《数据挖掘》模拟卷(国外英语资料).doc

《数据挖掘》模拟卷 Data mining simulation volume First, fill in the blanks (1 points per grid, 20 points) 1, in data mining, the commonly used clustering algorithms include: division method Hierarchical approach Density based approach Grid based approach and model-based approach. 2, the data warehouse multidimensional data model can be divided into three different forms, namely, star model, snowflake model and fact constellation 3, from the point of view of data analysis, data mining can be divided into two categories: descriptive data mining and predictive data mining 4, given the basic square, there are three options for the materialization of the square: no materialization , Total materialization and Partial materialization 5, the three main research directions in current data mining research are: Database technology, Statistics、 and machine learning 6, there are four types of concept hierarchy, namely: , , and 7, two commonly used large data sets of data generalization method is and Two, radio questions (please choose a correct answer, fill in brackets, 2 points per subject, a total of 20 points) 1. which of the following classifications belongs to the neural network learning algorithm? ( ) A. decision tree induction B. Bayes classification C. backward propagation classification D. case-based reasoning The 2. confidence measure (confidence) is an index of interestingness measures. A, simplicity B, deterministic C., practicality D, novelty 3. which of the following situations does outlier mining apply? A and target market analysis B, shopping basket analysis C and pattern recognition D, credit card fraud detection 4. the cube that holds the lowest level is called: A, vertex cube Lattices of B and squares C, basic cube D and dimension 5., the purpose of data reduction is () A to fill the vacancy values of data types B, integrating data from multiple data sources C, get the compressed representation of the data set D, normalized data 6. which of the following data preproces