Supporting keyword search in temporal databases
Dr Ling Tok Wang, Emeritus Professor, School of Computing
COM2 Level 4
Executive Classroom, COM2-04-02
Abstract:
Many organizations, especially in regulated industries such as finance and healthcare, need temporal databases to manage and maintain data that changes over time. Keyword search over temporal databases provides an easy and convenient way for non-export users, like financial analysts and doctors, to querying temporal databases without constructing complex SQL queries. However, the existing works did not consider the Object-Relationship-Attribute (ORA) semantics of the temporal databases in both database schema design and keyword query processing aspects, which leads to the inability of database to capture intended temporal and non-temporal semantics in real world, and the inability of the keyword query processing to capture intended query interpretations and return incorrect search results.
To solve these problems, we propose semantic approaches to design temporal database schema and process temporal keyword query. In the temporal database schema design, we first use a temporal ER diagram to capture the temporal and non-temporal ORA semantics in real world, and propose two sets of mapping algorithms to generate database schema in either normal form relations or nested relations, which preserving both temporal and non-temporal ORA semantics and avoiding data redundancy caused by the temporal attributes. Our database schema also separates the storage of current and historical data, which accelerates the speed of queries over current records. In the temporal keyword query processing, we adopt the Object-Relationship-Mixed (ORM) schema graph to capture the temporal and non-temporal ORA semantics of the database. With the ORM schema graph, we study the various ways a time condition can be applied to the objects, relationships or attributes in the database to capture users' search intention, and also support processing complex temporal keyword queries involving temporal joins and implicit time periods.