Artificial Intelligent Course (CS 302) PDF

Artificial Intelligent Course (CS 302) Computer Science Dept. Artificial Intelligent (AI) In short, it is the behavior of certain properties characteristic of the software makes it simulates human mental abilities and patterns of work. The most important of...

Artificial Intelligent Course (CS 302) Computer Science Dept. Artificial Intelligent (AI) In short, it is the behavior of certain properties characteristic of the software makes it simulates human mental abilities and patterns of work. The most important of these properties the ability to learn and the conclusion and the reaction to the situation was not programmed in the machine. Or is an effort to develop systems based on the computer to give it the ability to perform the functions mimic PEOPLE him the human mind in terms of learning languages, complete administrative tasks, and the ability to think, learn and understand. Philosophy of Intelligent Intelligent has been defined in many different ways such as in terms of one's capacity for logic, abstract thought, understanding, self-awareness, communication, learning, emotional knowledge, memory, planning, creativity and problem solving. It can also be more generally described as the ability to perceive information and retain it as knowledge for applying to itself or other instances of knowledge or information, thereby creating referable understanding models of any size, density, or complexity, due to any conscious or subconscious imposed will or instruction to do so. I think it is the ability to take information, process it and then be able to apply it to new situations. If one cannot apply what one has learned then they aren't very intelligent. It takes intelligent to use facts and old knowledge in new situations so that you can create new results. AI properties 1- Learning ability The ability to learn one of the advantages of intelligent behavior and whether learning in humans is by observation or take advantage of the mistakes of the past, the artificial intelligent programs must rely on strategies to machine learning. The ability used to improve performance by taking into account previous mistakes. This susceptibility associated with susceptibility to disseminate information and conclusion similar selective cases of neglect and some extra information. Page 1 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. 2-Empirical research Artificial intelligent programs geared towards the problems do not have the solutions to be found depending on the specific logical steps. It follows the method of empirical research, these programs hack into matters that have no general way to solve known. This means that the programs do not use sequential steps that will lead to the right solution, but choose a certain way to resolve look good while retaining the possibility to change the way if it turns out that the first option does not lead to a solution quickly, any focus on adequate solutions and not to confirm the optimal or the precise solutions as is the case in the current conventional programs. With this in mind, the solution to the equations of the second degree is not from artificial intelligent programs because the method known but chess is one of the good examples of artificial intelligent software programs, to the absence of a clear and definite way to determine the next move. 3-Embracing knowledge and representation Because of the important characteristics in artificial intelligent programs to use the symbolic representation method in the expression of information, and follow the ways of empirical research in finding solutions to the artificial intelligent programs must possess to build a large base of knowledge containing a link between the cases and results, that artificial intelligent programs unlike statistical programs contain a method to represent the information as it is used to describe a private knowledge restructuring. This restructuring includes Facts and relationships between these facts and rules that bind these relationships. And a range of cognitive structures are among the knowledge base and this rule provides as much information as possible about the problem you want to find a solution to it. 4-Data unconfirmed or incomplete On programs that are designed in the field of artificial intelligent to be able to give solutions if the data is uncertain or incomplete, and that does not mean that you give the solutions, whatever the solution is wrong or correct, and should it be to the good performance to be able to give solutions accepted and not become deficient. Of the other qualities that can artificial intelligent programs carried out usability to find some solutions, even if the information is available in the entire time the solution requires that the consequences of lack of integration of information leads to a less realistic conclusions or less deserved Page 2 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. 5-Symbolic representation The representation of information through symbols and this representation is approaching the form of human representation of his information in his daily life, since they are dealing with non-numerical symbols and the opposite of what is known and acceptable in most computers today dealing with numerical quantities and numbers. Of course, there is nothing to prevent that the artificial intelligent programs as usual calculations using the extracted values in the highest level of decision-making. These characteristics enable the programs to deal with the knowledge discourse naturally helping to carry out processing software discretionary rather than digital processing routine known in the field of computers. 