Feature Engineering versus Feature Extraction: Game On!
Feature Engineering versus Feature Extraction: Game On!
"Feature engineering" is a fancy term for making sure that your predictors are encoded in the model in a manner that makes it as easy as possible for the model to achieve good performance. For example, if your have a date field as a predictor and there are larger differences in response for the weekends versus the weekdays, then encoding the date in this way makes it easier to achieve good results.
However, this depends on a lot of things.
First, it is model-dependent. For example, trees might have trouble with a classification data set if the class boundary is a diagonal line since their class boundaries are made using orthogonal slices of the data (oblique trees excepted).
Second, the process of predictor encoding benefits the most from subject-specific knowledge of the problem. In my example above, you need to know the patterns of your data to improve the format of the predictor. Feature engineering is very different in image processing, information retrieval, RNA expressions profiling, etc. You need to know something about the problem and your particular data set to do it well.
Here is some training set data where two predictors are used to model a two-class system (I'll unblind the data at the end):
There is also a corresponding test set that we will use below.
There are some observations that we can make:
- The data are highly correlated (correlation = 0.85)
- Each predictor appears to be fairly right-skewed
- They appear to be informative in the sense that you might be able to draw a diagonal line to differentiate the classes
Depending on what model that we might choose to use, the between-predictor correlation might bother us. Also, we should look to see of the individual predictors are important. To measure this, we'll use the area under the ROC curve on the predictor data directly.
Here are univariate box-plots of each predictor (on the log scale):
There is some mild differentiation between the classes but a significant amount of overlap in the boxes. The area under the ROC curves for predictor A and B are 0.61 and 0.59, respectively. Not so fantastic.
What can we do? Principal component analysis (PCA) is a pre-processing method that does a rotation of the predictor data in a manner that creates new synthetic predictors (i.e. the principal components or PC's). This is conducted in a way where the first component accounts for the majority of the (linear) variation or information in the predictor data. The second component does the same for any information in the data that remains after extracting the first component and so on. For these data, there are two possible components (since there are only two predictors). Using PCA in this manner is typically called feature extraction.
Let's compute the components:
> library(caret)
> head(example_train)
PredictorA PredictorB Class
2 3278.726 154.89876 One
3 1727.410 84.56460 Two
4 1194.932 101.09107 One
12 1027.222 68.71062 Two
15 1035.608 73.40559 One
16 1433.918 79.47569 One
> pca_pp <- preProcess(example_train[, 1:2],
+ method = c("center", "scale", "pca"))
+ pca_pp
Call:
preProcess.default(x = example_train[, 1:2], method = c("center",
"scale", "pca"))
Created from 1009 samples and 2 variables
Pre-processing: centered, scaled, principal component signal extraction
PCA needed 2 components to capture 95 percent of the variance
> train_pc <- predict(pca_pp, example_train[, 1:2])
> test_pc <- predict(pca_pp, example_test[, 1:2])
> head(test_pc, 4)
PC1 PC2
1 0.8420447 0.07284802
5 0.2189168 0.04568417
6 1.2074404 -0.21040558
7 1.1794578 -0.20980371
Note that we computed all the necessary information from the training set and apply these calculations to the test set. What do the test set data look like?
These are the test set predictors simply rotated.
PCA is unsupervised, meaning that the outcome classes are not considered when the calculations are done. Here, the area under the ROC curves for the first component is 0.5 and 0.81 for the second component. These results jive with the plot above; the first component has an random mixture of the classes while the second seems to separate the classes well. Box plots of the two components reflect the same thing:
There is much more separation in the second component.
This is interesting. First, despite PCA being unsupervised, it managed to find a new predictor that differentiates the classes. Secondly, it is the last component that is most important to the classes but the least important to the predictors. It is often said that PCA doesn't guarantee that any of the components will be predictive and this is true. Here, we get lucky and it does produce something good.
However, imagine that there are hundreds of predictors. We may only need to use the first X components to capture the majority of the information in the predictors and, in doing so, discard the later components. In this example, the first component accounts for 92.4% of the variation in the predictors; a similar strategy would probably discard the most effective predictor.
How does the idea of feature engineering come into play here? Given these two predictors and seeing the first scatterplot shown above, one of the first things that occurs to me is "there are two correlated, positive, skewed predictors that appear to act in tandem to differentiate the classes". The second thing that occurs to be is "take the ratio". What does that data look like?
The corresponding area under the ROC curve is 0.8, which is nearly as good as the second component. A simple transformation based on visually exploring the data can do just as good of a job as an unbiased empirical algorithm.
These data are from the cell segmentation experiment of Hill et al,and predictor A is the "surface of a sphere created from by rotating the equivalent circle about its diameter" (labeled as EqSphereAreaCh1 in the data) and predictor B is the perimeter of the cell nucleus (PerimCh1). A specialist in high content screening might naturally take the ratio of these two features of cells because it makes good scientific sense (I am not that person). In the context of the problem, their intuition should drive the feature engineering process.
However, in defense of an algorithm such as PCA, the machine has some benefit. In total, there are almost sixty predictors in these data whose features are just as arcane as EqSphereAreaCh1. My personal favorite is the "Haralick texture measurement of the spatial arrangement of pixels based on the co-occurrence matrix". Look that one up some time. The point is that there are often too many features to engineer and they might be completely unintuitive from the start.
