Rule-based approach for topic maps learning from relational databases

摘要: Relational databases RDBs have been widely used as back end for information systems. Considering that valuable knowledge interwoven in between stored data, how to access, represent and share this becomes an important challenge. Topic maps TMs emerge a good solution problem. However, manual development of is difficult, time-consuming subjective task if there no common guideline. The existing building approaches mainly consider the meta-information contained RDB, without considering residing database content its current state. Other require predefined configuration applying specific data transformation. This paper proposes automatic method TM construction based on learning rules. Our considers background during process was implemented applied representative set 15 RDBs. resulting were validated syntactically using standard tool semantically through inference formal query language. In addition, analysis relational input representation output conducted. results found our experiments are encouraging put evidence soundness proposed method.