Artificial intelligence is a marginal discipline, used to simulate people's thinking, has attracted many disciplines, and there are more and more practical significance, and many scientists in different professional backgrounds are in artificial intelligence. Some new Thinking and new ways. As a computer science involves a branch of a smart computer system, these systems present the characteristics associated with intelligent behavior of humanity. The main areas of artificial intelligence include problem solving, language processing, automatic theorem certification, intelligent data retrieval, and other fields. These comprehensive concepts have important applications in natural language processing, intelligence retrieval, automatic programming, mathematical proven. The first big achievement of artificial intelligence is to develop a chess program that can solve the problem. Other techniques applied in the play chess program also include dividing difficult problems into some relatively easy sub-problems, developing artificial intelligence basic techniques such as search and problem statutes. Today's computer programs are capable of defeating the world championship of human beings, and has shown the power of artificial intelligence. Problem Solutions and Search is a big topic of artificial intelligence, which refers to many core concepts involving regulations, infer, planning, and related processes. The problem solving is a very vulnerable topic, and it is generally aware of all computer science. Here we only discuss the narrow problem solving. After people analyzing the problem of problem-use of artificial intelligence research, it is found that many methods are solved by the method of trying to search. Among them, the problems and game issues provide a rich source, the following chess problems are examples to analyze the artificial intelligence principles represented by the problem. In order to achieve a program that can play chess, we use the state space to solve the problem. The first thing to establish a mathematical model for chess, using one or more suitable data structures to represent chess. This has a logic problem that establishes a model, and the appropriate logic will play an important role in the subsequent quizzes. A relatively simple method is to give different values to different chess pieces, which give "king" to exceed the weights of other chess pieces, then determine a target function to reduce the total weight of the other party, you can Get a relatively simple way. Of course, if such a model is too simple, the effect will not be very good. After we get a mathematical model, the problem is broken down into a child problem that can be understood by this model. The process of seeking the question of the question is a process of tempting the search, combining the rules of chess and the might of operation of the target function, can guide the next child of the next child, that is, a collection of a state. Then, the next state set is then derived from this status set, so repeatedly, you can get a tree structure, and use a series of rules and search techniques in this structure, it is possible to determine a reasonable way. . Obviously, if the accuracy of the state space is, the higher the complexity of the attribute structure. As can be seen from the above, the main techniques similar to the above state space solve problem include the state description, describing the target status and search strategies. The way of searching the strategic simulator's thinking process is a key part of the expression of algorithm. The main search strategy includes a width priority search, depth priority search, and unpreacted search. The status space method uses the powerful computing power of modern computers, as far as possible, all possible states, is the most application-applied artificial intelligence theoretical branch. Another method different from the status space method is a method of the problem statute. In the method of the problem statute, the problem description or goal is its main data structure. The description of the problem is known, then through a series of transformations, this problem will eventually become a collection of sub-problems; the solution of these sub-problems can be obtained directly, thereby solving the initial problem. For example, the famous "Vatican issue" is a problem that can be solved like this. It can be seen that there is a problem with the issue of the problem indicating that there can be three parts: a description of an initial problem, a set of operators that turn problems into sub-problems, a description of this original issue. Artificial intelligence includes a wide range of people, and the solution is only an important aspect. Other aspects include a series of issues such as predicate calculations, rules interpretation systems, robotics, and expert systems.
