Ian MacLeod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....contexts. As such, a number of formalisms are available to specify a conceptual description. In traditional database design support for the speci3 fication of entity and relationship types are featured. In the world of textual information, context free grammars are a well known formalism. See [GT87, BvdW92, Mac91, ISO86]) The ISO document standard SGML uses context free grammars as the core of its so called document type definitions. Basically, a document type definition is a specification of a set of permitted hierarchically structured documents, or said equivalently in terms of formal language theory: the DTD ....

I.A. MacLeod. A Query Language for Retrieving Information from Hierarchic Text Structures. The Computer Journal, 34(3):254--264, 1991.

....is simpler than, for example, hypertext, even in this simpler case the problem of mixing content and structure is not satisfactorily solved. The models include (in increasing expressiveness order) the Hybrid Model [BY96] PAT Expressions [ST92] Overlapped Lists [CCB95] Lists of References [Mac91] Proximal Nodes [Nav95, NBY95b, NBY97] Parsed Strings [GT87] and Tree Matching [KM93] In most cases, the time complexity of queries has a trade off with the expressiveness of the model. The characteristics of these models are summarized in Table 1, and in this work we focus on the PN model, ....

....[CCB95] list of possibly uniform way, by an inverted express inclusion, union overlapping segments. list metaphor. and combinations. A single hierarchy Text queries can only be Results are flat and from Lists of References with attributes in used to restrict other the same constructor. Can [Mac91] nodes and hypertext queries. express inclusions, links. complex context conditions and set manipulation. A set of disjoint Text is a special view. Can express inclusion, strict trees (views) Text queries are leaves positions, direct and Views can overlap. of query syntax trees. transitive ....

I. MacLeod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....permit the dynamic definition of different structural components in the text, but these, as well as intermediate results, must be disjoint text areas, so one cannot make the union of chapters and sections. Overlapped lists [CCB95] permit overlapping but still not nesting. Lists of References [Mac91] permits nesting but only the top level elements are returned as the result of a query (so the union of chapters and sections returns just the chapters) Sgrep and other similar models [DSDT96, JK96] permit nesting and overlapping in the results, but they restrict the query operations to the ....

I. MacLeod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....of data definition languages. Several models combining grammar based data definition capabilities with operational capabilities on trees have also been introduced; consider, for example, the models of Gonnet and Tompa (1987) Furuta and Stotts (1988) Gyssens, Paredaens and Van Gucht (1989) Macleod (1990, 1991), Christophides (1994) and Blake et al. 1994) In fact, some of these models are hybrid models: text structure is defined by a grammar but the text operations may also be applied to text having no explicit grammar. The p string model (Gonnet Tompa 1987, Blake, Bray Tompa 1992) the model of ....

.... and . A production whose left side is t is called a t production. Example 2.1 A familiar text structure is a collection of papers. Throughout this paper we will use the example grammar shown in Figure 2. 1, where the structure for a paper has been defined following the example used by Macleod (1991). The start symbol of the grammar is Papers. Following Macleod, we suppose that unspecified non terminal symbols in the sample grammar represent word sequences: for each unspecified non terminal t, there is an implicit production t : Word , where the production for Word produces a terminal ....

[Article contains additional citation context not shown here]

Macleod, I.A., A query language for retrieving information from hierarchic text structures, The Computer Journal 34, 3 (June 1991), 254-264.

.... of tree; that generalization too has proved fruitful, underlying the derivations of a number of tree algorithms, such as the parallel pre x algorithm for pre x sums [15, 8] Reingold and Tilford s algorithm for drawing trees tidily [21, 9] and algorithms for query evaluation in structured text [16, 23]. There are two varieties of accumulation on lists: leftwards and rightwards. Leftwards accumulation labels every node of the list with some function of its successors the tail segment starting at that node thereby passing information from right to left along the list; rightwards accumulation ....

Ian A. MacLeod. A query language for retrieving information from hierarchical text structures. Computer Journal, 34(3):254-264, 1991.

....a reasonable strategy of evaluating such queries is first to locate the occurrences of their string components in the database, and then concentrate on the ancestors of these occurrences in the evaluation of the rest of the pattern. This kind of an approach has been taken in the Maestro project [Mac91] where a prototype retrieval tool for hierarchic text structures has been implemented on top of the full text retrieval system Ful Text. Note that the patterns forming a query can be treated in any suitable order. In fact, patterns or pattern instances that do not have any variables in common do ....

I. A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....has an OID whose lifetime is independent from the existence of a corresponding object in main memory. Related work with regard to declarative access. The following issues are of interest with regard to work on declarative access to document collections having originated in the database community [Mac91, C 94, MMM96, QRS 95, B 94, ST94, O 95] ffl With our approach, expressiveness of the query language is achieved by using methods of the database schema, together with OQL query mechanisms [Cat94] Compared to other approaches, the expressive power is higher, while, on the other ....

I.A. Macleod. A Query Language for Retrieving Information from Hierarchic Text Structures. The Computer Journal, 34(3):254--264, 1991.

