KR100902255B1 - AN INSTALLATION FOR URIref COMPRESS IN WEB DOCUMENTS - Google Patents

Abstract

The present invention discloses an apparatus and method for compressing a URL reference to an RDF / MSL document capable of efficiently compressing a web document.

In accordance with another aspect of the present invention, a compression apparatus includes: a URI parser for receiving a web document and extracting data elements from a corresponding file; A URI analysis unit for operating a URI extraction and registration for addition to a new element, including whether to register a base for the URI information used as an attribute value and a reference registration among data extracted from the URI parser; A URI pre-registration unit for storing URI information, URI index information, and short name information to determine whether to register a base of the URI analysis unit; A URI reference dictionary register that stores fragment identifiers and URI index information to determine whether to register a reference of the URI parser; And a compression unit for substituting the web document file input to the URI parser based on the URI index information registered with the URI preregistration unit. Therefore, the present invention can use the minimum storage space when storing the URI reference used as an attribute value in the Web document, thereby saving the storage space when storing the document, and saves network bandwidth when transmitting the document. Has the effect.

RDF / XML, URI, Reference, URIref, Attribute, Base, Index, Network

Description

AN INSTALLATION FOR URIref COMPRESS IN WEB DOCUMENTS}

The present invention relates to a method of compressing a URI reference in a web document. More specifically, the present invention relates to a URL reference for an RDF / MSL document capable of compressing a URI reference used as an attribute value of a web document. To a compression method.

In general, RDF uses Extensible Markup Language (XML) to represent RDF statements in a machine-processable way. XML is designed to allow anyone to design their own document formats and write documents in that format. RDF defines a special XML markup language called RDF / XML to represent RDF information and to exchange information between machines.

The above-described RDF, or Resource Description Framework (RDF), is a language for expressing information about resources in the World Wide Web. RDF is specifically intended to represent metadata about web resources, including the web page's title, author, and date of renewal, including the copyright and contractual information of the web document, or the date of access to some shared resources. However, by generalizing the concept of “web resources,” RDF can also be used to represent information about things identified on the web, even if they are not directly retrieved from the web, for example, information about products that can be purchased on an online shopping site. (E.g., product specifications, prices, inventory information) or preferences of web users for information delivery.

Therefore, in recent years, RDF / XML documents are widely used, and compression is being attempted to reduce network bandwidth due to the increased use of such documents. In particular, in the XML compression method disclosed in the prior art, XML compression techniques are first compressed by using a pre-coding method of each element of the tag, because it is inefficient to process the path expression for the compressed XML data, The inverse arithmetic coding technique transforms the path of each element into distinct intervals on the [0.0, 1.0) region. The path expression is also transformed into an interval on the [0.0, 1.0) region by using an inverse arithmetic coding technique, and the path expression is efficiently processed using a containment relationship between the interval of the path expression and the interval of the elements.

On the other hand, the efficient compression scheme can vary according to the data value type. Among the existing XML compression techniques, XMill defines the type of data value directly, and in XGrind, it is fixed as a string type so that only Hoffman encoding and precoding techniques are used. To compress the data values. Alternatively, apply coding techniques appropriate to the data value according to the inferred data type. For partial data that is inferred as integer or float type, this is the partial data that is generated during the region query processing using a combination of binary encoding and differential encoding. Reduce the burden of restoration

Therefore, by using inverse arithmetic coding and type-dependent compression in the conventional XML data compression, direct and efficient query performance is supported on the compressed XML data. In addition, the type inference engine is used to infer the type for each data value, and the numerical data (integer or real type) is applied to the partial data during the region query by applying a compression technique in which the size relationship of each data value is maintained. The reality is reducing the burden of restoration.

However, such a compression technique reduces disk space burden, query processing burden, and transmission burden incurred in storing, retrieving, and transmitting XML data, and thus will greatly contribute to XML applications such as e-commerce and Internet search. It only induces compression on the data. This causes a problem in that RDF / XML documents which have recently started to be used do not sufficiently compress the documents by the above-described conventional XML data compression. In other words, since it is unable to compress the URI ref, which takes up the largest portion of the RDF / XML document, up to now, not only does it require a lot of memory space for storing the RDF / XML document, but it also wastes network bandwidth during data transmission. It is pointed out that.

