KR100284574B1 - Automatic Object Storage System and How It Works - Google Patents

Abstract

The present invention relates to an automatic object storage system that supports a system for interpreting and verifying multimedia metadata and a method of operating the same. The purpose is to provide an automatic object storage system and its operation method to add necessary information in its standard in order to keep the information found on the Internet. Its configuration includes a metadata document type definer that defines metadata in SGML format, a metadata editor that lets you enter values for metadata elements, metadata document definitions, and metadata checking for grammatical errors in metadata documents. A parser, a metadata analyzer that receives the results of the metadata parser as inputs, and separates each element, a metadata manager that creates each metadata element in the form of an object, structural information about the metadata, information about each metadata element, It consists of a multimedia metadata server that creates a schema to store information about attributes and stores a metadata database, and a system controller that controls and manages all the previous means. In addition, anyone can easily write a metadata application, and the effect is that it can increase the economics and efficiency in metadata authoring.

Description

Automatic object storage system and its operation method

The present invention relates to an automatic object storage system that supports a system for interpreting and verifying multimedia metadata and a method of operating the same.

Recently, the use of the Internet has exploded due to the improvement of computer processing speed and the breakthrough of communication technology.

Currently, on the Internet, World Wide Web (WWW) information in the form of hypermedia document based on HTML (Hyper Text Markup Language), Archie that provides software information, Gopher that systematically organizes information of specific organizations, FTP information that can be used in common, It provides information on various fields in various forms such as WAIS.

The information provided on the Internet is different from many other books and documents printed in the past.

The first is that information providers can be anyone with access to the Internet, so the information on the Internet has become very diverse and extensive.

Second, information on the Internet is being delivered in the form of multimedia information including graphics, audio, video, as well as text and images.

Because of the advantages of such information on the Internet, the greatest interest of Yosai users is focused on how accurately they can get the information they want among a wide variety of internet information.

The technologies used to solve these problems are the various search engines on the Internet. Representative search engines include Inforseek, WebCrawler, Lycos, Alta Vita, and Exite in foreign countries, and dozens of search engines such as Simmani are provided in Korea.

Each search engine uses a robot agent to collect information scattered on the Internet, automatically indexes the contents of the information, and provides a keyword that a user wants to find. After searching, it provides the location information of the required wujdbo and a brief summary of that information.

Index information, location information, and summary information about the contents provided by such search engines are called 'metadata' data about Internet data.

This metadata is usually necessary for us to make full use of information resources, similar to the directory information provided by the file system, the catalog information of the database system, and the data dictionary moved by the CASE tool. It is an indispensable element.

However, various search engines currently used to find the desired information resources on the Internet do not provide accurate metadata, and it is difficult to add multimedia data and necessary metadata to be used as a system for ultimate metadata interpretation. Hard to do

When you need to handle various kinds of metadata like this, you need to easily and automatically save objects in batches. Because metadata document type definitions cover a wide range of information, it is virtually impossible to provide sufficient and accurate metadata for specific information resources.

First, the metadata editor provides a function of inputting values of metadata elements according to a metadata document type definition of a given metadata in standard generalized markup language (SGML) format. At this time, by providing tags to distinguish each element automatically in the form of a template (templet), it should be possible to prevent the error of the user generating the metadata and to provide convenience. Some commercially available SGML editors also provide editing capabilities to create the metadata document type definition itself.

Second, the role of a metadata parser can be broadly divided into two parts: the ability to check for errors in an already defined metadata document type definition, and the content of metadata input through an editor. This function is to check for errors. In most cases, currently available parsers provide both of these features.

Since the metadata parser only checks metadata errors, the metadata parser needs to analyze and separate the elements in order to store the input metadata for each element. The metadata analyzer that performs this function receives the result of the metadata parser as an input, separates each element from it, and passes it to the metadata manager to create objects corresponding to each element.

The metadata manager should be made in the form of an object so that each metadata element analyzed by the metadata analyzer can be stored according to the generated schema. In addition to the creation of such objects, the metadata manager should be responsible for managing changes or deletions of objects by grasping and maintaining the structural relationships among objects for each metadata element. Of course, the function of objectizing, storing and managing the original multimedia data collected from the Internet is also an important part of the role that the metadata manager should play. Multimedia metadata server technology can also be found in relational database systems.

