One way of making sense of the numerous audits information is to associate them with various categories. Two such categories are the “audit areas” and “audit issues.”
The public entities are involved in a wide range of business activities.
Sometimes, the AG’s office base the individual audits on a particular functional area like procurement, asset management, enforcement and so on. We term these collectively as the audit areas in the information graph.
By associating the audits’ information with one or more business activities, the user will be able to collate audits in the same or similar audit areas. This process produces a list of audits’ information that is recognized by the system. Manually, this is a tedious and time-consuming task but once the audit reports have been processed; this becomes a trivial task for the system. It is a case of doing it once and using it repeatedly.
Unfortunately, the National Audit Department has not published the lists of audit areas and audit issues. We have to construct these lists by processing the audits’ information to extract the items. In the parlance of the information management discipline, the items in the audit areas and issues lists are termed as “named entities.” We developed a named-entities recognition utility to speed up this work. The utility has to be trained to include a list of matching phrase with embedded wildcards for each named -entity. Based on the suggested named-entities, the user will have to verify the choice of these entities by searching for the phrases in the text.
Our lists are work-in-progress, and they will need to be updated from time to time as we process more audit reports.
What is in the Information Graph?
The above screen shows the list of audit areas. The alert numbers show the number of audits’ information associated with the particular item.
The above screen shows a partial list of the audit issues with the alerts indicating the number of associated audits.
The above screen shows part of the Agricultural programmes audit area. The 54 audit reports have been grouped against the associated audit issues. Click on the above image to view the live screen.
The above screen shows the full list of the audits with issue on Financial regulations. This list can be grouped in many ways using the dynamic list processing facilities provided. The first tabbed page above for Audit areas is a predefined processed list.
The above image shows the “Process list” form opened.
The above image shows the relationships pull-down list. For this demo, we select associate.
After selection of the relationship, the “Group by” pull-down list is populated.
The group-by options are generated from the relationships in the information graph. For this demo, we select “Federal Ministries.” You may try with the other items in the list.
The above screen shows the audits with the Financial regulations issues grouped by Federal Ministries. This process can be repeated for all the 19 group by options, and for all the items in the audit issues list. Similar list processing can be performed in the audit areas.
The above screen shows a typical cross tabulation of the State GLCs against the Human Resource management issues.
This illustrates the very powerful and flexible ways that information can be reorganized in the information graph for various purposes. Nevertheless, we still require the human to read and digest the reduced list of audits’ information. We could do with more categorization, but we are in some ways constrained by the content of the annual AG’s reports.