Objective: Nowadays, financial distress prediction is one of the most important research issues in the field of risk management that has always been interesting to banks, companies, corporations, managers and investors. The main objective of this study is to develop a high performance predictive model and to compare the results with other commonly used models in financial distress prediction Methods: For this purpose, sequential floating forward selection that is considered as the generalized form of sequential forward selection method and as one of the wrapper methods, and sequential forward selection method in combination with support vector machine were used. These models are combined models of feature selection and classifier. Logistic regression model which is a statistical classification models, has also been used in the present study. Results: After reviewing the important financial ratios, 29 financial ratios that were mostly used in previous researches were chosen. Paired T-test results showed that with a 95% confidence level. The proposed model provides higher accuracy than other models used in this study. Conclusion: Results showed that the proposed model of this research has significantly better performance in predicting financial distress than the sequential forward selection method and Logistic regression model in one year, two years and three years before financial distress.

Intrusion detection system (IDS) is one of the most important security tools, which is used for detecting computer attacks. This System reacts based on two methods: misuse-based and anomaly-based detection. The time limitation to responding and using low efficiency algorithm is the biggest challenge for researchers to promote detection of attacks in IDS. One of the most significant stages in intrusion detection process is the accurate selection of features of IDS to promote the detection, based on these features. In this article, a new method is presented to determine the most effective features in IDS, based on misuse detection method. In this method, the features of NSL-KDD data set have been reduced by ant colony optimization in sequential forward feature selection algorithm, utilizing PART classification algorithm. For evaluating success rate of this method, a specific software in Java language was implemented, using the functions of the library of WEKA. The results compared with other successful methods show that this method increases detection accuracy rate, with concurrent detection of attack category, from 84. 1% to 85. 35%. Also, the detection time decreases from 0. 31 seconds to less than 0. 25 seconds in a data set of approximately twenty thousand members.

Background: In this paper we compare a highly accurate supervised to an unsupervised technique that uses breast thermal images with the aim of assisting physicians in early detection of breast cancer.Methods: First, we segmented the images and determined the region of interest. Then, 23 features that included statistical, morphological, frequency domain, histogram and gray-level co-occurrence matrix based features were extracted from the segmented right and left breasts. To achieve the best features, feature selection methods such as minimum redundancy and maximum relevance, sequential forward selection, sequential backward selection, sequential floating forward selection, sequential floating backward selection, and genetic algorithm were used. Contrast, energy, Euler number, and kurtosis were marked as effective features.Results: The selected features were evaluated by fuzzy C-means clustering as the unsupervised method and compared with the AdaBoost supervised classifier which has been previously studied. As reported, fuzzy C-means clustering with a mean accuracy of 75% can be suitable for unsupervised techniques.Conclusion: Fuzzy C-means clustering can be a suitable unsupervised technique to determine suspicious areas in thermal images compared to AdaBoost as the supervised technique with a mean accuracy of 88%.

In this paper, in the first step, the present weather of METAR reports of the year 2013 in Mehrabad synoptic station was studied and the period with most occurrences of the instability producing the Gusty wind was identified. This period is from January to June of every year. Then, all data of selected period, except the data of Gusty wind direction and speed, were normalized to interval 0. 1– 0. 9. The considered data for training, testing and validation were 60%, 20% and 20%, respectively. The related features of Gusty wind direction and speed were selected from 58 features recorded by 3 sensors located on the runway. The Mehrabad runway direction is from the east to the west with 4000 meters long and 45 meters wide. The sensor No. 29 was on the east end of band, the sensor No. 11 was on the west edge of the band, and location of the mid sensor was on the middle of band which its distance from the band is 600 meters to the north direction. The feature selection methods in this study are mutual information (MI) with the Maximum-Relevance Minimum-Redundancy criterion (filter type) and sequential floating forward selection (SFFS) (wrapper type) with the k Nearest Neighbors (kNN) algorithm. Selected features for Gusty wind speed at each band are the maximum and mean wind speed in 2 and 10 minutes, and the momentary wind speed by the MI method. The selected feature by SFFS method is the wind direction deviation in past 10 minutes on band No. 11 and mid band, momentary pressure on mid band and maximum wind speed in 10 minutes on band No. 29. For Gusty wind direction by first method, the selected features are minimum, mean and maximum wind direction in 2 minutes, minimum and mean wind direction in 10 minutes and momentary wind direction on band No. 29. Selected features with second method are the wind direction deviations in past 10 minutes on the band No. 29 and mid band, and the mean sea level pressure and mean wind direction in 10 minutes on band No. 29. In the final step, these selected features were used as inputs of the multilayer perceptron neural network in different modes such as: layer number, neuron number, learning rate and threshold value for weight of neuron. The model output results were compared to predict the Gusty wind direction and speed and the best model was selected. The results show that to predict the wind speed, the best model is a multilayer perceptron neural network with four layers: input layer with 4 neurons, two hidden layers with 4 neurons in the first layer and 2 neurons in the second layer and 1 neuron in the output layer; learning rate of 0. 1 and initial weight neurons of 0. 5. For predicting the wind direction, the best model has four layers, 6 neurons in the first and second layers and 3 neurons at the third layer and one neuron at the fourth layer with the same learning rate and initial threshold. The MLP performance is better in predicting the Gusty wind speed.

