tl = the duration of the loading process.
te = time elapsed between the moment when data are available for querying in
the data warehouse, and the moment when the query is posed.
tq = the query response time.
??? Volatility:.It represents the length of the interval during which data are valid
in the real world (Wang & Reddy, 1992). Pipino, Lee, and Wang (2002) define
Timeliness as a function of currency and volatility:
Timeliness (d)=MAX [1??“currency (d)/ volatility (d), 0] s , where s > 0
The coefficient s (not considered in our methodology) is denoted sensitivity;
it reflects the criteria of the analyst, and depends on the task being performed.
Timeliness ranges between 0 (worst case) and 1 (desirable value).
??? Data source availability: It is the time during which the data source is available
(Jarke, Lenzerini, Vassiliou, & Vassiliadis, 2003).
??? Expected.query.response.time: It is the maximum accepted time for getting
the answer to a query.
??? Query.frequency: It is the minimum time between two successive queries.
Measuring.Quality.
There are many different ways of analyzing and measuring the required data quality
parameters. Thus, it is necessary to define a common way of specifying user
needs and measuring whether the DSS or the data warehouse will be able to fulfill
the minimum levels of quality required. To this end, we propose to apply GQM to
each one of the dimensions defined previously.
Pages:
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159