study, we have increased error performance employing multilevel structure to AES and LDPC. In all communications systems, phase discontinuities of modulated signals result in extra bandwidth requirements. CPFSK, which is a special type of continuous phase modulation, is a powerful solution for this problem. In this paper, we simulate error performance of ML/AES‐LDPCC‐CPFSK for regular LDPC codes. Simulation results are drawn for 4CPFSK, 8CPFSK and 16CPFSK over wireless cooperative sensor networks. Using this scheme, we are able to improve bit error performance, channel throughput, security level of communication and reduction in complexity compared with related schemes such as various turbo code structures. Copyright © 2009 John Wiley & Sons, Ltd.     

Negative capacitance peculiarities in a-Si:H/c-Si rectifier structure

Nontrivial negative capacitance (NC) effect, observed in a-Si:H/c-Si heterostructure devices, is discussed emphasizing the theoretical interpretation of experimental data. To explain NC effect, we have performed dark current voltage (I-V) and admittance measurements (C-V, G-V, C-f and G-f). The calculated values of series resistance (R_s) and barrier height (Φ_Bo) have the values from 100 to 114.7 Ω and 0.94 to 0.83 eV, respectively. Also, below 50% helium dilution rate, diode ideality factor (n) becomes bigger than 2, because tunneling at junction interface plays a major role. The measured room temperature (294 K) dark I-V result has been used during the fitting process for suggested capacitance model (Eq. (18)). The measured NC values exhibit strongly voltage depended behavior. This unexpected behavior is attributed to the presence of inductively coupled space charge region which might possibly be stemmed from the helium diluted a-Si:H material. It is seen that the measured NC values are well fitted with suggested capacitance model (Eq. (18)). Application of suggested correction formula on to experimental C-V data yields satisfactory results. It is shown that the calculated inductance values of the investigated device range from 10 to 42 μH and after correction, NC values are no longer observed in the C_d-V data.     

Effects of starter cultures and additives on the quality of Turkish style sausage (sucuk)

The effects of starter cultures and additives on the quality of Turkish style sausage (sucuk) were investigated during ripening and storage. Total aerobic plate counts (aerobic bacteria), mould and yeast counts, pH, thiobarbituric acid (TBA) value and biogenic amine formation were followed. Aerobic bacteria increased (P<0.05) during the first 10 days of ripening to 9.26 log CFU/g and they decreased (P<0.05) to 5.70 at the end of the storage. Larger reductions (P<0.05) were observed in sausages made with high levels of potassium sorbate, nitrite and nitrate than in those containing low levels of additives and those without additives at the end of storage. During the first 3 days of ripening, the pH values of all sausages decreased (P<0.05) from 5.98 to about 4.53. Later, the pH values increased slowly, due to decomposition of acids to an approximately constant value of about 5.20. TBA values were higher (P<0.05) in the sausages made without additives than in the others. Sausages prepared using high concentration of additives had lower (P<0.05) concentrations of biogenic amines than the others. Addition of starter culture with high concentration of additives (nitrite, nitrate, α-tocopherol, ascorbic acid, potassium sorbate, potassium pyrophosphate and di-potassium hydrogenphosphate) decreased the formation of biogenic amine.