当前位置：网站首页>Graphical data analysis | business analysis and data mining

Graphical data analysis | business analysis and data mining

2022-06-12 02:01:00 【ShowMeAI】

ShowMeAI research center

author ： Han Xinzi @ShowMeAI
Tutorial address ：http://www.showmeai.tech/tutorials/33
This paper addresses ：http://www.showmeai.tech/article-detail/139
Statement ： copyright , For reprint, please contact the platform and the author and indicate the source

The core steps of data analysis are divided into ： Business cognition and data exploration 、 Data preprocessing 、 Business cognition and data exploration Wait for three core steps . This article introduces the third step —— Business cognition and data exploration .

One 、 Business analysis model

1.1 AB test

AB test , Simply speaking , It is to make two plans for the same product goal （ For example, two pages have a red button 、 The other one uses a blue button ）, Let some users use A programme , Another part of users use B programme , Then log the user's usage , And analyze relevant indicators through structured log data , Such as click through rate 、 Conversion rate, etc , So that the scheme is more in line with the expected design objectives , And finally switch all traffic to the scheme that meets the target .

1.2 RFM analysis

RFM Model is an important tool to measure customer value and customer profitability , It's the most popular 、 The most simple 、 One of the most effective customer segmentation methods .

Recency Last consumption ： Time from the user's last consumption to the present . for example ,1 The number of users who spent a week ago is more than 1 The user value consumed years ago is great .
Frequency Consumption frequency ： The number of times a user purchases goods in the statistical period . for example , The value of users who purchase frequently is greater than that of customers who come once in a while .
Monetary Consumption amount ： The total amount consumed by the user in the statistical period . for example , The more you consume, the more valuable you are .

1.3 Funnel analysis / AARRR

Funnel analysis model is a set of process analysis model , It has been widely used in traffic monitoring 、 In daily operation and data analysis, such as product target transformation , It can help us grasp the efficiency of each transformation node , Can intuitively find the problem , So as to optimize the whole business process .

AARRR It is a product life cycle , It describes the depth of user participation in different stages , namely ： Acquisition（ Get users ）、 Activation（ Activate ）、Retention（ Improve retention ）、 Revenue（ increase income ）、 Referral（ Spread recommendations ）. It can pass the conversion number of users between layers , I.e. conversion rate , To locate the problem .

1.4 Simultaneous group analysis

Simultaneous group analysis , It is through analysis that the properties are exactly the same 、 The change of comparable groups over time , To analyze what factors affect the retention of users . Just a simple chart , It directly describes the retention or loss of users over a period of time . It is very important in the field of data operation , In particular, Internet operations require careful insight into retention .

1.5 comparative analysis

Comparative analysis mainly refers to the comparison of two interrelated index data , Show and explain the scale of the research object in terms of quantity , Level , Speed and other relative values , Through the comparison of indicators under the same dimension , You can find , Identify problems at different stages of the business . Common comparison methods include time comparison , Spatial contrast , Standard comparison .

（1） Time contrast

The most commonly used are year-on-year and month on month , By comparing data over time periods , Understand the current level of data .

Year on year ： Compare the same period of the previous cycle . for example , This year, 6 Months are better than last year 6 month .
Chain ratio ： Make a comparison between two periods of equal duration, for example , This year, 6 Months are better than last year 5 month .

（2） Standard comparison

Through the comparison between the current data and the set goal plan , Understand the current development process , Completion schedule, etc , After understanding the gap, you can adjust the strategy in time . for example ： Set the target value in the chart 、 Average 、 Median and so on , Compare with the actual data , Analyze the data .

（3） Spatial contrast

Compare with the data of different spatial indicators within the same time range, for example ： Comparison of order sales data among provinces , It can be concluded that the advantages of the product areas, key breakthroughs , Balance manpower and material resources

1.6 Source analysis

Source refers to how the users who visit our website get to our website . To analyze different channels in depth 、 Different stage effects , Can pass SEM Cross analyze the source channels such as paid search and the user's region , Get the details of customers in different regions . The finer the dimension , The more valuable the analysis results are , So as to guide the optimization of the website , Finally, the goal of improving the conversion rate of users .

1.7 Subdivision analysis

（1） Multi layer drilling

Nesting each layer of data , Click on different dimensions of data , Do a breakdown analysis , Drill through multiple layers , Click directly in the chart to see the breakdown data , Each layer of data can be displayed by selecting the appropriate chart type .

（2） Focus on running in

For some key data in the data , Focus analysis , In a holistic analysis , Want to see some of the data details of particular concern , You can use the focus and drill in functions , Free analysis .

1.8 User analysis

Common user analysis methods include ： Active analysis , Retention analysis , User segmentation , User portrait , User detailed inspection, etc .

With 『 Active analysis 』 For example , User activity can be subdivided into browsing activity 、 Active interaction 、 Active trading, etc , Through the subdivision of active behavior , Master key behavioral indicators . then , Sequence of events through user behavior , Group user attributes , Observe the access of clustered users 、 Browse 、 register 、 Interaction 、 Trading, etc , So as to truly grasp the characteristics of different user types , Provide targeted products and services .

1.9 Clustering analysis

Cluster analysis is an analytical method that divides data into relatively homogeneous groups . Clustering in website analysis is mainly divided into ： User clustering 、 Page or content clustering or source clustering . User clustering is mainly reflected in user clustering , User tagging page clustering is mainly similar 、 Related page grouping , Source clustering mainly includes channel clustering 、 Key words .

Two 、 Data mining and machine learning applications

2.1 Supervised learning

classification
- Logical regression
- Naive Bayes
- Decision tree
- Random forests
- K a near neighbor
- Support vector machine
Return to
- Linear regression

2.2 Unsupervised learning

clustering
- K Mean clustering
Dimension reduction
- Principal component analysis PCA

Data and code download

The code for this tutorial series can be found in ShowMeAI Corresponding github Download , Can be local python Environment is running , Babies who can surf the Internet scientifically can also use google colab One click operation and interactive operation learning Oh ！