The present invention was created to solve the above problems, and an object of the present invention is to minimize the storage space for a web document and to reduce waste of network bandwidth even when the document is transmitted. It provides a compression apparatus and method of reference.

In addition, another object of the present invention is to provide an apparatus and method for compressing a UI reference for a web document that can shorten the transmission time of the web document by compressing a URI reference used as an attribute value of the web document. .

According to a first aspect of the present invention, there is provided a device for compressing a web document for a web document. The apparatus for compressing a web document, comprising: receiving data from the web document and extracting data elements from the file; A URI parser for; A URI analysis unit for operating a URI extraction and registration for addition to a new element, including whether to register a base for the URI information used as an attribute value and reference registration among data extracted from the URI parser; A URI pre-registration unit for storing URI information, URI index information, and short name information to determine whether to register the base of the URI analysis unit; A URI reference dictionary register that stores fragment identifiers and URI index information to determine whether to register a reference of the URI parser; And a compression unit for substituting the web document file input to the URI parser based on the URI index information registered with the URI pre-registration unit.

Specifically, whenever arbitrary URI information is added to the URI dictionary register, a unique integer value is assigned to the index of the URI dictionary, and a frequently used base URI among the URI information arbitrarily defines the index number of the URI dictionary. It is characterized in that assigned to the specified number.

On the other hand, in the method for compressing a web document reference to a web document according to a second aspect of the present invention for achieving the above object, in a method for compressing a web document, a) scanning the web document to be used as an attribute value Determining whether the URI is a base URI, and if the determination result is a base URI, setting the URI index information to a randomly defined specific number; b) as a result of the determination in step a), if the URI used as the attribute value is not a base URI, determining whether there is a corresponding URI information based on a URI dictionary; c) As a result of the determination in step b), if there is matching URI information in the URI dictionary, it is determined whether there is a matching URI reference (URIref) in a URI reference dictionary, but if the URI reference does not exist, step f). Proceeding; d) if it is determined in step c) that the URI reference exists, extract the URI information from the URI reference dictionary and register it in the URI dictionary, extract the URI index information registered in the URI reference dictionary, Replacing the URI information scanned from the web document with the URI index information; e) if the URI information corresponding to the URI dictionary does not exist as determined in step b), registering the URI information in the URI dictionary and generating new URI index information corresponding thereto; And f) registering a new element in advance of a URI reference dictionary using a fragment identifier of the URI information based on the newly registered URI index information, and then feeding back to the step d).

The apparatus and method for compressing a URL reference to a web document according to the present invention saves storage space when storing a document by using a minimum storage space when storing a URI reference used as an attribute value in a web document. In addition, it has the effect of saving the network bandwidth during document transmission.

In addition, the present invention, unlike the existing formatted compression scheme, by generating and registering a URI reference used as an attribute value and compressing the document based on the registered URI reference, not only the compression algorithm is simplified but also the compression efficiency is increased. It has an effect that can be made.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

First, the Uniform Resource Identifier (URI) applied in the present invention is an identifier system that can identify anything, even those that cannot be directly searched on the web. It contains a Uniform Resource Locator (URL), and a URI reference (URIref) can optionally have a fragment identifier, which can be used to express information in a web document, e.g. RDF / XML. Identifies Hereinafter, as an embodiment of the present invention, the RDF / XML document will be collectively referred to as a web document, and the RDF / XML document is a basis for processing metadata, and interoperability between applications exchanging machine understandable information on the web is provided. This is because it provides the most efficient.

As described above, the above-described URI reference may be in various forms. For example, attribute values used in lines 9, 12, 14, and 15, such as an RDF / XML document using the URI reference illustrated in FIG. URI references in shortened form using ': //www.w3.org/wine#'. In other words, the complete URI reference for 'WineGrape'-line 9- is' http://www.w3.org/ wine # WineGrape'. Here, the base URI is used differently depending on the attribute type. For example, 'WineGrape'-line 9- and' # WineGrape'-line 15- are the same URI references, but are expressed in different ways. In addition, the attribute value in line 10 is represented using an object reference, the attribute value in line 13 is represented using a complete URI reference, and so on.

