Based On The Data Of The Data Space Activities Relationship Discovery And Evaluation Of The Importance Of Data

A DataSpace is composed of data and the relations of data. It is designed to provide users with more convenient data management. DataSpace support a variety of data sources, and provide more convenient help for users to store, find, update, and manage data. Meanwhile, it also has the ability to self-evolution, with data sources and the increase in the amount of data. The system analyzes the user's use of the DataSpace system and adjusts itself automatically. Therefore, the DataSpace must be able to discover and extract new data model, find the relationship between data and the important data for the user. The main contents of this paper include:1. Research on using user's activity data to find relationships between data. We propose an algorithm that finds relationship between data by analyzing user's activities. An activity collection system is used to collect user activity information. Then the algorithm analyzes user's activity, extracts correlation between activity information based on user activity recording information, and saves as activity correlation document. Finally, the relationships between data based on the activity correlation document are extracted2. Research on using the user's activity data to evaluate the importance of data. This paper proposes an algorithm which found the important data for user based on user's activity. The algorithm uses activity information records and data related documents as data sources, and through analysis the frequency of data which store in data related document, the frequency of related-data, user relevance of data and the recent use of the user to calculate the importance of data for users.This paper regards user's activity as evidence to find relationships between data and evaluate the important data for the user. Preliminary experiments show that the data relationship discovery subsystem and the important data evaluation subsystem can find relationships between data and important data respectively, which provide more effective service for DataSpace user.