Artificial Intelligence

 Association Rules |Clustering |Decision Trees| Naive Bayes| Regression Analysis| Neural Networks| Time Series |Text Mining |

Genetic Programming

Genetic Programming

Genetic Programming is Evolutionary Programming Technique where computer programs automatically Evolved, traditionally represented in memory as tree structures. Trees can be easily evaluated in a recursive manner. Every tree node has an operator function and every terminal node has an operand, making Algorithms (mathematical expressions) easy to evolve and evaluate. The main operators used in evolutionary algorithms such as GP are crossover and mutation. Crossover is applied on an individual by simply switching one of its nodes with another node from another individual in the population. With a tree-based representation, replacing a node means replacing the whole branch. This adds greater effectiveness to the crossover operator. The expressions resulting from crossover are very much different from their initial parents. Mutation affects an individual in the population. It can replace a whole node in the selected individual, or it can replace just the node's information. To maintain integrity, operations must be fail-safe or the type of information the node holds must be taken into account. A Functions library is created which can be used within the Programs or later on for other Program. Automatically defined functions are very popular in Designing of latest Computer Chips or Designing of space or defense applications. Automatically defined functions can also Evolutionary that these functions evolved with time and become more fit as the evolved.

ADF facility consists of several classes: ADF, EDF, ADM, ADFStack, ADFContext, and ADFArgument. ADFs, EDF (Evolving Defined Functions) and ADMs (Automatically Defined Macros), appear as typical function nodes in a GP tree. However, they have a special associated tree in the individual's tree forest which they evaluate as a kind of a subfunction. As the program evolves the more more and more hardware resources are consumed and on very large systems this becomes a technology constraint.

In stock market applications of Genetic Programming 95% forecast done using Genetic Programming more accurate than Time Series and our experience has been for short term like next day values for last 3 years GP forecast values are more accurate by as much 5 to 10 %. This could be useful stock market traders, commodity traders and foreign exchange traders. In sudden rise or fall GP automatically tracks in 1 or 2 days and again forecast values have uncanny accuracy like 1% that is predicted values are out from actual values within 1%. Neural Networks are very effective in finding long term trends  hidden data regular trend analysis and all equations and algorithms used conventional technical and fundamental analysis are way out from Genetic Programming predictions.

Genetic programming is a recent development in the field of evolutionary algorithms which extends classical genetic algorithms by allowing the processing of non-linear structures. A genetic algorithm is a randomized search procedure working on a population of individuals or solutions encoded as linear bit strings. This population evolves over time through the application of operators which mimic those found in nature, namely, selection of the fittest, crossover and mutation. Genetic programming allows the evolution of programs encoded as tree structures. These programs are constructed from a predefined set of functions and terminals (which may be variables, like the state variables of a particular system, or constants, like integer 3 or boolean False). The evolution of programs within the genetic programming framework can be summarized as follows:
Initialization. Create an initial random population of P programs and evaluate the fitness of each program by applying it on a set of fitness cases (examples). Set the current population to this initial population.
Selection. Select P programs in the current population (with replacement). The selection process is probabilistically biased in favor of the best programs.
Modification. Apply reproduction or crossover to the selected programs.
Evaluation. Evaluate the fitness of each offspring in the new population.
Set the current population to the new population of offspring.
Repeat steps 2–6 for a predefined number of generations or until the system does not improve anymore.
The final result is the best program generated during the search.