However, since the prior art as described above relies on the independent characteristics of each module, it is possible to add the information required by its own criteria when it is desired to keep the information found on the Internet for future use by each information user. The problem was that it became very difficult to find a way.

The present invention for solving the above problems is to provide an automatic object storage system and its operation method to add the necessary information in its own criteria in order to keep the information found on the Internet.

A feature of the present invention for achieving the above object is a metadata document type definer defining a given metadata in SGML format, and a value of metadata elements according to the metadata document type definition in SGML format of the metadata. A metadata editor, a metadata parser that checks the metadata document type definition and a grammatical error of the metadata document, a metadata analyzer that receives the results of the metadata parser as input, and separates each element, and the metadata A metadata manager which makes each of the metadata elements analyzed by the analyzer into an object form, structural information of the metadata expressed in document form among information to be maintained by the metadata schema, and information of each metadata element A schema so that you can store information about attributes and attributes. A multimedia metadata server for storing a metadata database, and a system for controlling and managing the metadata document type definer, the metadata editor, the metadata parser, the metadata analyzer, the metadata manager, and the multimedia metadata server. It includes a controller. In addition, the metadata editor may provide tags for distinguishing each element in a template form in order to prevent an error of the user who generates the metadata, and to change the metadata document type definition. Interpret this and insert or delete new elements. In addition, the metadata parser stores the parsing result and grammar error, displays the grammar error on the screen, and stops loading the metadata. The metadata manager is to grasp and maintain the structural relationship between objects of each metadata element, to manage the change or deletion of objects, and to store and manage the original multimedia data collected from the Internet.

Another feature of the present invention for achieving the above object is to form a metadata document type definition by modeling the hierarchical data in the form of a complex object, and to view the parts and other contents that can be seen hierarchically on the screen Editing the metadata by dividing it into portions, parsing the metadata by checking the metadata document definition for errors and transitioning the metadata to a state suitable for parsing, and analyzing the metadata; And defining the metadata management method by making the analyzed metadata elements into an object form so that they can be stored according to the generated schema. In addition, the composite object has a property of aggregating multiple data items to create another data, and when storing, clustering can be stored, and there is an IS-PART-OF relationship between upper data and lower data on a layer. Can be treated as one unit in the operation operation of the object. The editing may include applying an editing function to the metadata, determining whether the metadata is suitable for the editing data, generating a metadata document type definition if the determination is appropriate in the determining process, and determining If it is determined that it is not suitable in the process, it is terminated. In the editing step, an internally modified metadata document type definition is verified, but a series of operations are hidden from the user, and an input form is communicated so that a person who has no knowledge of SGML grammar or metadata document type definition can easily write metadata. It is boxed and provides help by using menu or status bar. By applying spatial and connection constraints, it checks for consistency by preventing unnecessary errors. The parsing step includes applying a rule of a metadata parser to the metadata, determining whether the metadata is suitable for metadata parsing, and parsing the parsed data if it is determined to be suitable in the determination process. If it is determined that it is not suitable in the process of generating and the determination process, the process is terminated. The analyzing may include receiving a result of metadata parsing as an input, applying the analysis rule of the metadata to the parsed metadata, determining whether the metadata is suitable for the analysis object, If it is determined that it is appropriate in the determination process, each element of the input metadata is analyzed and separated, a process of generating structured information, and if it is not suitable in the determination process, is terminated. The defining method may further include applying the management rule of the metadata to the structured information, determining whether the structured information is suitable for managing the metadata, and if it is determined to be suitable in the determination process, each element. Generating objects corresponding to the data, storing the object in the multimedia metadata server, and ending the process if it is determined to be unsuitable in the determination process. In addition, in the management method definition step, the structural relationship between objects of each metadata element is grasped and maintained, the change or deletion of the object is managed, and the multimedia data collected from the Internet is stored and managed.

Another feature of the present invention for achieving the above object is a metadata loading step of loading a document by receiving a metadata file name and a file name storing the parsed results as parameters, and develops elements commonly used in document type definitions. A common element object generation step, a meta information object generation step of generating objects for meta information and metadata authors, a primitive information object generation step of generating objects for original information, original authors, co-authors, and document information; Digital information object creation step for creating objects for digital information, media type, media size, and media creator, and for retrieving information, title information, classification information, document content information, image content information, and restriction information. Consists of creating a search information object and generating a metadata instance as a GLO object. The generating of the GLO object may include receiving a file name of the metadata instance as a parameter, storing an instance file in a GLO class, and returning an object name.

