The first paper on intelligent computers
Research on the Application of Computer in Artificial Intelligence
Abstract: In recent years, with the continuous development of information technology and computer technology, the application of artificial intelligence in computers has become more and more in-depth, and it has been widely used in various fields of computers. This paper studies the application of computer in artificial intelligence, expounds the theoretical concepts of artificial intelligence, analyzes the problems existing in its application at present, and introduces the application of artificial intelligence in some fields.
Keywords: computer; Artificial intelligence; application research
I. Introduction
Artificial intelligence, also known as machine intelligence, came from Dartmouth Society in 1956, where it was first proposed? Artificial intelligence? This word. As a comprehensive subject, artificial intelligence is developed under the mutual infiltration of computer science, information theory, psychology, neurophysiology and linguistics. In computer application systems, artificial intelligence specializes in how to make intelligent systems or intelligent machines imitate the ability of human beings to carry out intelligent activities, thus extending human scientific intelligence. Artificial intelligence is a challenging science, and people engaged in this work must understand computer knowledge, psychology and philosophy. Artificial intelligence is at the technical application level of thinking science and is one of its application branches. Mathematics is often regarded as the basic science of many disciplines, and it has also entered the field of language and thinking, so the subject of artificial intelligence must borrow mathematical tools. Mathematics plays a role in the scope of standard logic and fuzzy mathematics, and enters the discipline of artificial intelligence, which will promote each other and develop rapidly.
Second, the problems of artificial intelligence in computer applications
(A) the weakness of computer language understanding. At present, computers can't accurately understand the complexity of language. However, computer language translators, which are in the initial stage of development, have shown extremely high sentence-making ability and understanding of algorithmic standardized sentences. However, it has not achieved remarkable success in understanding the meaning of sentences. Most of the information we get comes from the context and our own knowledge. People's personal views, social views and cultural views in daily life have great influence on the additional meaning of sentences.
(2) Doubt about pattern recognition. Computer research and pattern recognition have achieved good results to some extent, and some of them have been used as products, but their theories and methods are absolutely different from human sensory recognition mechanisms. People's thinking ability in images and recognition means can't be achieved even by the most advanced recognition system in computers. In addition, in the real society, life as a loosely structured task can be easily handled by ordinary livestock, but not by machines. This does not mean that it will never be, but it will be temporary.
Third, the application of artificial intelligence in some fields.
With the rapid development of AI technology, the development of various information technologies is closely related to artificial intelligence technology, which means that artificial intelligence has been widely used in various fields of computers. The following is the author's exposition on the application of artificial intelligence in some fields of computer. Details are as follows.
(A) artificial intelligence symbol calculation. As an important use of computers, scientific computing can be divided into two categories. The first is the calculation of pure value, such as finding the function value. Secondly, the calculation of symbols, also called algebraic operation, is an intelligent and fast calculation, and all the contents are symbols. Symbols can represent real numbers, integers, complex numbers and rational numbers, as well as sets, functions and polynomials. With the continuous development of artificial intelligence and the gradual popularization of computers, multifunctional computer algebra system software has appeared one after another, such as Maple or Mathematic. Because these softwares are written in C language, they can be used on most computers.
(2) Artificial intelligence is used for pattern recognition. Pattern recognition means that the computer studies the interpretation and automatic processing of patterns through mathematical techniques. The realization of computer pattern recognition is a breakthrough in the development of intelligent machines, which enables human beings to deeply understand their own intelligence. Its identification features are accurate, fast and efficient. Computer pattern recognition process is similar to human learning process, such as speech recognition. Speech recognition is to make the computer understand what people say.
Automatic translation, such as an automatic oral translation system in seven languages. The realization of this system enables people to buy air tickets, book hotels and exchange foreign currency when they go abroad, and they can communicate with them by mobile phones or mobile phones only through the Internet and telephone networks. Foreigners? Have a dialogue.
(3) The application of artificial intelligence in computer network security. At present, the common technologies in computer network security management mainly include intrusion detection technology and firewall technology. As one of computer network security devices, firewall plays an important role in computer network security management. The previous firewall didn't have the function of detecting encrypted Web traffic, because it couldn't see the data in the encrypted SSL stream, so it couldn't get the data in the SSL stream quickly and decrypt it. Therefore, the previous firewall can not effectively prevent the attack of the application. In addition, after the general application is encrypted, it can easily avoid the detection of the previous firewall. Therefore, it is difficult to prevent new attacks because the previous firewall cannot fully monitor the application data flow. The new firewall uses intelligent methods of statistics, probability and decision-making to identify data, so as to achieve the purpose of access authorization. But this method is mostly taken from the discipline of artificial intelligence, hence the name? Intelligent firewall? .
