The first phase of the Question Bank Management System

After intense thinking and discussion last week, the core part of the question library management system is basically solved, that is, the ratio of question scores with different difficulty coefficients should be determined based on the question Difficulty coefficient, in addition, the scores of students must be normally distributed.

-- Question

Finally, we have time to sit down and summarize the activities of the last week.

First, start from
3
Month
26
The question bank management system needs to be analyzed, and then the requirements are constantly modified. After that, I use
PD
(
Powerdesigner
Modeling the data to obtain the database design documents. Then the system is designed as a whole, using
MVC
Architecture, or use
PD
Model the system
3
Month
29
Complete the overall modeling of the system.

But these are not perfect. Why? Later, we discussed some aspects of the system, including requirements and database design. After some column discussions, I finally reached an agreement. Because the system was relatively simple, Xue Yu completed one side from the beginning to the end and constructed the general "framework. The trouble begins at this time.

The first is the core of the system, that is, the question extraction algorithm. To be honest, these things related to mathematics are indeed very technical, so it is necessary to spend some time thinking. The biggest headache was to find a paper on question extraction from Wanfang, and then thought that the paper was correct. Then we started the research and the results were studied for three or four days, I found that the algorithms and formulas provided by the paper are inconsistent with the intention of the author to draw questions, that is, the saying goes, "the deviation ".

Because we have no basis, we can only find an algorithm to implement such an algorithm that satisfies the normal distribution. Under the guidance of Mr. Mi (this sentence is very official), we continue to discuss and finally found a solution.

The description is very simple, but the process is complicated. For this reason, I and xueyu brain have been frequently used for several days. Fortunately, xueyu brain has a higher frequency and finally finds a method.

 

In addition to the question extraction algorithm, there is also a difficulty in editing the questions online and storing questions. I have been thinking about this for a long time, and I have also tried two widgets (
Free textbox
And
Cuteeditor
), But the results are not ideal. So far, it is still in the exploratory state.

However, in any case, technical issues are not a problem. The only problem is non-technical ones. For example, what kind of status should the results be displayed in real life, or abstract the actual problems in life into technology.

In short, last week was a really tough week. Because we are always working to solve algorithm problems. But such a life is also very interesting. At our present level and some things we have done, sometimes it is exciting to encounter something challenging.

PS
: I should have read several literary books. I couldn't help but think of a sentence: If I didn't read a book for three days, I felt the language was tasteless and ugly.