KR20020096519A - System for collecting information based on internet - Google Patents

Abstract

If you specify items as HTML elements that indicate the information you want from the web server of the information server through the information gathering agent provided to each client, the contents of these items are displayed to the user collectively and at the same time Performs a browser function and continuously updates the contents of the specified items, while continuously storing the updated contents of the items. Moreover, the management server collects user-specified item information via the Internet, maintains / manages this information through a database, updates items designated by the administrator, and transmits the contents directly to the client information collecting agent.

Therefore, the user is provided with the convenience of being able to collectively view the contents continuously updated without any effort by selecting only the information he / she wants from various web pages on the Internet, and the administrator to the information of interest to the user. Keep accurate data about your business to help you develop your marketing or site operations.

Description

System for collecting information based on internet}

The present invention relates to an information collection system on the Internet, and more particularly, to an information collection system configured to selectively and update only information desired by a user from respective information providing servers connected to the Internet.

In general, when a user wants to collect information through the Internet, the user accesses a server providing the information and obtains the information through a search. For example, a user who mainly deals with day trading regularly visits news service sites operated by various media to check headline articles and spot news. However, it is quite difficult to repeat such information collection every few minutes or seconds, and the more the information site for the information collection, the more. Moreover, it is practically impossible if such work must be done at the same time.

Therefore, it is necessary for the user's convenience to provide only the information of interest to users in real time from a plurality of information providing servers connected to the Internet at the same time, and to continuously update by a predetermined time unit.

However, some prerequisites must be made in order to implement these functions.

First, a user interface that can easily set which items in the information providing server corresponding to the information the user wants to obtain, and how many cycles are required to check whether the corresponding items are updated should be provided.

Second, the web page of the information providing server is composed of HTML language, and HTML has a structure of multiple trees with elements (nodes) as nodes, and remembers the position in the multiple trees for the items specified by the user. We need an algorithm to make and find. That is, in the case of a web page where the content of HTML is continuously updated, the shape of this multi-tree may change. For example, to find a particular leaf on a specific branch of a tree, remember the features and paths of all the branches from the root to the leaf, regardless of the deformation, addition or deletion of the other branches. I can make it.

Third, in order to analyze web pages of multiple information providing servers at the same time, an algorithm is needed to quickly make HTML documents into multiple trees and access each node. In other words, create a multi-tree form where each tag is based on nodes so that you can programmatically access the general HTML document consisting of tags, and make it possible to know the characteristics and contents of each node regardless of the grammar of the original HTML document. There is a Document Object Model (DOM) that allows you to make changes and additions. Most of the DOM is designed around HTML output and a user interface on a web browser, and it takes a relatively long time to create multiple trees for many functions.

Fourth, there is a need for a process of continuously updating information of items designated by a user, and a storage function for preserving updated information so that all contents from a predetermined time point to the present can be collectively understood.

On the other hand, identifying and analyzing Internet users' dispositions and areas of interest is essential to company operations, but it is actually quite difficult and expensive.

However, if the user is provided with the functions described above, and receives data on the item of interest set by the user, the user's propensity or interests can be analyzed efficiently and at a low cost based on the data, and the actual error rate. This lower quality result can be obtained.

Moreover, if the above functions are performed entirely on the client side, the functions are more likely to be restricted by the user's Internet infrastructure such as Internet connection speed or computer processing speed. By collecting information through the user and transmitting the information to the user, the user can be provided with high quality information beyond his or her Internet environment.

SUMMARY OF THE INVENTION An object of the present invention is to provide an information collection system that allows a user to selectively and simultaneously obtain only information of interest from a plurality of information providing servers connected through the Internet, and continuously updates the data by a predetermined time unit.

Another object of the present invention is to receive data on items collected and set by the users themselves, and to efficiently analyze the propensity and interest of each user based on the data at a low cost, thereby obtaining a high quality result with low error rate. Instead, it provides an information collection system that collects and provides information.

Other objects and features of the present invention will become more apparent through the preferred embodiments described below.

