Question Correlating Similar Entries for Database (1 Viewer)

Bill Bisco

Custom User Title
Local time
Today, 13:08
Joined
Mar 27, 2009
Messages
92
Dear all,

I run a database for defect detail, tracking, and resolution. Often times we have the same type of defect occur multiple times before the problem is resolved. Currently the process is that every time a defect occurs, it gets entered into the database, and whenever it gets solved, those who have solved it have to go to the database and mark it as complete.

However, since every time a problem occurs, a record is created, the people who mark issues as complete have to go through and find all similar records of problems which are basically the same problem and mark those as complete as well.

Has anyone implemented a solution to something similar to the above? Let me know; any help would be greatly appreciated.

--Bill
 

spikepl

Eledittingent Beliped
Local time
Today, 20:08
Joined
Nov 3, 2010
Messages
6,142
Did you mention what exactly you are unhappy about?
 

Bill Bisco

Custom User Title
Local time
Today, 13:08
Joined
Mar 27, 2009
Messages
92
I just need a way to be able to easily tell when someone is typing if there is already a similar issue for that problem. That way I can just mark the issue as an incident of one problem.
 

spikepl

Eledittingent Beliped
Local time
Today, 20:08
Joined
Nov 3, 2010
Messages
6,142
Define "similar". You have not provided any specifics, so it is difficult to conjure any tricks. Machines are not very good at "similar". People are much better. If your "issues" have some discrete characteristics, then the machine can easily match the ones with same characteristics. A narrative is much more difficult to match.
 

jdraw

Super Moderator
Staff member
Local time
Today, 14:08
Joined
Jan 23, 2006
Messages
15,385
I agree with spikepl's responses -- it's difficult to see exactly the problem you are describing(quite general). We don't have any idea of the "problems" or "similar" etc.

Have you considered some sort of categories or keywords---to help the person doing entry or marking completed to look at several/many individual records.

If you put/tagged a problem with 1 or more keywords, you might be able to see more of the "similar" issues at one time.
Now you should have a defined set of keywords/tags and some education/training for all involved as to what these mean and when/how to use them.

Suppose your "problems" dealt with some switches ( dreamed up example)
Code:
[COLOR="Blue"]Problem   [/COLOR]                          [COLOR="Blue"]Tags[/COLOR]
turns off by itself                function, on-off
threads are stripped            function
color is too dark                 color, appearance
color is too light                 color, appearance
color is inconsistent            color, appearance

By reviewing Tags, you may be able to see "similarly tagged problems".

My thinking is that there must be some sort of hierarchy -- real or practical -- by which "defects" are identified. Going up one level in the hierarchy should show a number of "similar (by some means)" defects.
 
Last edited:

spikepl

Eledittingent Beliped
Local time
Today, 20:08
Joined
Nov 3, 2010
Messages
6,142
I agree with jdraw, but wanted to hear some specifics before guessing too much. Defect reporting and tracking can be a minefield, because causes can be mixed with root causes and interspersed with symptoms etc etc, which is why "similar" is not necessarily obvious.
 

Users who are viewing this thread

Top Bottom