6. Scalability heuristics It is the ability to devise possible solutions to a particular problem and the reality known and a previous experience, particularly for problems that cannot be with the use of conventional means known to solve the data. This capability is recognized on a computer storing all the possible solutions in addition to the use of the laws or the laws of logic and reasoning strategies. Targets of AI The great progress that the world is witnessing in all fields, but some of the credit for which is due to the computing devices. It may be too early to talk about the virtues of Smart Computing, but undoubtedly the Smart Computing (that may use this expression) play a growing role in many areas for the time being and wait for them to reach a big affair in the near future in areas such as: 1) Engineering field in terms of the ability to develop and examine the design and implementation method steps. 2) Medical field in terms of diagnosis of pathological cases and prescribing them. 3) Military field in terms of decision-making time of the outbreak of the fighting and attitudes and preparation of plans and supervise their implementation analysis. 4) Educational field in terms of carrying out the teacher and give consultancy in the field of education. 5) In numerous other areas in factories control of production processes, and the displacement of workers in difficult environmental conditions, and in trade and Page 3 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. business as an analysis of the market situation and forecasting and the study of prices, and other fields. Applications of AI Artificial intelligent has been used in a wide range of fields including: 1. Computer science 2. Finance 3. Hospitals and medicine 4. Heavy industry 5. Online and telephone customer service 6. Transportation 7. Telecommunications maintenance 8. Toys and games 9. Music 10. Aviation 11. News, publishing and writing Characteristics of the languages of AI Characterized languages artificial intelligent characteristics fit the nature of artificial intelligent systems and characteristics are: 1-Viability knowledge representation It is intended to use special rules to describe the knowledge (Facts, Relations, Rules, and Frames). 2- Symbolic processing: Characterized languages AI to address the possibility of symbols and shapes. 3- Flexibility of control Traditional languages such as Pascal and C are addressing the problem through a serial follow the instructions of the program are always going to be unable to treat artificial intelligent problems so it came to languages artificial intelligent possibility of more flexible control. Overall Considered Languages Artificial Intelligent more efficient than traditional languages and mean time efficient implementation of the program and reduce the size of storage in memory, but we need an effort by the programmer to determine all the facts and linking them to other to extract and objectives desired results. Page 4 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. AI problems Split the problem of simulating intelligent to a number of specific sub-problems. This consists of certain attributes or capabilities that researchers would like an intelligent system embodied. Features mentioned below have received the most attention. 1. Finding, and logical thinking, and the ability to solve problems The researchers first placed in the science of artificial intelligent algorithms that mimic the sequential logical thinking being done by humans when solving puzzles, playing board games or logical conclusions. In the eighties and nineties, artificial intelligent research led to the reach of the means of highly successful for dealing with unconfirmed information or incomplete, using concepts of probability and the economy. 2. Knowledge representation It is the focus of artificial intelligent research. Many of the problems that are expected to be solved by machines the will require extensive knowledge of the world. Among the things that need to be represented by Artificial Intelligent: objects and properties and groups taxonomic relationships between objects; and attitudes and events, states and time; causes and effects; knowledge of knowledge (what we know about what people know) and many other areas. 3. Default thinking and problem of qualification Is much that people know "assumptions." For example, at the mention of the birds in a conversation, usually draws a picture of the human brain in an animal the size of a fist, singing, fly. Of course not all of these specifications apply to all birds. John McCarthy identified this problem in 1969, the problem of qualifications: each logical base is interested in artificial intelligent researchers to represent, many exceptions. There is almost nothing can simply say that it is true or not in the manner required by the abstract logic. Artificial intelligent research has explored a number of solutions to this problem. 