The way of information processing

From different angles, there are various divisions of information processing methods. According to the depth of processing function, information processing can be divided into three types: pretreatment processing, business processing and decision processing. The first is the simple arrangement of information, which processes the previous information. The second is to analyze information and make comprehensive information to assist decision-making. The third category is statistical inference of information, which can produce decision-making information.

After processing, data becomes pre-information or statistical information, and statistical information becomes useful information for decision-making after processing. This transformation takes time and inevitably leads to time delay, which is also an important feature of information processing-lag, which must be paid attention to in use. The lag of information and the timeliness of information are contradictory. Information workers should understand the lag of information and try to reduce or even eliminate the restriction and influence of lag on timeliness. According to the different response time of processing, information processing methods can be divided into two types:

(1) One is "real-time processing type", which immediately processes the sent data and responds immediately. Generally, real-time processing systems are only allowed to process confirmed work, which is limited to routine operations to ensure timely response.

(2) The other is "batch processing", that is, the sent data is stored for a certain amount or time, and then processed centrally. This processing method is suitable for the following two statistical analysis services: if a certain amount of necessary data is not collected, there will be no processing significance and effect; There is no need to rush to get results.

From the perspective of development, information processing is developing from batch processing to on-line processing, focusing on post-processing to real-time processing. Traditional information processing is mainly carried out through the human brain, and then artificial equipment and computers appeared one after another. In other words, there are generally two ways to process information: manual processing and computer processing. Using manual management to process information is not only cumbersome and error-prone, but also takes a long time to process, which is far from meeting the needs of management decision-making. The continuous development and application of computer, artificial intelligence and other technologies greatly shorten the information processing time, meet the decision-making needs of managers, and at the same time make people get rid of the cumbersome manual management methods.

Computer information processing is the use of computers to process data. In the process of processing, a large number of mathematical models are used. The algorithms of these models are usually quite complex and usually contain a lot of iterations and loops. But there are many software packages to choose from, such as statistical software package, prediction software package, mathematical programming software package, simulation software package and so on. In the past, in management work, most of them relied on the experience of managers to process information, and the few operations needed were limited to simple arithmetic operations and simple statistical processing. With the development of management modernization, many methods in mathematical statistics and operational research have entered the field of economic management. In particular, the relationship between modern statistical methods and information processing is increasingly close. As a basic tool of information processing, modern statistical methods will play an increasingly important role. At present, many large computer data processing systems generally have three libraries, namely, database, model library and method library. Generally speaking, the model base is the core, the database provides it with necessary information, and the method base provides it with corresponding methods. There are many standard algorithms in the method base, but the models for different problems are all stored in the model base, and the data to be used are all in the database. Among them, the model and method are both related and different. The model is put forward on the basis of management theory and practice, which reflects the quantitative or qualitative relationship between various factors in management work and describes the specific laws in this field. Method refers to some mathematical calculation method (for example, regression analysis, trend extrapolation, game method, etc.). ), independent of management. Therefore, models can use various methods, and the same method can also be used by multiple models. From the application point of view, the model is the substantive content and the method is the tool.