Another plus for feature extraction is related to correlation. The predictors in this particular data set tend to have high between-predictor correlations and for good reasons. For example, there are many different ways to quantify the eccentricity of a cell (i.e. how elongated it is). Also, the size of a cell's nucleus is probably correlated with the size of the overall cell and so on. PCA can mitigate the effect of these correlations in one fell swoop. An approach of manually taking ratios of many predictors seems less likely to be effective and would take more time.
Last year, in one of the R&D groups that I support, there was a bit of a war being waged between the scientists who focused on biased analysis (i.e. we model what we know) versus the unbiased crowd (i.e. just let the machine figure it out). I fit somewhere in-between and believe that there is a feedback loop between the two. The machine can flag potentially new and interesting features that, once explored, become part of the standard book of "known stuff".
Feature Engineering versus Feature Extraction: Game On!的更多相关文章
- Feature Engineering and Feature Selection
首先,弄清楚三个相似但是不同的任务: feature extraction and feature engineering: 将原始数据转换为特征,以适合建模. feature transformat ...
- Discover Feature Engineering, How to Engineer Features and How to Get Good at It
Feature engineering is an informal topic, but one that is absolutely known and agreed to be key to s ...
- 学习笔记(四): Representation:Feature Engineering/Qualities of Good Features/Cleaning Data/Feature Sets
目录 Representation Feature Engineering Mapping Raw Data to Features Mapping numeric values Mapping ca ...
- 特征工程(Feature Engineering)
一.什么是特征工程? "Feature engineering is the process of transforming raw data into features that bett ...
- 【计算机视觉领域】常用的 feature 提取方法,feature 提取工具包
[计算机视觉领域]常用的 feature 提取方法,feature 提取工具包 利用 VL 工具包进行各种特征的提取: VL 工具包官网地址:http://www.vlfeat.org/index.h ...
- 机器学习-特征工程-Feature generation 和 Feature selection
概述:上节咱们说了特征工程是机器学习的一个核心内容.然后咱们已经学习了特征工程中的基础内容,分别是missing value handling和categorical data encoding的一些 ...
- feature.shape和feature.shapecopy的区别
以前在写AE代码的时候也没有注意到feature.shape和feature.shapecopy的区别,觉得两者也差不多: 今天写入库程序才明白过来. 如果取feature.shape,则得到的是该要 ...
- Google Play和基于Feature的过滤 —— Feature 参考手册
翻译自 Features Reference 下表列出了软/硬件Feature和权限的参考信息,它们被用于GooglePlay. 硬件feature 下面列出了被大多数当前发布的平台所支持的硬件功能描 ...
- tensorflow-TFRecord报错ValueError: Protocol message Feature has no "feature" field.
编写代码用TFRecord数据结构存储数据集信息是报错:ValueError: Protocol message Feature has no "feature" field.或和 ...
随机推荐
- Thinkphp 下面执行crond
thinkphp开启cli支持 1.tp正好支持cli命令模式,手册的路径为13.7.4 如果是用的其他框架不支持cli,那么只能直接写程序了,其实就是写面向过程的最基础的php代码. 2.在入口文 ...
- CentOS7.0安装JDK1.8.0_31
1.检查一下系统中的jdk版本 $>java -version java version "1.7.0_"OpenJDK Runtime Environment (IcedT ...
- Index of my articles
65:一个表格说明RelativeLayout中的几个重要属性[Written By KillerLegend] (2013-10-16 21:59) 64:win7修改软件[授权给…]后面的名称 ( ...
- delphi 单引号在字符串中使用方法
可以看delph的帮助,里面有这个问题详细说明:A character string, also called a string literal or string constant, consist ...
- 2.css选择器
由第一节的语法结构可以看出css语言的核心思想就是:找到对象,然后对选定的对象进行属性赋值.其中,css中对象的选择,依靠的就是选择器.当掌握了选择器以后,就能够指哪打哪,弹无虚发了. css的选择器 ...
- python使用简单http协议来传送文件
python使用简单http协议来传送文件!在ubuntu环境下,局域网内可以使用nc来传送文件,也可以使用基于Http协议的方式来下载文件我们可以使用python -m SimpleHTTPServ ...
- Ztack学习笔记(6)-广播组播点播
Zigbee网络中进行数据通信主要有三种类型:单播.组播.广播.那这三种方式如何设置呢,在哪里设置呢, 一. 广播 当应用程序需要将数据包发送给网络的每一个设备时,使用这种模式.广播的短地址有三种 0 ...
- rspec的一些常见用法
这里讲了如何安装rspec,安装使用rspec. 下面介绍一下rspec中常见的使用方法. 下面是一个最简单的测试用例,判断true是不是等于true,should_be是旧的用法,新用法推荐使用ex ...
- uninstall 11.2.0.3.0 grid & database in linux 5.7
OS: Oracle Linux Server release 5.7 DB: Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - ...
- .net 的生成操作
生成操作(BuildAction) 属性:BuildAction 属性指示 Visual Studio .NET 在执行生成时对文件执行的操作. BuildAction 可以具有以下几个值之一: 无( ...