1 is a block diagram of an automatic object storage system according to the present invention.

2 is a flow diagram for automatically creating an object in a system in accordance with the present invention.

Hereinafter, with reference to the accompanying drawings will be described in detail one of the preferred embodiments according to the present invention.

1 is a block diagram of an automatic object storage system according to the present invention. Referring to Figure 1, the configuration of an automatic storage system for creating a multimedia metadata database will be described.

The metadata document type definer 1 defines the given metadata in SGML format. In addition, the metadata editor 2 provides a function of inputting values of metadata elements according to the SGML format metadata document type definition of given metadata. At this time, by providing the tags (tags) for distinguishing each element in the form of a template (templet) automatically, it is possible to prevent the error of the user generating the metadata, and provide convenience. In consideration of future scalability, it may include functions for interpreting changes to metadata document type definitions and inserting and deleting new elements.

In object-oriented systems, objects are the basic units that make up a system. Many of these objects have the same characteristics, and classes that represent objects with the same characteristics are called classes. In other words, classes represent objects with common characteristics and behaviors. Each instance is called an object.

A class is a template that defines the commonalities that each object should have. It creates and executes each instance when executing a program. In other words, a template is a template for programming a class.

The metadata parser 3 also provides a metadata document type definition and a function for checking grammatical errors in the metadata document. At this time, the parsing result and the grammar error are stored. If an error occurs, display it on the screen and stop loading metadata. Then, since the metadata parser 3 only checks metadata errors, the metadata parser 4 needs to analyze and separate the elements in order to store the input metadata for each element. . The metadata analyzer 4 which performs such a function receives the result of the metadata parser 3 as an input, separates each element, and transmits the generated elements corresponding to each element to the metadata manager 5. . The metadata manager 5 should be made in the form of an object so that each metadata element analyzed by the metadata analyzer 4 can be stored according to the generated schema. In addition to the creation of such objects, the metadata manager 5 also grasps and maintains the structural relationships between objects for each metadata element, and can even manage the change or deletion of objects. Of course, the metadata manager 5 also functions to objectize, store, and manage the original multimedia data collected from the Internet. In addition, the multimedia metadata server 6 storing the metadata database stores structural information of the metadata, information of each metadata element, and information on attributes, which are expressed in document form among the information to be maintained in the metadata schema. Create a schema to help you. In addition, the system controller 7 includes a metadata definer 1, a metadata editor 2, a metadata parser 3, a metadata analyzer 4, a metadata manager 5, and a multimedia metadata server 6. Control and manage).

The process of automatically creating an object by the object storage system configured as described above will be described.

2 is a flowchart of automatically creating an object in a system according to the present invention. Referring to FIG. 2, a process of automatically creating an object in a system according to the present invention will be described.

First, metadata is recognized (S1), and in S2, an editing function is applied to the metadata (S2). Then, it is determined whether the metadata is suitable for the edited data (S3).

If it is determined that the result of the determination is not suitable, it is terminated. If it is determined that appropriate, S4 generates a metadata document type definition (S4) and applies the rules of the metadata parser to the metadata (S5). It is determined whether the metadata is suitable for metadata parsing (S6).

Here, applying a rule of a metadata parser to metadata means that some metadata is processed according to one rule (metadata parser) defined in parsing some metadata, and the metadata is suitable for metadata parsing. Whether or not a metadata has a DTD format that matches one of the rules set for processing metadata, namely the metadata parser.

If it is determined that the metadata is not suitable for parsing the metadata, the metadata is terminated. If it is determined that the metadata is not suitable, the parsed data is generated (S7), and the analysis rule of the metadata is applied to the parsed data (S8). Subsequently, it is determined whether the metadata is suitable for the analysis object (S9), and when it is determined that the metadata is not suitable, the structure is terminated. ).

It is determined whether the structured information is suitable for metadata management (S12). If it is determined that it is not suitable, the procedure is terminated. If it is determined that appropriate, an object is generated (S13), and the object is stored in the multimedia metadata server (S14).

