A. Theobald and G. Weikum. Adding relevance to XML. In WebDB 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....At first glance, extending XQL by a weighting mechanism seems to be straightforward. Assuming probabilistic independence, the combination of weights according to the di#erent Boolean operators is obvious, thus leading to an overall weight for any answer; such an approach has been described in [20]. However, there are two major problems that have to be solved first: 1) How should terms in structured documents be weighted 2) What are the probabilistic events, i.e. which term occurrences are identical, and which are independent Obviously, the answer to the second question depends partly on ....

A. Theobald and G. Weikum. Adding relevance to XML. In 3rd International Workshop on the Web and Databases (WebDB), 2000.

....data and XML in particular have been developed 26 during the last years (see [FSW99, BC00] for surveys) All languages support operators that allow to skip certain parts of the data. However, they require that the user knows which parts have to be skipped. To the best of our knowledge, XXL [TW00] and ELIXIR [CK01] are the only XML query languages stemming from the database community that support result ranking. Both XXL and ELIXIR add a similarity operator to a subset of XML QL [DFF 98] In XXL, the similarity operator can be applied to element names as well as to text sequences. The ....

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of 3rd International Workshop on the Web and Databases (WebDB'00), Dallas, USA, May 2000.

....similarity measures [AG97] Only few other papers investigate result ranking for structured queries against XML documents: Fuhr and Grojohann [FG00] as well as Hayashi et al. HTK00] focus on text rich documents, whereas our language is designed for data centric documents. Theobald and Weikum [TW00] propose the query language XXL, which supports a similarity operator. To use this operator, the user must know that similar keywords or element names exist. In contrast, the approXQL query processor transforms the query automatically in order to find appropriate results. Unlike all approaches ....

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of 3rd International Workshop on the Web and Databases (WebDB'00), Dallas, USA, May 2000.

....(i.e. contents) Furthermore, we investigate related techniques such as distance metric or selectivity estimation to support query relaxation for XML model. From Web perspective, several authors have proposed a language system that supports approximate pattern matching and answer ranking. In [TW00], a similarity operator # for XML is presented. In their work, # measures the similarity between the value of element attribute and the given constant helped by underlying thesaurus. In [FG01] authors describe XIRQL, an extension of XQL [RLS98] that integrates IR features by supporting ....

A. Theobald and G. Weikum. "Adding Relevance to XML". In Int'l Workshop on the Web and Databases (WebDB), Dallas, TX, May 2000.

....found, results similar to the query should be retrieved and ranked according to their similarity. The problem of similarity between keyword queries and text documents has been investigated for years in information retrieval [3] Unfortunately, the most models (with some recent exceptions, e.g. [15, 6, 7]) consider unstructured text only and therefore miss the change to yield a more precise search. Furthermore, it is not clear whether retrieval models based on term distribution can be used for data centric documents as considered in this paper. XML query languages, on the other hand, do ....

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of WebDB'00, 2000.

....results similar to the query should be retrieved and ranked according to their similarity. The problem of similarity between keyword queries and text documents has been investigated for years in information retrieval [BR99] Unfortunately, the most models (with some recent exceptions, e.g. TW00, CK01, FG01] consider unstructured text only and therefore miss the change to yield a more precise search. Furthermore, it is not clear whether retrieval models based on term distribution can be used for data centric documents as considered in this paper. XML query languages, on the other hand, ....

....to the deletion of nodes in our approach; the permutation of nodes has no correspondence in out model. Neither regular path expressions (with the exception of [FFS01] nor the models by Kanza et al. valuate the closeness between the query and its results. To the best of our knowledge, XXL [TW00] ELIXIR [CK01] and XIRQL [FG01] are the only XML query languages that support result ranking. Both XXL and ELIXIR add a similarity operator toasubsetofXML QL [DFF 98] In XXL, the similarity operator can be applied to element names as well as to text sequences. The query processor searches ....

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of 3rd International Workshop on the Web and Databases (WebDB'00), Dallas, USA, May 2000. 31

....or tree patterns of data to be retrieved. They do not support information retrieval (IR) capabilities, which are extremely valuable in particular for text rich XML sources. This de ciency has been recognized for some time [9] and several IR extensions to XML query languages have been proposed [7, 8, 9, 18]. However, these extensions are limited to simple keyword searches (e.g. 6, 8] and do not support a relevance based ranking of query results. Most importantly, current extensions do not provide fully orthogonal functionality in that precise queries can be arbitrarily mixed with IR style ....

....of structural queries through an IR style index but are limited by the expressiveness of their query languages and their lack in producing a weighted ranking of query results. Other work introduce a weighted ranking of XML document parts by integrating an IR operator into an XML query language [18, 9, 7]. Oracle SQL provides IR capabilities limited to a prede ned, xed set of table columns that do not exhibit any hierarchical structure [3] The semantics and applications of some of the language extensions remain vague. An ecient implementation in form of index structures that could also be used ....