....Tasks in the publishing process like authoring or editing are done on entire documents but also on smaller parts of them. The usage of SGML for the representation of text components offers the opportunity of not accessing the whole document for modification, but just a certain element of it [94]. This advanced feature requires that the structured document storage provides the functionality for accessing and modifying elements within documents. Furthermore, the document management system must provide some browser or structure view to identify specific elements [88] 3.3.3 Persistency of ....

Macleod, I. A. A Query Language for Retrieving Information from Hierarchic Text Structures. The Computer Journal 34, 3 (1991), 254--264.

....regions, the concept of background regions, and its direct implementation, but another language may be used instead. A variety of languages have been proposed for querying structured text databases, such as Proximal Nodes [19] GClists [5] p strings [8] tree inclusion [13] Maestro [16], and PAT expressions [23] A survey of structured text query languages is found in [3] Sgrep [12] is a variant of grep that uses a structured text query language instead of regular expressions, which helped inspire us to incorporate other Unix style tools into a structured text processing ....

MacLeod, I. "A query language for retrieving information from hierarchic text structures." The Computer Journal, v34 n3, 1991, pp 254264.

....information of documents can be represented 1 , but it is not possible to use this information in a specialized way. Research in recent years shows however that we would like to be able to retrieve documents in various ways based on the structure of the document as a composite object (e.g. [6, 2, 4, 10, 1]) There is a need for querying and retrieval restricted to parts of a document, for access to documents by their structure and for querying which combines structural data and nonstructural data. 1 A structural relation such as the relation between a document and its sections would be ....

MacLeod, I., `A Query Language for Retrieving Information from Hierarchic Text Structures', The Computer Journal, Vol 34(3), pp 254-264, 1991.

....by textual searches or by named regions (like chapters, for example) The idea is to unify both searches by using an extension of inverted lists, where regions and words are indexed the same way. The implementation of this model can be as efficient as that of PAT expressions. Lists of References [14]: is a general model to structure and query textual databases, including also hypertext like linkages, attribute management and external procedures. The structure of documents can be hierarchical (no overlaps) but answers to queries are flat (only the top level elements qualify) and all elements ....

....or only paragraphs) Answers to queries are seen as lists of references (i.e. pointers to the database) This allows to integrate in an elegant way answers to queries to hypertext links, since all are seen as lists of references. This model is very powerful, but hard to implement efficiently [14]. To make the model suitable for comparison, we consider only the portion related to querying structures. Even this portion is quite powerful. Parsed Strings [10] is in fact a data manipulation language. The language used to express database schemas is a context free grammar, that is, the ....

[Article contains additional citation context not shown here]

I. MacLeod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....textual searches or by named regions (like chapters, for example) The idea is to unify both searches by using an extension of inverted lists, where regions and words are indexed in the same way. The implementation of this model can be as efficient as that of PAT expressions. Lists of References [MacLeod 1991]: is a general model to structure and query textual databases, including also hypertext like linkages, attribute management and external procedures. The structure of documents is hierarchical (no overlaps) but answers to queries cannot nest (only the top level elements qualify) and all elements ....

....the models are almost orthogonal. Ours can represent hierarchies but not overlaps, while this one does the inverse. Disregarding the fact that each model cannot represent the most important structure of the other one, most operations can be translated between both models. Lists of References [MacLeod 1991]: our model can completely represent this model (recall that we consider the part of the model related to querying structures) On the other hand, a good portion of our model can be represented in this one, being the most important omissions the unability to have multiple hierarchies and to ....

MacLeod, I. 1991. A query language for retrieving information from hierarchic text structures. The Computer Journal 34, 3, 254--264.

....want to OCR. Therefore, we can begin to look at other aspects of IR to see how they can be made more workable with OCR text. For example, some IR systems have already incorporated the handling of images. Since it has been suggested that document structure can enhance document understanding[13][7][5] our future projects will involve accessing, interpreting, applying, and storing structured information to make it usable within an IR framework. ....

I. A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....have proposed a query language, called HyQ, for the HyTime documents. The HyQ is also a functional query language which is similar to the SGQL. The HyQ provides basic functions for simple queries, and they are combined to generate a complex function. It is also a procedural language. Macleod [20] [21] has defined a query language based on a tree structured document model. His language has used the concept of path in a document hierarchy. In order to make a structure query, users must know the element names and their hierarchy defined in the DTD. Christophides et al. 8] 9] have extended this ....

I. A. Macleod, "A Query Language for Retrieving Information from Hierarchical Text Structures, " The Computer Journal, 34(3), 1991.

....queries, being much more expressive than each mechanism by itself. By using a query language that integrates both types of queries, the retrieval quality of textual databases can be potentiated. However, despite all the work done on efficient and or expressive query languages for structured text [14, 1, 16, 4, 12, 9], the issue of the user s viewpoint has not been studied in detail. In a classic information retrieval system the query process is well defined. The user makes a query, the system answers it and the user evaluates the results. If the user needs are satisfied, the process is finished. Otherwise, ....