(d) Application of artificial intelligence in computer network system fault diagnosis. As an information processing system, artificial neural network is formed by simulating human cognitive process and the organizational structure of human brain. 1943, artificial neural network was first put forward and developed rapidly, becoming another branch of artificial intelligence technology. With its own advantages, such as associative memory, adaptive and parallel distributed processing, artificial neural network has been widely concerned in intelligent fault diagnosis, which has great potential and opens up a new road for the exploration of intelligent fault diagnosis. The diagnosis method of artificial neural network is different from that of expert system. It learns and trains through a large number of standard samples in the field, strengthens the adjustment of threshold and connection weight in artificial neural network, and makes the knowledge obtained from it hidden and distributed in the whole network, thus achieving the purpose of pattern memory in artificial neural network. Therefore, artificial neural network has a strong ability to capture knowledge, which can effectively deal with abnormal data and make up for the defects of expert system methods.
Four. Concluding remarks
In a word, artificial intelligence is the trend of computer technology, and its research theory and discovery determine the development prospect of computer technology. Nowadays, most of the research results of artificial intelligence have penetrated into people's daily life. Therefore, we should strengthen the research and development of artificial intelligence technology. Only by comprehensively analyzing the problems existing in its application in various fields and taking corresponding measures can it develop smoothly. The development of artificial intelligence technology will bring great influence to people's life, study and work.
References:
Angelababy. Realization and research of intelligent computer-aided teaching system [J]. Computer Knowledge and Technology, 2009,9
[2] Mao Yi. Research hotspot and development direction of artificial intelligence [J]. Technology and Market, 2008,3
[3] Li Deyi. Research and development of artificial intelligence in the network age [J]. Journal of Intelligent Systems, 2009, 1.
[4] Chen mai,. Research on the Application of Artificial Intelligence [J]. Journal of Xingtai Vocational and Technical College, 2008, 1
The second paper on intelligent computers
Reliability Analysis of Computer Network Based on Intelligent Computing
Abstract: Today's society is an information society, and the application of the network has spread throughout production, life, scientific research and other fields. Computer networking has become a trend, and the research on computer network reliability has been paid more and more attention by the computer industry. This paper mainly discusses the theoretical method of intelligent granular computing segmentation, realizes the minimum path set operation of computer network system by using dynamic array layering, and expounds the means of reliability analysis of computer network system.
Keywords: intelligent algorithm; Computer network; reliability analysis
1 Factors affecting the reliability of computer networks
1. 1 user equipment. User equipment is the terminal equipment provided for users. Whether its function is reliable or not has a profound impact on the user's experience and will also have an important impact on the reliability of the computer network. Ensuring the reliability of user terminals in use is an important part of daily maintenance in the process of computer network operation. The higher the interactive ability of the user terminal, the more reliable its network is.
1.2 transmission switching equipment. Transmission equipment includes transmission lines and transmission equipment. In practical work, if the computer network fault is caused by transmission lines, it is generally difficult to find out, and sometimes it takes a lot of work to find out the cause of this fault. Therefore, when installing transmission equipment, standardized communication lines and wiring systems should be adopted, and redundancy and fault tolerance should be fully considered to ensure the reliability of the network to the greatest extent. If conditions permit, it is best to use two-wire connection mode, so that network lines can be switched in case of failure.
1.3 network management. In some large-scale network equipment structures, the network products and equipment used are produced by different manufacturers, which are relatively large in scale and complicated in structure. Improving the reliability of computer network can ensure the integrity of information transmission, reduce the incidence of information loss, and reduce the incidence of failures and errors. In order to improve the reliability of computer network, it is necessary to adopt advanced network management technology to collect network operation parameters in real time and eliminate existing faults.
1.4 network topology. Network topology refers to the use of transmission media to connect various devices, mainly reflected in the physical interconnection between network devices. Computer network topology is related to the planning structure of the whole network and is one of the important determinants of computer network reliability. The performance of network topology is mainly influenced by network technology, network scale, user distribution and transmission medium. With the improvement of people's requirements for network performance, computer network topology now needs to meet more requirements, such as fault-tolerant diameter, wide diameter, limited connectivity, limited fault-tolerant diameter and so on. These parameters can measure the reliability and fault tolerance of computer network more accurately, so as to realize the scientificity and reliability of computer network planning.
2 Network reliability analysis based on intelligent computing
2. 1 The concept of network reliability based on intelligent computing. The components of computer network system include nodes and arcs connecting nodes, and nodes can be divided into input nodes (only output arcs but no nodes belong to input arcs), output nodes (only input arcs without output arcs) and intermediate nodes (non-input/output nodes). Networks can be divided into directed networks (all composed of directed arcs), undirected networks (all composed of undirected arcs) and mixed networks (including directed arcs and undirected arcs). In some complex network systems, in order to accurately analyze the reliability of the system, the network diagram is generally used to represent it. When analyzing network reliability, we usually simplify it as follows: there are only two States of a system or arc: normal and fault; The undirected arc has the same reliability in different directions; Failure of any arc will not affect the normal use of other arcs.