A. Theobald, G. Weikum. Adding relevance to XML. In Proceedings of the 3rd International Workshop on the Web and Databases (WebDB), LNCS 1997.

....implements a standard, static IR approach based on table columns [1] The operator is used within an SQL Where clause. In [6] only exact keyword search within certain XML elements is introduced into XML QL. Earlier related approaches for structured documents can be found in [11] The aim of [13] is an improved Web search by means of an IR enabled XML query language. The similarity operator of [3] is mainly a vehicle to execute fuzzy similarity joins based on atomic values within two DFs. Again, XML QL is used as host language. Based on XQL, 8] proposes an IR style operator that computes ....

A. Theobald and G. Weikum. Adding Relevance to XML. In Proceedings of the 3rd International Workshop on the Web and Databases, LNCS 1997.

....in the next version. When reintroduced, will be given a new EID. If only EIDs are compared, equality test will fail even if the entry for a restaurant is exactly the same in version 10 2 10 3. Introduce a similarity operator in the style of the approach of Theobald and Weikum [17]. There is not any perfect solution to this problem, but we consider a combination of shallow equality and a similarity operator to be the most interesting solution. 5 Discussion Temporal query processing is in general expensive, and in our context we note the following potential performance ....

A. Theobald and G. Weikum. Adding relevance to XML. In WebDB (Informal Proceedings), 2000.

....in the next version. When reintroduced, will be given a new EID. If only EIDs are compared, equality test will fail even if the entry for a restaurant is exactly the same in version V i and V I 2 . 3. Introduce a similarity operator in the style of the approach of Theobald and Weikum [14]. There is not any perfect solution to this problem, but we consider a combination of shallow equality and a similarity operator to be the most interesting solution. 8 Summary We have in this paper described how temporal queries can be executed in an XML database system. We have described issues ....

A. Theobald and G. Weikum. Adding relevance to XML. In WebDB (Informal Proceedings), 2000.

....data and XML in particular have been developed 26 during the last years (see [FSW99, BC00] for surveys) All languages support operators that allow to skip certain parts of the data. However, they require that the user knows which parts have to be skipped. To the best of our knowledge, XXL [TW00] and ELIXIR [CK01] are the only XML query languages stemming from the database community that support result ranking. Both XXL and ELIXIR add a similarity operator to a subset of XML QL [DFF 98] In XXL, the similarity operator can be applied to element names as well as to text sequences. The ....

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of 3rd International Workshop on the Web and Databases (WebDB'00), Dallas, USA, May 2000.

....in the next version. When reintroduced, will be given a new EID. If only EIDs are compared, equality test will fail even if the entry for a restaurant is exactly the same in version and , 0 . 3. Introduce a similarity operator 1 in the style of the approach of Theobald and Weikum [14]. There is not any perfect solution to this problem, but we consider a combination of shallow equality and a similarity operator to be the most interesting solution. 8 Summary We have in this paper described how temporal queries can be executed in an XML database system. We have described ....

A. Theobald and G. Weikum. Adding relevance to XML. In WebDB (Informal Proceedings), 2000.

....order for all three queries. 9 Related Work There have been few research results concerning the ranking of structured queries against XML documents. In [BYRN99] several approaches that combine content and structure are compared. However, none of the models considers relevance ranking. XXL [TW00] and XIRQL [FG00] introduce formalisms loosely based on XML QL and XQL, respectively, that incorporate the notion of relevance. In both cases the relevance of a document to a structured query is determined by combining the weights of the terms in the query leaves according to the query structure. ....

A. Theobald and G. Weikum. Adding relevance to XML. In WebDB'2000 (Informal Proceedings), 2000.

No context found.

A. Theobald and G. Weikum. Adding relevance to XML. In WebDB 2000.

No context found.

A. Theobald and G. Weikum. "Adding Relevance to XML". In Int'l Workshop on the Web and Databases (WebDB), Dallas, TX, May 2000.

No context found.

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of SIGMOD WebDB Workshop, 2000.

No context found.

A. Theobald et al. Adding Relevance to XML. WebDB 2000.

No context found.

A. Theobald and G. Weikum. Adding Relevance to XML. In WebDB 2000.

No context found.

A. Theobald et al. Adding Relevance to XML. WebDB 2000.

No context found.

A. Theobald and G. Weikum. Adding Relevance to XML. In WebDB 2000.

No context found.

A. Theobald and G. Weikum. "Adding Relevance to XML". In Int'l Workshop on the Web and Databases (WebDB), Dallas, TX, May 2000.

No context found.

Anja Theobald and Gerald Weikum. Adding Relevance to XML. In Proceedings of the Third International Workshop on the Web and Databases (WebDB), LNCS 1997.

No context found.

A. Theobald, G. Weikum. Adding Relevance to XML. WebDB'00.

No context found.

A. Theobald, G. Weikum. Adding Relevance to XML. WebDB'00.

No context found.

A. Theobald and G. Weikum. Adding relevance to XML. In Proceedings of SIGMOD WebDB Workshop, 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC