1. The main reasons why data mining has aroused the demand of experts, scholars and commercial manufacturers: the widespread use of large-scale data systems and the urgent need to convert data into useful knowledge.

2. Data mining technology changes with the development of the times.

3. Data, information, and knowledge can be regarded as different forms of generalized data expression.

1. Technical basis: the development of computers and related technologies.

2. Technical background arising from data mining: the development of information technologies such as databases, data warehouses and the Internet, the improvement of computer performance and the development of advanced architectures, and the research and application of statistics, artificial intelligence and other methods in data analysis.

1. The development of big data research can be roughly divided into three stages: 2000 and before, called the "budding stage of the big data concept", 2001-2010, called the "big data concept exploration stage", 2011 and later , the concept of big data has been further deepened, has become the focus of academic research, and has become a concept supported by many applications.

2. The four perspectives of the current big data concept: data theory, methodology, environment theory, and attribute theory.

Based on attribute theory, it shows that data mining technology is the core of big data analysis. New challenging technical methods that need to be strengthened include: large-volume data analysis requires data mining technology to support, and high-speed aggregation of big data proposes new methods for data mining. Challenging topics and various types of big data need to be supported by the development of relevant research branches of data mining. Big data with huge value is the research target of data mining technology.

Data mining is essentially a new business information processing technology

1. “Knowledge Discovery” (KDD) in Databases

2. Data Mining and Knowledge Discovery (DMKD)

3. The meaning of data mining technology: KDD is regarded as a special case of data mining. Data mining is a step in the KDD process. KDD has the same meaning as Data Mining.

Some important theoretical frameworks: pattern discovery framework, rule discovery framework, based on probability and statistical theory, microeconomic perspective, based on data compression theory, based on inductive database theory, visual data mining.

1. Definition: Generalized knowledge mining refers to generalized knowledge that describes category characteristics.

2. Concept description method: essentially summarizing the connotative characteristics of certain objects. Concept description is divided into characteristic description and distinctive description.

3. Multidimensional data analysis can be regarded as an effective method for generalized knowledge mining.

4. Multi-level concept description problems (currently using more concept layering methods): collection grouping layering, rule-based layering, operation derived layering, and pattern layering.

1. Definition: Relevant knowledge reflects the dependence or association between an event and other events.

2. Purpose: Find hidden information in the database.

3. Type: simple association, temporal association, causal association, quantitative association.

4. Broadly speaking, correlation analysis is the essence of data mining.

5. Association rule mining is the most commonly used method for association knowledge discovery.

1. Definition: Class knowledge describes a class of things that have common characteristics in a certain sense and are clearly distinguished from things of different classes.

2. Two basic methods for mining class knowledge

Definition: Predictive knowledge mining refers to knowledge generated from historical and current data and able to predict future data trends.

Modes: trend prediction mode, cycle analysis mode, sequence mode, neural network.

The following questions can help you understand the tasks and methods of specific knowledge mining: isolated point analysis, sequence anomaly analysis, and discovery of specific rules.

Some problems currently being studied: multi-dimensional knowledge mining problems, multi-table mining and quantity mining problems, multi-layer knowledge mining problems, knowledge evaluation problems, and constrained data mining problems.

Data types of data mining: structured data mining, unstructured data mining, semi-structured data mining.

Wed mining must face some key issues: heterogeneous data source environment, semi-structured data structure, and dynamically changing language environment.

There are three schools of research on Wed mining: Wed structure mining, Wed usage mining, and Wed content mining.

Rough set theory is a teaching tool for studying imprecise and uncertain knowledge. Rough sets abstract the objective world into an information system.

Reduction: Minimal set of attributes.

Application of data mining in sports competitions

Data mining applied to commercial banks: financial investment, fraud screening.

Data mining applied to telecommunications

Data mining applied to scientific exploration

Application of data mining in information security