I. MacLeod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....contexts. As such, a number of formalisms are available to specify a conceptual description. In traditional database design support for the speci3 fication of entity and relationship types are featured. In the world of textual information, context free grammars are a well known formalism. See [GT87, BvdW92, Mac91, ISO86]) The ISO document standard SGML uses context free grammars as the core of its so called document type definitions. Basically, a document type definition is a specification of a set of permitted hierarchically structured documents, or said equivalently in terms of formal language theory: the DTD ....

I.A. MacLeod. A Query Language for Retrieving Information from Hierarchic Text Structures. The Computer Journal, 34(3):254--264, 1991.

....similar grammar based model. While the languages are fairly simple and well defined, the operators of the languages are very primitive. Queries tend to become long and complicated when expressed in terms of such primitive operators. A query language for manipulating text structures was proposed in [14] by Macleod. The model on which the language is based allows the hierarchical representation of text (using a grammar based representation) as well as non hierarchical linkages between objects (using references) The model and the language are specifically designed for text applications and are ....

Macleod, I. A. A query language for retrieving information from hierarchic text structures. The Computer Journal 34, 3 (1991), 254--264.

.... can be partially supported in schema less databases by dynamic structural summaries as provided for Lore databases through DataGuides [23] 6 Grammars for Modelling Information in Text September 1998 Several models incorporating grammar based data definition capabilities have been proposed [25, 26, 36]. In fact, some of these models are hybrid models: text structure is defined by a grammar but the text operations may also be applied to text having no explicit grammar. Grammar based approaches are also common for text transformation systems [5, 16, 21, 27, 32, 37, 42] The purpose of our model ....

....operators and no parentheses, then the only variant is the right side itself. The example grammar shown in Figure 2.1 represents the text structure for a collection of papers. The start symbol of the grammar is Papers, and the structure for a single paper follows the example used by Macleod [36]. 1) Papers : Paper (2) Paper : Front Body Back (3) Front : Title Author Location Abstract (4) Abstract : Paragraph (5) Paragraph : Sentence (6) Body : Section (7) Section : SectionHeading (Paragraph Paragraph SubSection ) 8) SubSection : SectionHeading Paragraph (9) ....

[Article contains additional citation context not shown here]

I.A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34 (3): 254-264 (1991).

....schemes and present the analytical and experimental results. 1 Introduction Since the Standard Generalized Markup Language (SGML) 13] 15] was standardized, many structured document management systems have been built to manage structured documents including [1] 2] 3] 4] 5] 6] 17] 18] 20] [21] [23] In those systems, structure queries as well as content queries are supported. The content query is based on the content of documents. For example, the query that finds documents which contain a specific keyword is a content query. The structure query is based on the hierarchical logical ....

I. A. Macleod, "A Query Language for Retrieving Information from Hierarchic Text Structures," The Computer Journal, vol. 34, no. 3, pp. 254-264, 1991.

....Therefore, we can begin to look at other aspects of IR to see how they can be made more workable with OCR text. For example, some IR systems have already incorporated the handling of images. Since it has been suggested that document structure can enhance document understanding [Salton 83, Macleod 91, Fuller 93] our future projects will involve accessing, interpreting, applying, and storing structured information to make it usable within an IR framework. References [Croft 94] W. B. Croft, S. Harding, K. Taghva, and J. Borsack. An evaluation of information retrieval accuracy with simulated ....

....pages 204 213, Pittsburgh, PA, June 1993. ACM Press. Harman 93] D. Harman. Overview of the first TREC conference. In Proceedings of the Sixteenth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 36 47, Pittsburgh, PA, June 1993. ACM Press. Macleod 91] I. A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3) 254 264, 1991. Mauldin 89] Michael L. Mauldin. Information Retrieval by Text Skimming. Ph.D. dissertation, School of Computer Science, Carnegie Mellon University, 1989. Ott ....

I. A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

....provides query element independence. 4. 3 Related Work Work in the area of structured text retrieval generally either extends the relational algebra with support for structured text, uses a contextfree grammar to describe text structure and bases queries on this grammar, or attempts both [1, 2, 5, 8, 9, 10, 12, 14]. This heavy dependence on relations and or grammars to describe text structure and formulate queries is in direct contrast to our work. GCL owes some of its intellectual and cultural heritage to two earlier structured text retrieval languages developed at the University of Waterloo. The Pat text ....

I. A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

No context found.

Ian MacLeod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

No context found.

I. MacLeod, `A query language for retrieving information from hierarchic text structures', Computer Journal, 34(32), 254--264, (1991).

No context found.

Macleod, A. (1991). A Query Language for Retrieving Information from Hierarchic Text Structures. The Computer Journal 34(3), pages 254--264.

No context found.

I. A. Macleod. A query language for retrieving information from hierarchic text structures. The Computer Journal, 34(3):254--264, 1991.

First 50 documents

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