In the future, medical diagnosis by artificial intelligence is expected to become widespread. Therefore, we considered a method for diagnosing a disease based on a checklist using Bayesian inference. In this paper, we describe the development of basic theory and algorithms for that purpose. Nowadays, neural networks or deep learning have become synonymous with artificial intelligence (AI), and the idea that human intellectual work will soon be replaced by AI has been born. However, deep learning is not perfect. For example, it is known that there are many problems such as \"inference is a black box\", \"unexpected answer due to overfitting\", and \"large-scale network and long-time learning\". Bayesian inference can provide learning and inference that is completely different from neural networks. Therefore, it may be possible to overcome the problems of neural networks. In this paper, we discuss the application of Bayesian inference to disease diagnosis.
The relationship marketing approach in the service sectors, focusing on establishing relations with customers, has drawn considerable attention mainly because it paves the way for promoting Customer loyalty and keeping loyal Customers as a sustainable competitive advantage and a valuable asset in line with disposing with the risk of losing customers by companies due to the prevailing conditions governing business such as the unprecedented growth of service industries, globalization,the financial crises hitting the world economy, increasing competitiveness in markets, maturity of many of industries and finally an increase in Customers’ awareness and understanding of pecuniary values in the context of commodities and services.This paper concentrates on investigating the effect of service quality, price perception, brand image, and tangible rewards as relationship marketing tactics on customers’ loyalty in the mobile communication networks with the mediation of trust, and commitment as relationship quality indicators. The research is a descriptive applied research follows a survey methodology and the statistical population consists of all Irancell Sim Card Users in the city of Tehran. The data is compiled from questionnaire that were itemized based on the comparable previous studies. The data are analyzed based on the Partial Least Square approach within structural equation modeling method, using the Smart PLS 2.0Beta software.The study verified the indirect effect of price perception, brand image, and tangible rewards on customers’ loyalty while mediated by trust and commitment, but it could not verify the indirect effect of services quality on Customers’ loyalty , mediated by trust and commitment.
This work presents the design of an advanced N-250 Military version (N-250 TRD) configuration design. The N-250 TRD is designed from its high speed predecessor, the N-250 civil version. Like its predecessor, the N-250 TRD is equipped with two 2,439 kW (3,271 shp) Rolls-Royce Allison AE 2100C turboprops, each driving a Dowty Rotol R384/6123-FX8 six-blade propeller. Its main different from N-250 (civil version) is N-250 TRD adopt beavertail after body (CN-235 type) instead of upswept after body. Its mission capabilities are troops transport mission, paratroops dropping mission, cargo and LAPES (Low Altitude Parachute Extraction System).
The Digital Bridge is a new generation of DC resistance test instrument for cables and wires. It can automaticallyselect the electric current according to dierent cables and wires, and displays the test results rapidly. This paperdesigns a kind of digital electric bridge instrument with self-calibration based on ARM9 core processor. This instru-ment can complete high-speed data acquisition, data recording and processing functions, and possesses custom RS232based communication protocol for data communication with the host computer. In order to eliminate the error ofADC chip and improve the measurement accuracy, the instrument uses a self-calibration method on both hardwareand software.
Non-Ignorable missing values are often a great challenge for data mining analysts. Imputation of such missing values requires models which vary depending on the dataset and distribution of values within the dataset. Statistical techniques should be implemented on the outcome of the model to impute the values. The paper proposes a new algorithm called Bayesian Genetic Algorithm (BAGEL) which combines both Bayesian principles and Genetic Algorithm. The model is created by the Bayesian method using the available values and it drives the Genetic Algorithm to find the most suited value to be substituted for missing value. BAGEL is implemented on real datasets and the results are reported in this paper. It is found that BAGEL imputes both discrete and continuous missing values with more accuracy than other existing methods at different missing rates ranging from 5% to 50%.
Different stock markets are interrelated due to the common fundamentals which influence the movement in the respective indices. Using daily stock returns from 2000 to 2010 for 21 countries paired with Pakistan, and using the Geweke methodology, the paper investigates the degree to which these stock markets are integrated with the Pakistani stock market. The paper also explores the factors that have an effect on the level of economic integration by applying fixed effect model. The result demonstrates statistically significant and high percentage of contemporaneous association between the 21 economies of the world and Pakistan. Greater comovement was observed between the equity markets during the period when Pakistani capital market and economy experience performance but less comovement was noted when KSE was under crises. In broad, an elevated export share by other economies to Pakistan has good impact on stock market comovement as well as a larger difference in inflation rates, has an inverse effect on stock market comovement between pairs of countries. On the other hand volatility in bilateral exchange rate also influences stock market comovement.