2.2 The node traversal method of the minimum path set of the network system. There are generally three methods to find the minimum path set of a network system: First, adjacency matrix is also called contact matrix method, and its principle is to multiply a matrix and multiply it several times. This method is suitable for manual operation in the network with few nodes, but it is not suitable when the number of nodes is large, because it will require a lot of calculation, high computer capacity and long operation time, and it is not suitable for this method; Secondly, Boolean determinant method, similar to finding matrix determinant, is easy to understand and operate, and can be handled manually, but it is cumbersome to apply in networks with more nodes. Thirdly, node traversal method is widely used because of its clear organization and ability to solve complex networks with multiple nodes. However, this method has many judgment conditions, and it is easy to make mistakes due to insufficient consideration. The basic method of finding the minimum path set of network system is to traverse point by point from input node I to output point L until all the minimum path sets are found. In this process, the following judgments need to be made: judge whether the current node is duplicate with the previous node; Jud whether that minimum path set is found; It is judged whether all the minimum path sets have been found.
2.3 Minimum path set operation of computer network system based on intelligent granular computing partition. Granular computing is a cluster of points in the universe, which are often difficult to distinguish, approach or combine with a function. Granular computing is a big umbrella, covering many problem-solving methods in specific fields, which are embodied in interval analysis, divide-and-conquer method and rough set theory. The operation method of the minimum path set of the computer network system based on the improved node traversal method of intelligent granular computing is generally as follows: First, the two-dimensional array in the traditional node traversal calculation method of the network system is expressed in one dimension, which is easy to be expressed as n- 1, because the maximum path length of the minimum path set of the network system with n nodes is less than or equal to n- 1, that is to say, the one-dimensional dynamic array is enabled, and it is traversed one by one from the input node to the output node. When finding the minimum path set, you can write the result into the hard disk file, then continue to find the next minimum path set, and then write it into the hard disk file, and so on until all the minimum path sets are found and the one-dimensional dynamic array is released; Secondly, the array integrated into the operation is dynamically involved in the operation, and the operation function is released immediately after completion, which can save memory space and improve the overall operation speed; Then, according to the minimum path set file represented by nodes, it is transformed into the minimum path set represented by arcs and stored for subsequent correlation calculation. Finally, the object is divided by the theory of intelligent granular computing, which is realized by dynamic array layering, so as to realize the reliability analysis of computer network system.
3 the realization of computer network reliability
3. 1 computer network hierarchy and architecture design. A reliable computer network should not only be equipped with advanced network equipment, but also have advanced network hierarchy and architecture. Scientific and reasonable network hierarchy and architecture design can give full play to the performance of network equipment. Network hierarchical design is to build a large-scale high-speed network hierarchical design model for distributed network services with the increase of network throughput. The modular and hierarchical design of the network can increase the network capacity with the increase of network nodes in the future, thus increasing the certainty and facilitating the daily operation.
3.2 the realization of computer network fault tolerance. What are the guiding principles of fault-tolerant design? Parallel backbone, dual network center? The specific design is as follows: firstly, connecting the user terminal equipment and the server to the computer network center at the same time is generally realized by the method of parallel computer network and redundant computer network center; Second, connect the data links and routers in the WAN to ensure that any data link failure will not affect local network users; Third, try to use network equipment with hot plug function, which can not only make the networking mode flexible, but also replace the faulty module in time without cutting off the power supply, thus improving the ability of computer network to work for a long time; Finally, multi-processors and specially designed systems with fault tolerance function are used to run network management software to realize fault tolerance.
3.3 Redundancy measures are adopted. Improving the fault-tolerant ability of computer network system is the most effective way to improve the reliability of computer network. The fault-tolerant design of computer network is to find common faults, which can be strengthened by redundant measures to minimize the duration of faults and avoid data loss, errors or even paralysis in computer network, such as redundant data links between users and computer network centers.
4 conclusion
Studying the reliability of computer network system is of great significance to solving problems, so it is necessary to study its reliability, but it needs to be further discussed from the theoretical method. With the application of computer network system in every corner, its reliability analysis has attracted more and more attention from the industry. The method of network reliability analysis should be based on the principle of theory serving practice, and the reliability analysis theory should be applied to actual production to make the computer network construction more scientific and reasonable.
References:
[1] Liu Jun. Research on reliability optimization design of computer network [J]. China Science and Technology Information, 2011(18): 29.
[2] Deng Zhiping. On the Optimal Design of Computer Network Reliability [J]. Science and Technology Plaza, 20 10 (10): 52.
[3] Gao Fei. Network reliability analysis based on network state relation [J]. Communication Network, 20 12 (25): 19.