2.90:Reports

From Grooper Wiki
Revision as of 16:06, 2 January 2024 by Dgreenwood (talk | contribs) (Dgreenwood moved page Reports to 2.90:Reports)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
‼

WARNING OBSOLETE TECH DETECTED!!

Reports were removed from Grooper in version 2021.

The reporting functionality they produced was replaced by the "Stats" page in the Grooper web client in version 2023.

Reports are found and created in the "Reports" node.


Reports display information collected from various Grooper operations.


About

Individual Reports are created as Report Instances in Grooper. What information the Report Instance displays is defined by the Report Type property. There are several "System Reports" that ship with every Grooper install. Many of these are designed to track batch processing automation rates or the productivity of data entry clerks. For example, the "Keystrokes" report tracks the number of keystrokes logged during Data Review for Batches using a specified Batch Process.


There are twelve System Reports that are included in every Grooper installation.

  • Batch Statistics
  • Data Capture Automation
  • Data Review Operators
  • Event Log
  • Keystrokes
  • Pages Scanned
  • Production Batches
  • Project Billing
  • Query Report
  • Scan Operator Efficiency
  • Scan Volume
  • System Health

Each of these reports can be saved as a Word, PDF or Excel file.

Custom reports can also be generated in an Object Library using Microsoft Visual Studio and the Visual Studio Report Designer.

Report Types

The Batch Statistics report details information about each Activity in a Batch Process. This report will give different information, depending on the activity steps in the Batch Process fro a given date range. You can use this information to get a better idea of how good your accuracy, productivity and automation rates are.


1. To run a Batch Statistics report, first change the Report Type property to Batch Statistics

2. Select the Batch Process you wish to report. Here, it is a simple demo process named "Dummy Batch Process".

3. Select a date range for the reporting period using the Report Period property.

4. Press the "Run Report" button to run the report.
  • This will display various information about all batches using the selected process.
  • Each step will contain different information for the activity. For example, for the Recognize the total number of OCR characters recognized will be displayed. You can use this information to get a better idea of how good your accuracy, productivity and automation rates are. The Classify activity will show total folders classified and total folders confident. You can use the different between these values to see how well your Content Model is classifying your documents. The Extract activity will show you how many fields were extracted in error.
  • All activities will have "Tasks Processed", "Run Time", "Processing Rate", and "Processing Time" values.

Back to the top

The Data Capture Automation report returns information from the Data Review activity. This includes total records completed for a Batch Process and total records edited. The percentage of records edited of records completed gives you an "Automation Percentage". If you don't have to edit any field, your data extraction is perfect and your automation rate would be 100%. The report also displays the total number of keystrokes and keystrokes per record.


1. To run a Data Capture Automation report, first change the Report Type property to Data Capture Automation

2. Select the Batch Process you wish to report. Here, it is a simple demo process named "Dummy Batch Process".

3. Select a date range for the reporting period using the Report Period property.

4. Press the "Run Report" button to run the report.
  • This will display information about how many document records were edited during Data Review.
  • This will give you an idea about how well the data extraction portion of your Batch Process automates.
  • For this example, we didn't do too well. At least one field on every document had to be manually updated during data review (Either the field was blank and needed to be entered in, or the wrong value extracted and needed to be changed). So, the "Automation Percentage" is "0.00%". This gives me an idea that my extraction logic in my Content Model could be improved.

]

Sure enough, there was a Data Type extractor not picking up values for a field in my Content Model. After another batch runs through, the "Automation Percentage" changes to "33.33%", reflecting the change.

Back to the top

How To

Create a New Report

Add a New Report Instance

1. Right click the Reports folder in the Node Tree.

2. Select "Add" and "Report Instance..."

3. A pop up window will appear to name the Report Instance. Here, we chose to name it "Report Example". Press the "OK" button to add the Report Instance to the Reports folder.

Select the Report Type

4. Using the dropdown list on the Report Type property, select which report you would like to configure.

Using the Query Report to Generate Custom Reports

The Query Report allows you to perform a SQL query against any SQL database. This includes the ability to query the Grooper SQL database. If the information you want to is query-able in some way from the Grooper database tables, you can make a custom report targeting it!

