Database Reference
In-Depth Information
one person involved
.
The calculated time period from
Arrival-Date
and
Closed-Date
is than multiplied by the number of people yield from the
Responsible
attribute. We
have also experimented with discretizing this attribute with cutting points be one day
(1), half week (3 days), one week (7), two weeks (14), one month (30) and one
quarter (90 days), half year (180 days) and more than one year (360 days). The main
reason behind this exercise is to give an approximate estimate. It allows for a minor
change in human resource, and being not highly dependent on the exact human
resource involved.
PR_ID|Category|Severity|Priority|Class|Arrival-Date|Close-Date|Synopsis
a.
17358|bambam|serious|high|
sw-bug
|
20:50 May 25 CST 1999
|
11:35 Mar 24 CST
1999
| STI STR register not being reset at POR
b.
17436|bambam|serious|high|support|18:10 Mar 30 CST 1999|12:00 May 24 CST
1999| sequence_reg varable in the RDR_CHL task is not defined
c.
580 |bingarra|serious|low|doc-bug|10:10 May 31 May 1996| | In URDRT2 of
design doc, the word 'last' should be 'first'
d.
6205 |gali|serious|medium|
SW-bug
|14:30 Nov 5 1997|13:14 Dec 1 1997|
grouping of options in dialog box
Fig. 2.
Data examples from the original PR data set
Severity |Priority| Time-to-fix| Class |Synopsis
a.
serious, high, 61, sw-bug, STI STR register not being reset at POR
b.
serious, high, 56, support, sequence_reg varable in the RDR_CHL task is not
defined
c.
serious, low, ?, doc-bug, In URDRT2 of design doc, the word 'last' should be
'first'
d.
serious, medium, 24, sw-bug, grouping of options in dialog box
Fig. 3.
Data examples ready for mining
3.2 Data Modelling and Mining
We have chosen three data mining techniques to analyse the PR data:
•
Predictive modelling on the PR data to make estimation on the time spent to fix a
PR according to the PR properties.
•
Link analysis to discover association among various PR characteristics.
•
Text mining to analyse
Synopsis
field to find out most representative words in the
problem-discussion, showing the major cause of a problem, along with the
relationship of each frequent word with other words to show how are they related.
Search WWH ::
Custom Search