In the end, the URI reference is http: // www. ... / ... has a format, but depending on the type of attribute, the predefined form 'http', 'www', '/', etc. may be deleted, which is an object reference It can be expressed using. Thus, you can define compression for documents based on these URI references.

The present invention will compress a web document, that is, an RDF / XML document, based on such a URI reference. For this purpose, FIG. 1 shows a compression apparatus of an RDF / XML document. As shown, a URI parser 101 for receiving an RDF / XML document and extracting data elements from a corresponding file, and URI information used as attribute values among data extracted from the URI parser 101. URI analysis unit 103 for operating the URI extraction and registration for addition to a new element, including whether the base registration for the reference, reference registration, and URI to determine whether the base registration of the URI analysis unit 103 A URI dictionary register 105 that stores information, URI index information, and short name information, and a URI reference dictionary register 107 that stores fragment identifiers and URI index information to determine whether the URI parser 103 registers a reference. ) And a compression unit 111 for replacing the RDF / XML file input to the URI parser 101 based on the URI index information registered by the URI preregistration unit 105.

The URI dictionary described above is assigned a unique integer value as an index of the URI dictionary each time the URI information is added, and it is appropriate that the base URI is used as a fixed specific number because the base URI is frequently used. Here, the fixed specific number may be any number, character, and the like, including '0', but in the present invention, it is preferable to assign an index to '0'. It is also appropriate to register the namespace prefix declaration portion and base URI in the top-level element of the RDF / XML document with the URI pre-registration 105, where the base URI or namespace prefix URI is the URI of the URI reference. This is because it is likely to be used as.

As described above, the compression scheme configured as described above replaces the URI reference used as the attribute value with a unique integer value, and performs compression on the RDF / XML document due to the replaced information. Since a substantial part of the RDF / XML document is composed of URI references, the actual compression efficiency is high, thereby minimizing storage space when storing the document and reducing network bandwidth when transmitting the document.

Hereinafter, the operation of the present invention will be described in detail with reference to the accompanying drawings. 2 is a flowchart for explaining the main operation of the present invention. 3 to 6 are documents and dictionary contents shown as an embodiment of the present invention to explain a compression process for an RDF / XML document.

First, in step S201, the URI parser 101 scans an RDF / XML document to extract data elements such as tags, attributes, data values, and the like. After receiving the URI information used as an attribute value among the data elements through the URI analysis unit 103, it is determined whether the currently input URI information is base URI information. That is, when an RDF / XML document is received as shown in FIG. 3, it is determined whether to use a base URI by sequentially receiving URI information used as an attribute value of the document.

As shown, the base URI is 'xml: base' and the registration information of the URI dictionary is 'http://www.w3.org/wine#'. This base URI always has index '0' as shown in FIG. Of course, as described above, the base URI index number may be applied to a specific number other than '0'. Therefore, the URI index information illustrated in FIG. 4 is for explaining the exemplary embodiment of the present invention, and of course, a specific number or a specific character in which the URI index information is arbitrarily defined may be used. Thereafter, the index number is 'http://www.w3.org/food#' registered as the index number '1' and the URI dictionary registered as the index number '2' as an embodiment of the URI dictionary information. http://www.w3.org/2002/07/owl# 'may be registered, and the index number' 3 'may also be used as' http://www.w3.org/1999/02/22-rdf-syntax- ns # 'may be registered or' http://www.w3.org/2000/01/rdf-schema# 'may be registered as the index number' 4 '. In addition, 'http://www.kaist.kr/owl#' is registered as the index number '5', and a plurality of URI pre-registrations may be registered with unique numbers.

The URI analyzing unit 103 finds base URI information on the RDF / XML document. When the URI analyzing unit 103 determines that the currently input URI information is the base URI information, the URI analysis unit 103 enters the step S215. That is, when confirming that the URI information is used as the base information, the URI analysis unit 103 enables the compression unit 111 to notify that the URI index is '0'. Therefore, the compression unit 111 replaces the currently input base URI information with the index number '0' and outputs the replacement.

In addition, the URI analyzing unit 103 continuously checks the URI information used as an attribute value, and if it is determined in step S201 that the currently input URI information is not the base URI information, the process proceeds to step S203 and the URI is entered. The analysis unit 103 enables the URI preregistration unit 105. The URI preregistration unit 105 searches whether currently received URI information is already registered. The registration search of the URI dictionary registration unit 105 is searched based on the short name of the URI dictionary, as shown in FIG. 4.