Create a Data Connection to the Grooper Database

1. Navigate to the Data Connections folder in the Infrastructure folder. Traverse the Node Tree following this path.
Root Node > Infrastructure > Data Connections
2. Right click the Data Connections folder. Select "Add" and "Data Connection..."
3. A pop up window will appear to name the Data Connection. Since we are connecting to the Grooper Database, we named ours "Grooper Database". Press the "OK" button when finished to add the Data Connection to the Node Tree.

Enter Database Connection Settigns

1. Expand the Connection Settings properties. Use the Server Name property to point to the Grooper database's server.
  • If you installed Grooper using SQL Express, the server will be a named instance of the machine. You will enter Server_Name\GROOPER
  • For standard SQL installs, simply point to the server on which it is installed.
2. Select the database name for the Grooper Repository. This defaults to "Grooper" but you may have named it something different.
3. Enter the User Name and Password to access the database, if needed.

FYI

If you are unsure what your Grooper Database's Server Name or Database Name is, open Grooper Config and select the Grooper Repository you are attempting to query. The Server Name and Database Name will be listed in the property panel.

Once connected, you will be able to see the thirteen database tables in the Grooper Database

If you select one of these tables, the "Data Preview" panel will show you a preview of the data within that table.

Back to the top

Add a New Query Report

1. Add a new Report Instance.
2. Select the Report Type property and select Query Report.

3. Using the Database Connection property, select the Database Connection to the Grooper Database established in "Step 1" of this tutorial.

Next, you will use the Query Statement property to write a SQL query to return whatever information you desire from the Grooper Database. The next few tabs provide some examples of SQL queries and their results when the report runs.

Back to the top

Count Objects in a Repository

This is a simple query to return the number of Content Models in a Grooper Repository. The query is as follows: 

SELECT COUNT(TypeName) AS Content_Models
FROM TreeNode
WHERE TypeName='Grooper.Core.ContentModel'
1. Enter the SQL query using the Query Statement property. Expand the dropdown menu. This is a text editor for you to type the full query or paste it from SQL Server Management Studio or wherever you like to craft your queries.

2. Press the "Run Report" button. As you can see, there are five Content Models in this repository, and the report returns a value of "5".

You could use this query to count any object in the repository. Just change the value in the WHERE TypeName='Grooper.Core.ContentModel' clause to match the Grooper Type Name of the object you're looking for. For example, if you wanted to return the count of Data Types you would instead use WHERE TypeName='Grooper.Core.DataType'

Back to the top

Count Multiple Types of Objects in a Repository

This query is very similar to the one from the previous example, but counts multiple different types of objects instead of just one. Specifically, this will return the number of Content Models, Data Types and CMIS Connections. 

SELECT TypeName AS Object_Name, COUNT(TypeName) AS Obect_Count
FROM TreeNode
WHERE TypeName in ('Grooper.Core.ContentModel', 'Grooper.Core.DataType', 'Grooper.CMIS.CMISConnection')
GROUP BY TypeName


Just like in the previous example, enter the SQL query using the Query Statement property. The dropdown menu will allow you to type in the SQL query.

When the report runs, it returns a count of the three types of objects searched for in the Node Tree.

  • Note as well, the "Appearance" properties control how the report is formatted. You can change the column widths, header and data font sizes, page layout and the table cells' background colors.

Back to the top


Attended Activity Time Tracking

This query returns the time (in seconds) users have spent on the attended activities Classify Review and Data Review in the last day. 

DECLARE @today datetime=GETDATE()
SELECT UserName, StepName, DATEDIFF(second,SessionStats.StartTime,SessionStats.EndTime) AS Processing_Time, BatchName
FROM SessionStats
WHERE StepName='Classify Review' OR StepName='Data Review' AND EndTime>DATEADD(day,-1,@today)
ORDER BY UserName, StepName, Processing_Time DESC


Just like in the previous example, enter the SQL query using the Query Statement property. The dropdown menu will allow you to type in the SQL query.

When the report runs, it searches the "SessionStats" table, which collects the start and end time for Classify Review and Data Review activities from processed Batches and returns the total time spent for each instance.

Back to the top

If you're really serious about using the Query Report to build custom reporting, you may want to explore the "TreeNode" table. This single table has the most information on every single object in your Grooper Environment. This includes the values of all properties on every object. You can even see the regular expression pattern used by extractor objects!


Version Differences

There are no notable version differences at this time.

Retrieved from "https://wiki.grooper.com/index.php?title=2.90:Reports&oldid=17799"