Integrate Multiple Relevant Tuple Units to Effectively Answer Keyword Queries Over RDBMS

Existing studies on keyword search over relative information typically realize Steiner trees composed of connected database tuples as answers. Recently, tuple units square measure planned to boost search potency by classification structural relationships, and existing ways establish one tuple unit to answer keyword queries. However, in several cases, multiple tuple units ought to be integrated to answer a keyword question. Thus, these methods will involve false negatives. To address this downside, during this paper, we tend to study the way to integrate multiple connected tuple units to effectively answer keyword queries. To attain a higher performance, we conceive two bond indexes, SKSA index and KPSA index, and incorporate structural relationships between different tuple units into the indexes. We use the indexes to expeditiously establish the answers of integrated tuple units. We develop new ranking techniques and algorithms to increasingly realize the top-k answers. We have enforced our methodology in real information systems, and therefore the experimental results show that our approach achieves high search potency and result quality, and outperforms progressive ways considerably.