In this study, we apply several advanced machine learning techniques including extreme gradient boosting (XGBoost), support vector machine (SVM), and a deep neural network to predict bankruptcy using easily obtainable financial data of 3728 Belgian Small and Medium Enterprises (SME’s) during the period 2002–2012. Using the above-mentioned machine learning techniques, we predict bankruptcies with a global accuracy of 82–83% using only three easily obtainable financial ratios: the return on assets, the current ratio, and the solvency ratio. While the prediction accuracy is similar to several previous models in the literature, our model is very simple to implement and represents an accurate and user-friendly tool to discriminate between bankrupt and non-bankrupt firms.

No abstract

In this paper we reviewed the importance issues of the optical character recognition, gives more emphases for OCR and its phases. We discuss the main characteristics of Arabic language, furthermore it focused on the pre-processing phase of the character recognition system. We described and implemented the algorithms of binarization, dots removing and thinning which will be used for feature extraction phase. The algorithms are tested using 47,988 isolated character sample taken from SUST/ ALT dataset and achieved better results. The pre-processing phase developed by using MATLAB software.

Absract-This paper describes the efforts of SUST-ALT (Sudan University of Science and Technology-Arabic Language Technology group) research group towards building datasets for research in recognition of Arabic handwritten. The data sets contain: numerals datasets, isolated Arabic letters datasets, Arabic names datasets. Most of these datasets are off-line. The paper also describes some published results as well as future work.

Abstract. Optical Character Recognition (OCR) is one of the important branches. One segmenting words into character is one of the most challenging steps on OCR. As the results of advances in machine speeds and memory sizes as well as the availability of large training dataset, researchers currently study Holistic Approach "recognition of a word without segmentation". This paper describes a method to recognize off-line handwritten Arabic names. The classification approach is based on Hidden Markov models.. For each Arabic word many HMM models with different number of states have been trained. The experiments result are encouraging, it also show that best number of state for each word need careful selection and considerations.