The Data Mining of Lanzhou University of Finance and Economics

  Recently,The mass of data was  grabbed from the post bar of our school  by myself  using  web spider, and I got some interesting information from these data. Let’s see.

Summary

  There are  altogether 734,930 items grabbed from the post bar, and I just only selected some of them about 600,000 when is between from 2014 and 2016.

No.1 Some  interesting post

  It’s amazed that   the number of followed posts of some posts  is more than  2,000,and I pick out some of them that is not trading post .

title	url	the number of followed post
你敢自爆自己生日吗，同一天的做一个月情侣，男的就。。。。你懂的	http://tieba.baidu.com/p/3052538643	5077
有没有人陪我聊天，直到楼中楼过千？	http://tieba.baidu.com/p/3383923307	3615
[问答]200个暗恋问题你能答到不哭吗？	http://tieba.baidu.com/p/4092428927	2954
青春碎语	http://tieba.baidu.com/p/2576965665	2621
仙梦侠义,烽火扬洲,看英雄精英,豪情壮志,今朝谁主沉浮?	http://tieba.baidu.com/p/341221031	2601
你是什么类型的女孩子   答完请做好被人追的准备	http://tieba.baidu.com/p/2850779865	2572
晒晒你的高中校名	http://tieba.baidu.com/p/2940696342	2476
【心理测试】美国著名心理测试，看看你潜意识的自己	http://tieba.baidu.com/p/2708611306	2105
（转）分过手的进，一共69道题，不多，认真回答好么	http://tieba.baidu.com/p/3834605332	2022
若惜缘 ——水军第二阵地	http://tieba.baidu.com/p/1490523084	2010

 

No.2  The type of  smart phone  used by Lanzhou University of Finance and Economics

  At the first,it’s reasonable according statistics –estimating population by using sample–  that I estimate the type of smart phone used by Lanzhou University of Finance and Economics from the mass of data ,and let’s observe a chart as follow.

  It’s easy that we could get some information from the chart. This isusual Android which the measure is  greatest each year. And the following tightlytype is Apple, which the market shares of apple is increasing, It’s seems that thenumber of  rich students of our school isincreasing.

  But why am I  poor as always?

  The market shares of others, such as XiaoMi 3,fell nearly to 0 in the three years for the invading of  Apple and Android .

 

No.3  The Activity Regular Of The Student  

  I attained some charts and some result which can show the activity regular of students by the data analysis.

 

The green chart is the distribution of the posting time, and thewhite curve is Gaussian distribution. It  indicates that the activity regular of  students is not in accord with Gaussiandistribution—the Q-Q is also show the result. The second chart is boxplot, whichshows the  median is 14:36:25 and the 25percent of students  concentrate on  between from  19:54:46  and 00:00:00.  

There are some other results of data analysis here.

Mean:14:17:45

Mode:00:02:00

95% confidence interval :[14:16:45,14:18:46]

No.4  The Activist Of Post Bar

Members of top 10  are asfollow.

  According the three sigma theory, the member  of post bar  that the number of posts is three times greater than the  mean is exceptional. In particularly,I use the five standard error to find the exceptional, So these member isselected fond as follow.

 

No.5 The Post Bar Time Series Analysis

  The chart of the post bar data during three years is  draw.

In addition, the incrementsequence is draw by making the difference of values.

 

  It show that the activeness of our school’s post bar is stable during the three years. And the data fitted by using fitting of a polynomial, whichit  seems to indicates that the activeness will increase.