خرید و دانلود نسخه کامل کتاب Computational Intelligence Techniques for Trading and Investment – Original PDF

Machine intelligence dates back to 1936, when Turing proposed the idea of a universal processing machine (Turing, 1936), which composed the theoretical milestone in the theory of computability for the next years. Having Turing’s machine as a starting point, artificial intelligence (AI), in the last decades, focused on the principles, theoretical issues and design of algorith- mic methodologies inspired by observations of the natural environment. Some striking examples are the artificial neural networks (ANNs), which are inspired by the mammalian neural system, and evolutionary algorithms (EAs), inspired by natural selection and genetic evolution in biology. These techniques, along with numerous others, have found their way in solving real-world problems in biology, medicine, business, technology and finance. Computational intelligence (CI) (Engelbrecht, 2007) constitutes the sub- branch of AI that includes the design and development of theories and methods with a sound biological understanding alongside their application to solve real- world problems. According to Bezdek (1994): A system is computationally intelligent when it: deals with only numerical (low level) data, has pattern recognition components, does not use knowledge in the AI sense and additionally when it (begins to) exhibit i) computational adaptivity, ii) computational fault tolerance, iii) speed approaching human- like turn around and iv) error rates that approximate human performance. The above definition provides a central role for the definition of CI systems to the principles of computational adaptation, noisy data handling, computational speed and fault tolerance. The basic distinction of general AI systems and CI ones lies in their effort not only to imitate or surpass human intelligence, but also to aid humans to solve real-world problems more effectively, reducing their workload. The necessity for fulfilling human tasks more conveniently, costlessly and faster, and the need for processing high volumes of data has led to the growth of the CI field in the last two decades, with the development of a series of high- performance algorithms.