The URI analyzing unit 103 requests the URI dictionary registration unit 105 to search whether the currently scanned URI information is registered in the short name of the URI dictionary. As a result of the search of the URI preregistration unit 105, that is, the result of the determination in step S205, if the currently scanned URI information is registered in the URI preregistration unit 105, the process proceeds to step S207 to check the URI reference dictionary registration unit 107. Enable it. The URI reference dictionary registration unit 107 determines whether URI information previously registered with the URI dictionary registration unit 105 is registered in the URI reference dictionary.

For example, as in step S209, when the URI analysis unit 103 determines that the URI information exists in the URI reference dictionary registration unit 107, the process proceeds to step S215 and the URI index information registered as the URI reference dictionary is entered. Extract. In addition, the URI analysis unit 103 provides the URI index information extracted from the URI reference dictionary registration unit 107 to the compression unit 111, and the compression unit 111 provides a URI index to the currently scanned URI information. Replace with information.

On the other hand, if it is determined in step S209 that there is no UIR reference corresponding to the currently scanned URI information, that is, if there is no corresponding URI reference information in the URI reference dictionary registration unit 107, step S213 is entered. This procedure is a procedure for adding a URI reference when there is no index information on URI information, and is to replace a URI reference with a URI reference index using a URI reference dictionary.

To this end, the URI analysis unit 103 extracts a URI from the URI reference, which examines the URI reference dictionary using the URI index and the fragment identifier retrieved in step S203. Here, the URI reference pre-check using the fragment identifier extracts a URI excluding the fragment identifier when the URI reference includes a fragment identifier, that is, a '#' symbol. In addition, when the aforementioned '#' symbol is not included, a URI is extracted based on the last '/' symbol.

However, if the base URI is used, as described above, 'WineGrape' of the 9th line rdf: ID and '#WineGrape' of the 15th line rdf: resource of FIG. 3 may be checked in consideration of the same URI reference. will be. Through such a procedure, the URI analysis unit 103 extracts the fragment identifier as shown in FIG. 5 and registers it in the URI reference dictionary registration unit 107.

As shown, the URI information for 'WineGrape' or 'Wine', which is frequently used as base URI information, is already registered with '0' as the URI index number, but 'madeFromFruit', which was not registered in the URI reference dictionary, is a URI. In addition to registering a reference, URI index information is newly given. For example, the URI index information of 'madeFromFruit' may be set to '5'.

The URI analyzing unit 103 recognizes the completion of registration in the URI reference dictionary and notifies the compression unit 111 of this. The compression unit 111 performs data substitution based on URI index information of URI information additionally registered as a URI reference dictionary. The compression unit 111 replaces a plurality of URI information used as attribute values with index information of a URI reference through the above procedure. As shown in FIG. 6, the substituted result replaces the URI information for the RDF / XML document with digitized or characterized index information, thereby greatly reducing the capacity of the data. This represents a compression ratio of approximately 50% when comparing the capacity of the RDF / XML document of FIG. 3 described above with the capacity of an RDF / XML document compressed in accordance with the present invention as shown in FIG.

On the other hand, if it is determined in step S205 that the URI information is not registered in the URI dictionary, that is, since there is no URI index information and cannot compress the URI reference, the index information must be additionally registered in the URI dictionary. To this end, the process proceeds to step S211 and the URI analysis unit 103 enables the URI pre-registration unit 105. The URI dictionary registration unit 105 registers newly registered URI information in the above-described URI dictionary, and registers the short name and URI index information for the corresponding URI including the URI information.

The pre-registration method for the URI information is to extract the short name based on the URI information shown in FIG. 4 and register it. For example, when the URI information is 'http://www.w3.org/food#' As described above, when the '#' symbol is included, 'food', which is URI information excluding the fragment identifier, is set as a short name and URI index information thereof is assigned thereto. On the other hand, when the '#' symbol does not exist, for example, 'http://www.w3.org/2002/07/owl', URI information is extracted based on the last '/' symbol. owl-This is set to a short name and given a new URI index.

