数据挖掘 第二章 认识数据研讨.pptVIP

* * * * * * * * * * * * * * * * * * * Note to self (MK): last part of slide does not clearly match slides that follow Introduction slides: In the current version, we only talk about numeric data. Can we add some materials about non-numeric data. - text statistics, TF/IDF - visualization of text statistics such as word cloud - distribution of Internet (IN, SCC, OUT, …) - visualization of social relationship (e.g., /) * * * * * * Redraw the boxplot! Yes, we need a more real boxplot graph! * * JH: Maybe we can remove Loess curve * * Note: We need to label the dark plotted points as Q1, Median, Q3 – that would help in understanding this graph. Tell audience: There is a shift in distribution of branch 1 WRT branch 2 in that the unit prices of items sold at branch 1 tend to be lower than those at branch 2. * Need a better and more MEANINGFUL scatter plot! -JH * * CC (conference call) 08.11.02: Need to check Wiki to update references. * CC (conference call) 08.11.02: Need to check Wiki to update references. * CC (conference call) 08.11.02: Need to check Wiki to update references. * CC 08/11/02: Need a good survey. JH or JP may find a good book on amazon in order to update this slide. JH’s colleagues may generate some images for book. We may consider the following books. Visualizing Data: Exploring and Explaining Data with the Processing Environment - Paperback - Illustrated (Jan 11, 2008) by Ben Fry. JH: I have ordered the book. * /companions/1558606890/pictures/Chapter_01/fig1-6b.gif * * * * * * “two attributes mapped to axes, remaining attributes mapped to angle or length of limbs”. Look at texture pattern. * MK 08/10/18: Is it “Treemap” or “Tree-Map”? - inconsistent in this file. * * Picture: /hcil/treemap-history/ * * 切尔诺夫图 用二维的脸表示18维的多维数据（赫尔曼·切尔诺夫） 切尔诺夫脸利用脸的眼耳口鼻等要素的不同形状，大小，位置和方向代表维的值。利用人的思维能力，识别面部特征的微小差异来理解许多面部特征，有助于数据的规律性和不规律性的可视化。 Data Mining: Concepts and Techniques * X和Y轴映射两个维 used by permission of G. Grinstein