As an interdisciplinary subject. Data mining integrates databases, artificial intelligence statistics and other fields, and abstracts databases, human intelligence and mathematical statistics are the main pillars of data mining technology. The main reason for data mining is intelligence. Association rules, decision-making, clustering and feint-based learning. 1J! Yeasian learning, rough set, mustache network, genetic algorithm, statistical analysis and other technologies. Adopt f1j data sampling {select data samples), data exploration, taxable data exploration and cluster analysis and selection 1, data adjustment (data group subdivision and splitting), modeled [human] neural network ．Decision-making model, mathematical statistics analysis and time sequence analysis, and evaluation (conclusion synthesis and evaluation, whether to repair the ship, and whether new problems arise) and other five basic processes may need to be repeated. Obtain the water quality of things and constantly solve the problem. Through correlation analysis, classification analysis, prediction and deviation detection, the relationship between the data and the pattern of the data are currently the most common. FHn: J Data mining technologies include: modular logic and rough set methods, genetic algorithms, proximity search algorithms, etc. Functionally speaking, the analysis methods of data mining are divided into four types: correlation analysis, sequence analysis, partition analysis and cluster analysis. Association rules: association rules that represent data relationships are used in direct commercial applications_The most typical example is that a chain store discovered through data mining the intrinsic relationship between I flag diapers and N beer.
Expand