The automatic storage method for generating a multimedia metadata database according to the present invention is largely divided into five steps. In the first step, a metadata document type definition should be made. By analyzing the metadata instance described according to the metadata document type definition, it can be found that the values of several metadata elements are hierarchically gathered to form one metadata. The complex object form is suitable for modeling such hierarchical data. The composite object developed in the present invention has a characteristic of making another data by aggregating several data items, and when storing, clustering can be stored, IS between the upper data and the lower data on the hierarchy. -PART-OF relationship exists so that it can be treated as a unit in the operation operation of the object.

In the second step, the metadata editing function should be defined. It has a simple structure that contains only the functionality necessary to fully support the metadata document type. Since SGML has a hierarchical structure, it divides elements on the screen into parts that can be seen hierarchically and parts that can be seen. It should be easy to create, delete, and update elements or attributes in consideration of extensibility. Internally, a series of tasks such as verifying changed metadata document type definition are hidden from the user. Input forms are based on dialog boxes and help functions are available through menus or status bars so that anyone who has no knowledge of SGML grammar or metadata document type definitions can easily write metadata.

As a third step, apply metadata parser rules. Check for errors in the metadata document type definition, and switch to a state suitable for metadata parsing. At this point, complete the document type definition.

The fourth step defines the metadata analysis rule. Since the metadata parser only checks the metadata errors, it stores the input metadata for each element and analyzes and separates each element, and receives the result of the metadata parser as an input. Then ask the metadata manager to create objects for each element.

In the fifth step, a metadata management method is defined. Each metadata element analyzed by the metadata analyzer is in the form of an object so that it can be stored according to the generated schema. The metadata manager not only creates such objects, but also maintains the structural relationship between objects of each metadata element, manages the change or deletion of objects, and stores and stores multimedia data collected from the Internet. Can manage functions.

Referring to the metadata loading method developed in the present invention, the metadata is analyzed from the parsed result, and each element in the document type definition is created as an object and loaded into the database. This load can be divided into seven parts. The first is loading metadata, the second is creating objects for common elements, the third is creating objects for meta information, the fourth is creating objects for raw information, and the fifth is digital. The sixth part is to create an object for information, the sixth is to create an object for search information, and the seventh is to create a metadata instance as a general large object (GLO) object. The metadata loading section loads the document with the metadata file name and the file name where the parsed result is stored as parameters. The part that creates objects for common elements has developed elements that are commonly used in document type definition. The part that creates the object for the meta information creates the object for the meta information and the metadata author. The part that creates an object for the original information creates an object for the original information, the original author, the co-author, and the document information. The part that creates the object for the digital information creates the object for the digital information, the type of medium, the size of the medium and the media creator. The part for generating the object for the search information generates the object for the search information, title information, classification information, document content information, image content information and restriction information. The part that creates the metadata instance as a general large object (GLO) object receives the file name of the metadata instance as an argument, stores the instance file in the general large object (GLO) class, and returns the object name. A GLO object is a system defined class.

According to the present invention, an object can be easily stored when it is necessary to process various kinds of metadata, and a system controller can be used to define a metadata document type, a metadata editor, a metadata parser, a metadata analyzer, a metadata manager, and a metadata database. You can write reliable metadata applications that include storage. In addition, it supports a consistent metadata authoring process by authoring the entire application program instead of independent authoring.

In addition, when editing the metadata, tags are automatically provided in the form of templates to distinguish each element for storing the metadata, thereby preventing errors and providing convenience for the user who generates the metadata. have.

This makes it easy for users who do not know how to define SGML tags or metadata document types, metadata editors, metadata parsers, metadata analyzers, metadata managers, and metadata databases, to create metadata applications. And increase efficiency.

Claims (18)

