Article3101: Unterschied zwischen den Versionen
Lzh (Diskussion | Beiträge) |
Lzh (Diskussion | Beiträge) |
||
Zeile 23: | Zeile 23: | ||
{{Publikation Details | {{Publikation Details | ||
|Abstract=The problem of rewriting keyword search queries on graph data has been studied recently, where the main goal is to clean user queries by rewriting keywords as valid tokens appearing in the data and grouping them into meaningful segments. The main solution to this problem employs heuristics for ranking query rewrites and a dynamic programming algorithm for computing them. Based on a broader set of queries defined by an existing benchmark, we show that the use of these heuristics does not yield good results. We propose a novel probabilistic framework, which enables the optimality of a query rewrite to be estimated in a more principled way. We show that our approach outperforms existing work in terms of effectiveness and efficiency w.r.t. query rewriting. More importantly, we provide the first results indicating query rewriting can indeed improve overall keyword search runtime performance and result quality. | |Abstract=The problem of rewriting keyword search queries on graph data has been studied recently, where the main goal is to clean user queries by rewriting keywords as valid tokens appearing in the data and grouping them into meaningful segments. The main solution to this problem employs heuristics for ranking query rewrites and a dynamic programming algorithm for computing them. Based on a broader set of queries defined by an existing benchmark, we show that the use of these heuristics does not yield good results. We propose a novel probabilistic framework, which enables the optimality of a query rewrite to be estimated in a more principled way. We show that our approach outperforms existing work in terms of effectiveness and efficiency w.r.t. query rewriting. More importantly, we provide the first results indicating query rewriting can indeed improve overall keyword search runtime performance and result quality. | ||
− | |||
|Link=http://www.vldb.org/pvldb/vol6/p1642-zhang.pdf | |Link=http://www.vldb.org/pvldb/vol6/p1642-zhang.pdf | ||
|Projekt=XLike | |Projekt=XLike | ||
|Forschungsgruppe=Wissensmanagement | |Forschungsgruppe=Wissensmanagement | ||
}} | }} |
Version vom 15. Februar 2014, 00:19 Uhr
Probabilistic Query Rewriting for Efficient and Effective Keyword Search on Graph Data
Probabilistic Query Rewriting for Efficient and Effective Keyword Search on Graph Data
Veröffentlicht: 2013 September
Journal: Proceedings of the VLDB Endowment (PVLDB), the 39th International Conference on Very Large Data Bases (VLDB)
Nummer: 14
Seiten: 1642-1653
Volume: 6
Referierte Veröffentlichung
Kurzfassung
The problem of rewriting keyword search queries on graph data has been studied recently, where the main goal is to clean user queries by rewriting keywords as valid tokens appearing in the data and grouping them into meaningful segments. The main solution to this problem employs heuristics for ranking query rewrites and a dynamic programming algorithm for computing them. Based on a broader set of queries defined by an existing benchmark, we show that the use of these heuristics does not yield good results. We propose a novel probabilistic framework, which enables the optimality of a query rewrite to be estimated in a more principled way. We show that our approach outperforms existing work in terms of effectiveness and efficiency w.r.t. query rewriting. More importantly, we provide the first results indicating query rewriting can indeed improve overall keyword search runtime performance and result quality.
Weitere Informationen unter: Link