Viewing Quality Reports

Querying Business Quality Reports

The full quality score can be set to 5, 10, or 100 points. By default, a five-point scale is used for quality scoring based on table-associated rules. The scores in different dimensions, such as tables, business objects, and subject areas, are calculated based on the weighted average values of rule scores in different dimensions.

You can query the quality scores of subject area groups, subject areas, business objects, tables, and table-associated rules. For details on the calculation formulas, see Table 1.

**Table 1** Formulas for calculating scores
Object	Formula
Rule	When a quality job that contains a percentage-related rule (either built-in or custom) is created, a quality report can be generated. Percentage-related rules can be classified into positive rules and negative rules. For a positive rule, the higher the percentage is, the better the data quality is. For a negative rule, the higher the percentage is, the poorer the data quality is. Rules that contain the unique value percentage, duplicate value percentage, and valid percentage are positive rules, and rules that contain the null value percentage are negative rules. Positive rule score = Number of data rows that meet the rule/Total number of data rows x Full score (5, 10, or 100 points). Negative rule score = (1 – Number of data rows that meet the rule/Total number of data rows) x Full score (5, 10, or 100 points). If the table is empty (the total number of rows is 0), the positive rule score is fixed at the full score and the negative rule score is fixed at 0 points.
Table	The table score is calculated as follows: ∑(Scores of all rules associated with the table x Rule weight)/∑Rule weight.
Business object	Weighted average value of the scores of all tables under the business object, that is, ∑Scores of all tables under the business object/Number of tables.
Subject area	Weighted average value of scores of all business objects in the subject area, that is, ∑Scores of all business objects in the subject area/Number of business objects.
Subject area group	Average weighted value of the scores of all subject areas in the group, that is, ∑Scores of all subject areas in the group/Number of subject areas.

On the DataArts Studio console, locate an instance and click Access. On the displayed page, locate a workspace and click DataArts Quality.
Choose Quality Monitoring > Quality Job in the left navigation bar.
Click the Business Reports tab, and select a subject and an end date to query the quality scores of the end date and the previous seven days, as shown in Figure 1.

Figure 1 Business object
- Take the full score 5 points as an example. Points 4 to 5: excellent; 3 to 4: good; 2 to 3: fair; 1 to 2: qualified; 0 to 1: unqualified.
- The quality score data of a day is generated in the early morning of the next day.
- In the Quality Scoring Changes area, the solid line consists of the quality scores of the end date and the previous seven days, and the dashed line indicates the average quality score of these days.
- If the job is executed multiple times on a day, the last score is used as the quality score of the day.
Click the score link in the Table Score column to expand the scores of the rules associated with the table.
Click the score link in the Rule Score column to expand the scores of the fields associated with the rule.

Figure 2 Table-associated rule scores

Viewing Data Quality Reports

The full quality score can be set to 5, 10, or 100 points. By default, a five-point scale is used for quality scoring based on table-associated rules. Scores in different dimensions, such as tables and databases, are calculated based on the weighted average values of rule scores in different dimensions.

You can query the scores of databases, tables, and table-associated rules. For details on the calculation formulas, see Table 2.

**Table 2** Formulas for calculating scores
Object	Formula
Rule	When a quality job that contains a percentage-related rule (either built-in or custom) is created, a quality report can be generated. Percentage-related rules can be classified into positive rules and negative rules. For a positive rule, the higher the percentage is, the better the data quality is. For a negative rule, the higher the percentage is, the poorer the data quality is. Rules that contain the unique value percentage, duplicate value percentage, and valid percentage are positive rules, and rules that contain the null value percentage are negative rules. Positive rule score = Number of data rows that meet the rule/Total number of data rows x 5. Negative rule score = (1 - Number of data rows that meet the rule/Total number of data rows) x 5.
Table	The table score is calculated as follows: ∑(Scores of all rules associated with the table x Rule weight)/∑Rule weight.
Database	Weighted average value of the scores of all data tables in the database, that is, ∑Scores of all data tables in the database/Number of tables.

On the DataArts Studio console, locate an instance and click Access. On the displayed page, locate a workspace and click DataArts Quality.
Choose Quality Monitoring > Quality Job in the left navigation bar.
Click the Technical Report tab, and select a data connection and an end date to query the quality scores of the end date and the previous seven days, as shown in Figure 3.

Figure 3 Selecting a data connection
- Take the full score 5 points as an example. Points 4 to 5: excellent; 3 to 4: good; 2 to 3: unqualified; 1 to 2: poor; 0 to 1: very poor.
- The quality score data of a day is generated in the early morning of the next day.
- In the Quality Scoring Changes area, the solid line consists of the quality scores of the end date and the previous seven days, and the dashed line indicates the average quality score of these days.
- If the job is executed multiple times on a day, the last score is used as the quality score of the day.
Click the score link in the Table Score column to expand the scores of the rules associated with the table.
Click the score link in the Rule Score column to expand the scores of the fields associated with the rule.

Figure 4 Table-associated rule scores