4. Widening logical knowledge The average person knows a large number of facts about corn. Research projects that seek to build a complete knowledge base of logical, it must be built in a traditional manner where it is building the complex concepts, one by one. One of the main objectives that the Page 5 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. computer understands a multitude of concepts to be able to learn to read through sources such as the Internet. 5. Planning You must be smart agents will be able to set goals and achieve them they needed a way to visualize the future (you must have the ability to represent the case of human beings in this world, and be able to predict the extent of their ability to change it), and be able to choose to maximize the benefit. 6. Learning The machine learning pivotal in artificial intelligent researches since the beginning. Unsupervised learning is the ability to find patterns in a large number of inputs. Supervised learning includes both classification (ability to determine to which category something belongs in, after seeing a number of models for a number of things from several categories), and retreat (the discovery of the mechanism of continuing that will generate the outputs of the input, in light of the range of input and output numerical examples). 7. Mechanism of natural language Natural language processing gives machines the ability to read and understand the languages spoken by humans. Many researchers hope that the natural language processing system that is strong enough to gain knowledge of his own, by reading the existing text available over the Internet. Some direct applications of natural language processing include information retrieval (or text) analysis. 8. Movement and the possibility of change The field of robotics is closely related to artificial intelligent. Necessary robots intelligent to be able to handle tasks such as object manipulation and navigation, in light of localization sub-problems (that you know where you are), mapping (to know what is around you) and plan the movement (to know how to get there). 9-perception Machine perception is the ability to use input from sensors (such as cameras, microphones, sonar and other more exotic) to draw out aspects of the world. Computer vision is the ability to input visual analysis. Sub few of the problems: speech recognition face recognition and identification of objects. Page 6 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. Intelligent measures How can determine whether the worker is smart or not? In 1950, Alan Turing proposed a general procedure to test the intelligent of a factor now known as the Turing test. This procedure allows Baanbar most of the major problems of artificial intelligent. But it is a very difficult challenge at the moment and all the factors Aaty Houdat him failed. It can also be evaluated artificial intelligent in accordance with specific problems such as small problems in chemistry, handwriting recognition, and games. These tests are called Turing tests expert. The smaller scale of the problems, the number of achievable goals, and there is a growing number of positive results. Classified as artificial intelligent test results into the following groups: Optimization: it cannot be for better performance Man of supernatural powers: the performance of the best of all human beings Man of supernatural: better performance than most humans Less than human: worse than most humans performance For example, the performance of checkers (drafts) is optimal, performance in chess, falls under the "extraordinary man" and close to "human extraordinary powers" and performance in many everyday tasks performed by humans falls under the category of "less than human." There is a completely different approach based on measuring machine intelligent through derived from mathematical definitions of intelligent tests. Examples started on this kind of testing in the late nineties; Kachtbarat intelligent using concepts Andrey Kolmogorov such complexity and pressure as Marcus Hutter gave similar definitions of machine intelligent in his book Universal Artificial Intelligent (Springer 2005), which was developed again by Legg and Hutter of mathematical definitions features, it can be applied to non-human intelligent, and in the absence of examiners of humans. Why we are study AI Some futurists claim that artificial intelligent will exceed the limits of progress and humanity will change fundamentally changed. Use Ray Ray Kurzweil Moore's Law (which describes the significant improvement in digital precision hacks technology) to calculate that the computer will have the same processing power in human brains by the year 2029, and by Page 7 of 8 Artificial Intelligent Course (CS 302) Computer Science Dept. 2045 artificial intelligent will reach a point becomes then able to improve itself at more than everything imaginable in the past, a science fiction writer who coined by Vernor Vinge and he called the "technological singularity" scenario. Fredkin says that "artificial intelligent is the next stage in evolution," and went to great lengths to talk about George Dyson in his book of the same name in 1998. The expectation of many futurists and science fiction writers that humans and machines will merge in the future and become a cyborg (any system combines natural recipes artificial and recipes so he will more capable and powerful than both. this is called the idea "beyond the human," that's why we study artificial. Page 8 of 8