This paper presents a novel image segmentation algorithm, which uses a biologically inspired para... more This paper presents a novel image segmentation algorithm, which uses a biologically inspired paradigm known as swarm intelligence to segment images. A more efficient MRF based clustering algorithm that incorporated the Markov Random Field (MRF) theory and the Quantum Particle Swarm Optimization (QPSO) algorithm is proposed. the QPSO algorithm is ised to optimize the energy function which is a combinatorial optimization problem because of the large search space. The QPSO algorithm has few parameters and is simpler and more powerful than the classical version of the Particle Swarm Optimization algorithm. The experiments performed on both synthetic and real images show that the QPSO-MRF algorithm generates good results in clustering images and outperforms those obtained with other optimization methods.
Credit scoring has become very important issue due to the recent growth of the credit industry, s... more Credit scoring has become very important issue due to the recent growth of the credit industry, so the credit department of the bank faces a large amount of credit data. Clearly it is impossible analyzing this huge amount of data both in economic and manpower terms, so data mining techniques were employed for this purpose. So far many data mining methods are proposed to handle credit scoring problems that each of them, has some prominences and limitations than the others, but there is no a comprehensive reference introducing most used data mining method in credit scoring problem. The aim of this study is providing a comprehensive literature survey related to applied data mining techniques in credit scoring context. Such reference can help the researchers to be aware of most common methods in credit scoring evaluation, find their limitations, improve them and suggest new method with better capabilities. At the end we notice the limitation of the most proposed methods and suggest the more applicable method than other proposed.
This paper presents a novel image segmentation algorithm, which uses a biologically inspired para... more This paper presents a novel image segmentation algorithm, which uses a biologically inspired paradigm known as swarm intelligence to segment images. A more efficient MRF based clustering algorithm that incorporated the Markov Random Field (MRF) theory and the Quantum Particle Swarm Optimization (QPSO) algorithm is proposed. the QPSO algorithm is ised to optimize the energy function which is a combinatorial optimization problem because of the large search space. The QPSO algorithm has few parameters and is simpler and more powerful than the classical version of the Particle Swarm Optimization algorithm. The experiments performed on both synthetic and real images show that the QPSO-MRF algorithm generates good results in clustering images and outperforms those obtained with other optimization methods.
Credit scoring has become very important issue due to the recent growth of the credit industry, s... more Credit scoring has become very important issue due to the recent growth of the credit industry, so the credit department of the bank faces a large amount of credit data. Clearly it is impossible analyzing this huge amount of data both in economic and manpower terms, so data mining techniques were employed for this purpose. So far many data mining methods are proposed to handle credit scoring problems that each of them, has some prominences and limitations than the others, but there is no a comprehensive reference introducing most used data mining method in credit scoring problem. The aim of this study is providing a comprehensive literature survey related to applied data mining techniques in credit scoring context. Such reference can help the researchers to be aware of most common methods in credit scoring evaluation, find their limitations, improve them and suggest new method with better capabilities. At the end we notice the limitation of the most proposed methods and suggest the more applicable method than other proposed.
Uploads
Papers by Mohamed batouche