A metadata document type definer defining the given metadata in SGML format; A metadata editor capable of inputting values of metadata elements according to the SGML format metadata document type definition of the metadata; A metadata parser for checking the metadata document type definition and a grammatical error of the metadata document; A metadata analyzer for separating the elements by receiving the results of the metadata parser as inputs; A metadata manager which makes each of the metadata elements analyzed by the metadata analyzer into an object form; Multimedia meta to store a metadata database by creating a schema to store structural information of the metadata, which is represented in a document type among information to be maintained in a metadata schema, and information on the information and attributes of each metadata element. Data server; And And a system controller for controlling and managing the metadata document type definer, the metadata editor, the metadata parser, the metadata analyzer, the metadata manager, and the multimedia metadata server. . The method of claim 1, The metadata editor, in order to prevent the error of the user generating the metadata, the object auto-save system, characterized in that for providing a tag in the form of a template for distinguishing each element. The method according to claim 1 or 2, And the metadata editor interprets the change in the metadata document type definition and inserts or deletes a new element. The method of claim 1, And the metadata parser stores a parsing result and a grammar error. The method according to claim 1 or 4, If the metadata parser finds a grammar error, the metadata parser displays on the screen and stops loading the metadata. The method of claim 1, The metadata manager, Identify and maintain structural relationships between objects of each metadata element; Manage changes or deletions of objects; Automatic object storage system, characterized in that the object of the original multimedia data collected from the Internet to store and manage. Modeling hierarchical data in the form of a composite object to form a metadata document type definition; Editing the metadata by dividing the element into a hierarchical view and other parts of the screen; Parsing said metadata by checking for errors in defining a metadata document type and transitioning said metadata to a state suitable for parsing; Analyzing the metadata; And And defining a metadata management method by forming the object in a form of an object so that the analyzed metadata elements can be stored according to the generated schema. The method of claim 7, wherein The composite object, Several data items are aggregated and have the property of making another data; When storing, clustering can be stored; An IS-PART-OF relationship exists between hierarchical data and lower data on a hierarchy, and thus can be handled as a unit during an operation operation of an object. The method of claim 7, wherein The editing step, Applying an editing function to the metadata; Determining whether the metadata is suitable for edited data; Generating a metadata document type definition if it is determined appropriate in the determination process; And If it is determined that it is not suitable in the determination process, the operation method of the automatic object storage system characterized in that it comprises a process of terminating. The method of claim 7, wherein In the editing step, a method of operating an automatic object storage system, characterized in that the internally modified metadata document type definition is verified but a series of tasks are hidden from the user. The method of claim 7, wherein In the editing step, the object is characterized in that the input form as a dialog box so that even people who have no knowledge of the SGML grammar or metadata document type definition can easily write the metadata and provide help using a menu or a status bar. How to use the auto save system. The method of claim 7, wherein In the editing step, the method of operating an automatic object storage system, characterized in that by applying the spatial constraints and connection constraints to check for consistency by preventing unnecessary errors. The method of claim 7, wherein The parsing step, Applying a rule of a metadata parser to the metadata; Determining whether the metadata is suitable for metadata parsing; Generating parsed data by parsing if it is determined appropriate in the determination process; And If it is determined that it is not suitable in the determination process, the operation method of the automatic object storage system characterized in that it comprises a process of terminating. The method of claim 7, wherein The analysis step, Receiving a result of metadata parsing as an input; Applying the analysis rule of the metadata to the parsed metadata; Determining whether the metadata is suitable for an analysis object; Analyzing and separating the elements of the input metadata when it is determined appropriate in the determination process; Generating structured information; And If it is determined that it is not suitable in the determination process, the operation method of the automatic object storage system characterized in that it comprises a process of terminating. The method of claim 7, wherein The management method definition step, Applying the management rule of the metadata to the structured information; Determining whether the structured information is suitable for metadata management; Generating objects corresponding to each element if it is determined appropriate in the determination process; Storing the object in a multimedia metadata server; And If it is determined that it is not suitable in the determination process, the operation method of the automatic object storage system characterized in that it comprises a process of terminating. The method of claim 7, wherein In the management method definition step, Identify and maintain structural relationships among objects for each metadata element; Manage changes or deletions of objects; Method of operating an automatic object storage system, characterized in that the object to store and manage the multimedia data collected from the Internet. A metadata loading step of loading a document by receiving the metadata file name and the file name storing the parsed result as parameters; Creating a common element object for developing elements commonly used in the document type definition; A meta-information object generation step of generating objects for meta-information and metadata author; Primitive information object generation step of generating objects for original information, original author, co-author and bibliographic information; A digital information object creation step of generating objects for the digital information, the type of the medium, the size of the medium and the creator of the medium; A search information object generation step of generating an object for search information, title information, classification information, document content information, image content information, and restriction information; And A method of loading a database of objects, comprising the steps of creating a metadata instance as a GLO object. The method of claim 17, Creating the GLO object, Receiving a file name of the metadata instance as a parameter; Storing the instance file in a GLO class; And A method for loading a database of an object, comprising the step of returning an object name.

Images