-
Qing Fei commented on the post, Weekly Question #8: Complete by November 2, 2017, on the site 7 years, 6 months ago
The most important takeaway from this course is how to analyse data and visualize data in an effective way. Data science plays a more and more important role in today’s society, so it is necessary for everyone, including who are not data science major, to learn how to use data to tell good stories, and make it simple and clear for the audiences to…[Read more]
-
Qing Fei commented on the post, Weekly Question #10: Complete by April 17, 2017, on the site 7 years, 6 months ago
A data-driven service I use is online banking. Each column would be different accounts, and each row would represent detailed account information, such as account number, current balance, credit line, credit available, transaction history, and the information about each transaction, like transaction amount, posting date, description, transaction…[Read more]
-
Qing Fei commented on the post, Weekly Question #8: Complete by April 10, 2017, on the site 7 years, 6 months ago
https://fivethirtyeight.com/features/one-slow-month-doesnt-change-the-strong-trend-on-jobs/
This article talked about the slowdown in U.S. job increasing rate in March, which is out of expectations. However, the author explained that there was no need to worry too much about these data, and the author listed some possible reasons about this…[Read more] -
Qing Fei commented on the post, Weekly Question #7: Complete by March 27, 2017, on the site 7 years, 7 months ago
My alarm clockis an example of a KPI. It’s specific and measurable because I can set spcific alarm time, like seven hours later. It is achievable because it rings at designed time points. It’s relevant it determines my sleeping time. It’s time-variant because the alarm time in weekends is later than in weekdays.
-
Qing Fei commented on the post, Weekly Question #6: Complete by March 20, 2017, on the site 7 years, 7 months ago
I haven’t made these mistakes, and I think the third mistake “start working on the database without doing a full backup first” is the most important to avoid. If you made some mistakes other than the third one, you could waste some time to fix the problem, but if you started working on the database without doing a full backup, the mistake might…[Read more]
-
Qing Fei commented on the post, Weekly Question #5: Complete by March 6, 2017, on the site 7 years, 8 months ago
This article is about a new marketing strategy of Fitbit. Fitbit was only used to track how many steps the wearers walk every day, but now the company intends to track the wearers’ sleeping quality. The company collects the data of the users’ sleeping time, and wants…[Read more] -
Qing Fei commented on the post, Progress Report for Week Ending, March 1, on the site 7 years, 8 months ago
I think “compare” is the most principle in data visualization. It is important to show the relationship between different data in data visualization. Single datum does not have too many meanings, so for data sets, it is the distribution of data that matters. Thus, we can see the big picture and have better understanding of the data.
-
Qing Fei commented on the post, Progress Report for Week Ending, February 9, on the site 7 years, 8 months ago
I think data science can affect food industry, and data can either make it better or make it worse. For the good side, because there are data websites in which customers can rate, restaustants will make more efforts to preserve their business images and improve their foods and services. For the bad side, because of the easy access to these…[Read more]
-
Qing Fei posted a new activity comment 7 years, 9 months ago
https://fivethirtyeight.com/features/trump-could-really-mess-up-mexicos-economy/
This article talks about how Trump would affect Mexicos’ economy. A series of anti-Mexico policies that Trump has made would cause a great loss to Mexico’s GDP. The policies include limiting the flow of remittances to Mexico, increasing tariffs, stopping free trade…[Read more] -
Qing Fei posted a new activity comment 7 years, 9 months ago
It is said that drunk people are more likely to tell the truth. This can be tested by examine a group of people’s words when they are sober and drunk, and compare whether their words are more reliable when they are drunk than when they are sober. However, it is difficult to judge if a person tells a truth or a lie, which may affect the result of…[Read more]