当前位置:网站首页>Data analysis - Thinking foreshadowing
Data analysis - Thinking foreshadowing
2022-07-05 23:13:00 【Dutkig】
Three parts of data analysis
Data collection
data mining
—— The core of data mining is to mine the commercial value of data , That is what we are talking about business intelligence (BI)
You need to master and understand the following contents :
① The basic flow
② Ten algorithms
③ A certain mathematical foundationData visualization
This part is mainly to learn how to use relevant tools
Two principles
- Try to use third-party class libraries to complete your own ideas
- Try to choose the tool with the most users ,bug Less , All the documents , Many cases
The basic flow
- Business understanding : Understand project requirements from a business perspective , Better serve the business ;
- Data understanding : Explore the data , Including data description , Data quality verification , So as to have a preliminary understanding of the data ;
- Data preparation : Data cleaning and inheritance ;
- model : Apply the mining model and optimize , In order to get better classification results ;
- Model to evaluate : Evaluate the model , Check every step of building the model , Confirm whether the model has achieved the business objectives ;
- Launch online
Ten algorithms of data mining
For different purposes , The above ten algorithms are divided into the following four categories :
- Classification algorithm :C4.5 , Naive Bayes ,SVM,KNN,Adaboost,CART;
- clustering algorithm :K—Means,EM
- Correlation analysis :Apriori
- Connection analysis :PageRank
First of all, let's have a preliminary understanding of the above 10 Algorithms :
C4.5
A decision tree algorithm , Prune in the process of creating the decision tree , And can handle continuous attributes , It can also process incomplete data .
Naive Bayes
Based on the principle of probability theory , Want to classify the given unknown objects , We need to solve the probability of each category under the condition of occurrence , Which is the biggest , Which classification do you think it belongs to .
SVM
Support vector machine (Support Vector Machine) Build a hyperplane classification model .
KNN
K Nearest neighbor algorithm (K-Nearest Neighbor) Each sample can use its latest k A neighbor represents , If a sample , its k The closest neighbors belong to the classification A, So this sample also belongs to classification A
AdaBoost
AdaBoost A joint classification model is established in the training , Build a classifier Lifting Algorithm , It allows us to form a strong classifier with multiple weak classifiers , therefore Adaboost It is also a commonly used classification algorithm .
CART
CART Represents classification and regression trees , English is Classification and Regression Trees. Like English , It builds two trees : One is a classification tree , The other is the regression tree . and C4.5 equally , It is a decision tree learning method .
Apriori
Apriori Is a kind of mining association rules (association rules) The algorithm of , It does this by mining frequent itemsets (frequent item sets) To reveal the relationship between objects , It is widely used in the fields of business mining and network security . Frequent itemsets are collections of items that often appear together , Association rules imply that there may be a strong relationship between the two objects .
K-Means
K-Means Algorithm is a clustering algorithm . You can think of it this way , Finally, I want to divide the object into K class . Suppose that in each category , There was a “ Center point ”, Opinion leader , It is the core of this category . Now I have a new point to classify , In this case, just calculate the new point and K The distance between the center points , Which center point is it near , It becomes a category .
EM
EM Algorithm is also called maximum expectation algorithm , It is a method to find the maximum likelihood estimation of parameters . The principle is : Suppose we want to evaluate parameters A And parameters B, In the initial state, both are unknown , And got it A You can get B Information about , In turn, I know B And you get A. Consider giving... First A Some initial value , So as to get B Valuation of , And then from B Starting from the valuation of , Reevaluate A The value of , This process continues until convergence .
PageRank
PageRank It originated from the calculation of the influence of the paper , If a literary theory is introduced more times , It means that the stronger the influence of this paper . Again PageRank By Google It is creatively applied to the calculation of web page weight : When a page chains out more pages , Description of this page “ reference ” The more , The more frequently this page is linked , The higher the number of times this page is referenced . Based on this principle , We can get the weight of the website .
边栏推荐
- audiopolicy
- Basic knowledge of database (interview)
- CJ mccullem autograph: to dear Portland
- Global and Chinese markets for children's amusement facilities 2022-2028: Research Report on technology, participants, trends, market size and share
- leecode-学习笔记
- Leetcode buys and sells stocks
- Yiwen gets rid of the garbage collector
- grafana工具界面显示报错influxDB Error
- Debian 10 installation configuration
- 一文搞定class的微觀結構和指令
猜你喜欢
[speech processing] speech signal denoising based on Matlab GUI Hanning window fir notch filter [including Matlab source code 1711]
Data type, variable declaration, global variable and i/o mapping of PLC programming basis (CoDeSys)
CJ mccullem autograph: to dear Portland
Fix the memory structure of JVM in one article
Use of grpc interceptor
Go language implementation principle -- lock implementation principle
第十七周作业
Hainan Nuanshen tea recruits warmhearted people: recruitment of the product experience recommender of Nuanshen multi bubble honey orchid single cluster
Expectation, variance and covariance
【Note17】PECI(Platform Environment Control Interface)
随机推荐
Basic knowledge of database (interview)
Krypton Factor purple book chapter 7 violent solution
regular expression
一文搞定JVM常见工具和优化策略
3: Chapter 1: understanding JVM specification 2: JVM specification, introduction;
LeetCode102. Sequence traversal of binary tree (output by layer and unified output)
[digital signal denoising] improved wavelet modulus maxima digital signal denoising based on MATLAB [including Matlab source code 1710]
Detailed explanation of pointer and array written test of C language
SPSS analysis of employment problems of college graduates
Multi sensor fusion of imu/ optical mouse / wheel encoder (nonlinear Kalman filter)
Development specification: interface unified return value format [resend]
Roman numeral to integer
C Primer Plus Chapter 9 question 9 POW function
audiopolicy
从 1.5 开始搭建一个微服务框架——日志追踪 traceId
Nacos installation and service registration
Masked Autoencoders Are Scalable Vision Learners (MAE)
February 13, 2022-4-symmetric binary tree
【原创】程序员团队管理的核心是什么?
Openresty ngx Lua regular expression