Biomedical datasets usually include a large number of features relative to the number of samples. However, some data dimensions may be less relevant or even irrelevant to the output class. selection of an optimal subset of features is critical, not only to reduce the processing cost but also to improve the classification results. To this end, this paper presents a hybrid method of filter and wrapper feature selection that takes advantage of a modified method of sequential forward floating search (SFFS) algorithm. The filtering approach evaluates the features for predicting the output and complementing the other features. The candidate subset generated by the filtering approach is used by k-fold cross validation of support vector machine (SVM) with user-defined classification margin as a wrapper. Applications of the proposed SFFS method to five biomedical datasets illustrate its superiority in terms of classification accuracy and execution time relative to the conventional SFFS method and another previously improved SFFS method.

Machine learning-based classification techniques provide support for the decision making process in the field of healthcare, especially in disease diagnosis, prognosis and screening. Healthcare datasets are voluminous in nature and their high dimensionality problem comprises in terms of slower learning rate and higher computational cost. Feature selection is expected to deal with the high dimensionality of datasets in terms of reduced feature set. Feature selection improves the performance of classification accuracy particularly performing with less number of features in decision making process. In this paper, Random Forest (RF) is employed for the diagnosis of cardiovascular disease. The first phase of the proposed system aims at constructing various feature selection algorithms such as Principal Component Analysis (PCA), Relief-F, sequential forward floating Search (SFFS), sequential Backward floating Search (SBFS) and Genetic Algorithm (GA) for reducing the dimension of cardiovascular disease dataset. The second phase switched to model construction based on RF algorithm for cardiovascular disease classification. The outcome shows that the combination with GA and RF delivered the highest classification accuracy of 93. 2% by the help of six features.

This paper proposes an innovative band selection method based on fuzzy clustering of bands for hyperspectral images.The main novelties of this research lie in two issues: i) bands representation in a new space called prototype space (PS) where bands take characteristic vector in terms of class reflectivity (where bands' feature vectors are defined in terms of class reflectivity), and ii) using uncertainty and angle measures to distinguish highly correlated and informative channels. Having clustered channels by Fuzzy C-Means (FCM) clustering in PS, the highly correlated bands fall in a cluster by uncertainty measure, then a nearest band to the cluster center is discerned as representative of those bands. Moreover uncertain bands as isolated bands are separated in PS where among isolated bands the bands that have large angle with diagonal of PS are indicated as informative channels. Since finding representative and informative bands depends on random initialization of FCM clustering, the bands clustering algorithm conducted many times. Accordingly, the proper bands are selected by maximizing overall accuracy of validation data set. Benchmarking on the challenging hyperspectral data set demonstrated relative merit of proposed method respect to the conventional band selection methods like sequential forward floating and sequential backward floating band selection.

We extend the method of adaptive two-stage sequential sampling to include designs where there is more than one criteria used in deciding on the allocation of additional sampling e ort.These criteria, or conditions, can be a measure of the target population, or a measure of some related population. We develop Murthy estimator for the design that is unbiased estimators for the population mean, and propose another, more efficient, estimator. We investigate asymptotic properties of this estimator. We use a simulation study to investigate design properties of the multi-criteria adaptive stratified sequential sampling scheme and also some estimator properties under the design.