The URI analysis unit 103 provides the URI information obtained through the above procedure to the URI dictionary registration unit 105. The URI preregistration unit 105 stores newly input URI information and assigns an index number for URI indexing. The index number is unique to the new number, and may be given as a letter, number, alphanumeric string, or the like, including an integer, if necessary. In addition to the URI dictionary registration, the URI analyzing unit 103 also registers URI index information in the URI reference dictionary through step S213.

In addition, the compression unit 111 proceeds to step S215 described above and performs data substitution based on newly allocated URI index information. That is, the data capacity is drastically reduced by replacing URI information having a large data capacity with numeric or characterized URI index information. 3 and 6 show the pre-compression and post-compression states for the RDF / XML document, indicating that the capacity of the compressed document is reduced by near 50% to increase the compression rate.

As described above, the present invention proposes a compression method for an RDF / XML document as a popularized web document, and this compression method significantly reduces the capacity of data, thereby reducing waste of network bandwidth during data transmission. As a result, it is possible to contribute to the construction of a communication infrastructure, and thus, industrial use value is sufficiently given.

1 is a block diagram for explaining the main function of the compression apparatus of the URF reference to the RDF / XM document according to the present invention.

FIG. 2 is a flowchart for explaining a main operation of FIG. 1.

3 is an RDF / XML pre-compression document used for explaining the compression method according to the present invention.

4 illustrates a URI dictionary according to the present invention.

5 is a diagram illustrating a URI reference dictionary according to the present invention.

6 is an RDF / XML document compressed in accordance with the present invention.

<Explanation of symbols for main drawings>

101: URI parser 103: URI analysis unit

105: URI dictionary register 107: URI reference dictionary register

111: URI reference dictionary addition

Claims (8)

An apparatus for compressing a web document, the apparatus comprising: A URI parser for receiving a file corresponding to the web document and extracting data elements from the file; A URI analysis unit for operating a URI extraction and registration for addition to a new element, including whether to register a base for the URI information used as an attribute value and reference registration among data extracted from the URI parser; A URI pre-registration unit for storing URI information, URI index information, and short name information to determine whether to register the base of the URI analysis unit; A URI reference dictionary register that stores fragment identifiers and URI index information to determine whether to register a reference of the URI parser; And And a compression unit for substituting the web document file input to the URI parser based on the URI index information registered by the URI pre-registration unit. The method of claim 1, Each time arbitrary URI information is added to the URI dictionary registration unit, an integer value different from an integer value corresponding to an index previously assigned to the URI information previously stored in the URI dictionary registration unit is assigned to the index of the URI dictionary. Compressor for the UI reference to a web document. The method of claim 2, The base URI, which is frequently used among the URI information, allocates the index number of the URI dictionary to a randomly defined specific number. The method according to any one of claims 1 to 3, And the web document is an RDF / XML document. In the method for compressing a web document, a) scanning the web document to determine whether a URI used as an attribute value is a base URI, and if the determination result is a base URI, setting URI index information to a randomly defined specific number; b) as a result of the determination in step a), if the URI used as the attribute value is not a base URI, determining whether there is a corresponding URI information based on a URI dictionary; c) As a result of the determination in step b), if there is matching URI information in the URI dictionary, it is determined whether there is a matching URI reference (URIref) in a URI reference dictionary, but if the URI reference does not exist, step f). Proceeding; d) if it is determined in step c) that the URI reference exists, extract the URI information from the URI reference dictionary and register it in the URI dictionary, extract the URI index information registered in the URI reference dictionary, Replacing the URI information scanned from the web document with the URI index information; e) if the URI information corresponding to the URI dictionary does not exist as determined in step b), registering the URI information in the URI dictionary and generating new URI index information corresponding thereto; And f) registering a new element in a URI reference dictionary using a fragment identifier of the URI information based on the newly registered URI index information, and then feeding back to step d). How to compress a reference. The method of claim 5, wherein And URI information extracted from the URI reference dictionary of step d) is URI information excluding '#' symbol among fragment identifiers. The method of claim 5, wherein And URI information extracted from the URI reference dictionary of step d) is URI information extracted based on the last '/' symbol of the URI information. The method according to any one of claims 5 to 7, And the web document is an RDF / XML document.

Images