Russian Federal Property Deductor

The main objective of the Russian Federal Property Fund – carrying out competitive selection of the best scientific projects from number of those that are provided to Fund by scientists in an initiative order, and the subsequent organizational and financial providing the supported projects.

The organization of activity in Fund allows each researcher regardless of a rank, age and departmental accessory on equal terms to participate in tenders and to get a financial support in cases of success to the researches. Important achievement of the Russian Federal Property Fund can be considered creation of an effective system of examination of scientific projects.

Information was going to the Russian Federal Property Fund many years during which both the technology base, and the principles of data collection, and information systems, and DBMS and many other things changed. All this and also need of input of personal information and data on researches by scientists objectively led to the fact that the saved-up base contained a large number of errors: duplicates, typos, anomalies, admissions and so forth.

However the value of collected data even taking into account the errors which collected for years of use of an information system is very big. For quality improvement and accuracy of the data and also for solving of tasks on development of information systems of Fund it was necessary to book audit of the saved-up data, to estimate their quality, to classify types of errors, to develop recommendations about improvement of a current status and to prevention of quality degradation of data in the future.

The solution of an assigned task was developed based on the analytical Deductor platform. In a work progress the procedure of automatic data validation on existence of errors on the following categories was implemented:

1. Technological level. The problems connected with violation of structure and integrity of data, existence of not informative fields, use of different formats of input in one field, violation of logic of fields, etc. were estimated.
2. Analytical level. Abnormal values, contradictory and duplicative records, typos, dummy values, admissions, input errors, etc. came to light.

On the basis of the received results the following reports were prepared:

• Comparative characteristic of the used databases, the description of a current status of information systems.
• Errors in structure of data collection, including recommendations about its change for ensuring better collecting of data.
• Errors in filling of the available databases: classification of types of problems, statistics, criticality and origins of errors, possibility of correction, recommendation about improvement of quality of data.

The received recommendations will allow to improve significantly quality of information filling of bases of Fund and will form the basis of a monitoring system of quality of the data arriving and accumulated in the Russian Federal Property Fund.