Data mining is an extraordinary process to discover hidden, regular and unknown information and knowledge from a large number of incomplete, noisy, fuzzy and random practical application data, but it is potentially useful and ultimately understandable. Any place that needs data management and knowledge discovery can use data mining technology to solve problems. This paper studies the application of data mining algorithm and data mining technology, and studies the application of data mining technology.

Keywords: data mining; Technology; Application;

Introduction: Data mining technology is the result of long-term research and development of database technology. At first, all kinds of business data were stored in the database of computer, and then developed to query and access the database, and then developed to traverse the database in real time. Data mining makes database technology enter a more advanced stage. It can not only query and traverse past data, but also find out the potential relationship between past data, thus promoting the transmission of information.

First, an overview of data mining

Data mining is an extraordinary process to discover hidden, regular and unknown information and knowledge from a large number of incomplete, noisy, fuzzy and random practical application data, but it is potentially useful and ultimately understandable.

Second, the basic process of data mining

(1) Data Selection: Select the data related to the data mining target. Processing data according to different data mining targets can not only eliminate unnecessary data interference, but also greatly improve the efficiency of data mining. (2) Data preprocessing: mainly including data cleaning, data integration and transformation, data reduction, discretization and concept hierarchy generation. (3) Pattern discovery: The process of discovering patterns that users are interested in from data is the main processing process of knowledge discovery. (4) Pattern evaluation: get a pattern that truly represents knowledge through certain measurement. Generally speaking, enterprise data mining mainly follows the following processes: data preparation, that is, data collection and accumulation. At this time, enterprises need to know what kind of data they need, and get objective and clear target data through classification, editing, cleaning and pretreatment. Data Mining This is the most critical step. The main purpose is to further mine the preprocessed data to obtain more objective and accurate data before introducing the decision. Different enterprises may adopt different data mining technologies, but at present, they cannot get rid of the above mining methods for the time being. Of course, with the advancement of technology, big data will definitely become the foundation of enterprises and has been applied in many fields. For example, marketing, which is the earliest application field of data mining, aims at mining users' consumption habits, analyzing users' consumption characteristics, and then conducting precise marketing. Take the hateful pop-up advertisement as an example. When consumers have the habit of online shopping, they search for their favorite products online, and then search, many products aiming at consumers' consumption habits will pop up.

Thirdly, data mining methods.

1. Aggregation discovery.

Clustering is to divide the whole database into different groups. Its purpose is to make the differences between groups obvious and the data of the same group as similar as possible. The typical application of clustering in e-commerce is to help market analysts find different customer groups from customer groups and describe the characteristics of different customer groups with purchase patterns. In addition, cluster analysis can be used as a preprocessing step for other algorithms, such as features and classification, and then these algorithms are processed on the generated clusters. Unlike classification, you don't know how to divide the data into several groups before you start clustering, and you don't know how to divide it (according to which variables). Therefore, after grouping, people who are familiar with the business should explain the significance of this grouping. In many cases, the cluster you get at one time may not be good for your business. At this time, you need to delete or add variables to affect the clustering method. After several iterations, we can finally get an ideal result. There are two main clustering methods, including statistical method and neural network method. Self-organizing neural network method and K- means are commonly used clustering algorithms.

2. Decision tree.

This is very powerful for solving the problems of classification and prediction. Through a series of questions, rules are formed and expressed, and then through constant questions, the required results are obtained. A typical decision tree has a root at the top and many leaves at the bottom. Records are broken down into different subsets, and each subset may contain a simple rule.

Fourth, the application fields of data mining

4. 1 marketing

The application of marketing data mining in sales industry can be divided into two categories: database sales and shopping basket data analysis. The task of the former is to select potential customers through interactive inquiry, data segmentation and model prediction, so as to sell products to them, instead of blindly selecting customers to sell products as before; The latter's task is to analyze market sales data to identify customers' buying behavior patterns, so as to help determine the layout and launch of store shelves and promote a certain commodity.

4.2 Financial investment

Typical financial analysis fields include investment evaluation and stock market prediction, and the analysis method generally adopts model prediction method. There are also Fidelity stock selection and LBS capital management. The former's task is to use neural network model to select investment, while the latter uses expert system, neural network and genetic algorithm technology to assist in the management of securities up to 600 million US dollars.

Conclusion: Data mining is a new intelligent information processing technology. With the rapid development of related information technology, the application fields of data mining are constantly broadened and deepened, especially in the fields of telecommunications, military, bioengineering and business intelligence, which will become a new research hotspot. At the same time, the application of data mining also faces many technical challenges. How to mine complex types of data, the integration of data mining with database, data warehouse and Web technology, the visualization and data quality of data mining need further research and exploration.

refer to

[1] Meng Qiang, Li Haichen. Web data mining technology and its application research [J]. Computer and Information Technology, 2017,25 (1): 59-62.

[2] Gao Haifeng. Research on the Application of Data Mining Technology in Intelligent Transportation System [J]. Digital Technology and Application, 2016 (5):108-108.

;