1 is a block diagram of the Internet to which the present invention is applied;

2 is a block diagram showing the configuration of a client information gathering agent according to the present invention;

3 is a block diagram showing the configuration of a server system of the present invention;

4 illustrates a process of specifying an HTML element from a web page of an information providing server,

FIG. 5 illustrates browsing the item specified from FIG. 4 as an item, FIG.

6 is a flowchart illustrating a process of generating an item by the item generating module according to the present invention;

7 is a block diagram illustrating a tree structure of an HTML document of a specific web page.

8 is a flowchart showing a method for collecting information from a client according to the present invention.

According to one aspect of the present invention, a plurality of clients, a management server that registers and manages the clients as members, and a plurality of information providing servers are interconnected via the Internet, and each of the clients needs information required by the user. An information collection agent is built in to collect only periodically from the information providing servers. The information collection agent performs a general web browser function and allows the user to specify HTML elements, which are the unit information of the web page of the information providing server. It provides the ability to access the multi-tree DOM (Document Object Model) mechanism for HTML documents, the web browser control module that shows the contents of items corresponding to user-specified elements, and the multi-tree structure of HTML documents. HTML tree generation module for quickly and efficiently An item generating module that makes a specified element a single item, an item checking module that checks whether an item has a valid value in multiple trees created by the HTML tree generating module, and when the contents of the element pointed to by the item is updated With the item storage module and item list, which keeps updated contents in the database, add or remove items from the list, download the HTML of the item to be checked through the timer through the internet, save and load the item list, Disclosed is an information gathering system on the Internet including a client item manager for managing transmission to a management server and production of content to be shown to a user.

Preferably, the management server is a network input and output control module for direct information input and output with the client information gathering agent through the Internet, a server item storage module for storing the item information transmitted from the client to the DB server, and the HTML document of Server HTML tree generation module that generates multi-tree structure quickly and efficiently, and item checking module and management server to check whether one item has valid value in multiple tree created by server HTML tree generation module. It includes a server item manager that takes a list of files, downloads an HTML document of items to be inspected from the Internet, and manages production of content to be sent to a client.

The item of the present invention stores the location information in the web page, the URL of the web page to which the corresponding element is assigned, the text of the element and the inquiry period of the item.

Preferably, the location information of the item includes information on the relationship between the node feature and the parent node from the node corresponding to the HTML element of the item to the top node of the tree structure of the web page.

According to another aspect of the present invention, a plurality of clients, a management server that registers and manages clients as members, and a plurality of information providing servers are interconnected through the Internet, and each of the clients has only information required by the user. Applied to an information gathering system on the Internet in which an information gathering agent for collecting information periodically and periodically from the information providing servers is embedded, and a user designates an HTML element selected from a specific web page of the information providing server as an item. Generating the tree structure of the web page through the agent's HTML tree generation module, obtaining the information of the HTML element specified by the user through the dynamic HTML object model, and then creating the target element node corresponding to the HTML element. Steps found in the structure, tree constructs from matched nodes Storing information about the feature of the node and its relationship with the parent node up to the top node of the item; and storing the URL of the web page to which the HTML element is assigned, the text of the element, and the lookup period of the item in the item. Disclosed is a method of collecting information on an image.

In addition, by transmitting the item information of the client to the management server to distinguish the items to receive information from the management server, to determine whether there is an item to be inspected in the client, if present, the HTML string of the URL specified in the item to the Internet After downloading through the HTML tree generation module and generating the tree structure of the downloaded HTML, checking the item validity with the item inspection module of the information collecting agent, and determining whether the element is invalidated, Outputting and notifying, and if it is not invalidated, determines whether it has been changed, and if it is changed, outputting the updated item contents to the user and determining whether there is item information received from the management server, and if there is item information, updating the corresponding item. Printed content to the user It may further comprise the step.

In describing the present invention, terms used in advance are defined as follows.

In the expression that a user specifies a particular item in a web page on a web browser, an item, a tag in an HTML document, an element in a DOM mechanism, a node in multiple trees, The element nodes and the like in the HTML tree generation module described below are all slightly different but generally have the same meaning.

In addition, as shown in Figure 4, the user selects the element 30 in a particular web page, as shown in Figure 5, as shown in his item 50, is continuously updated for that information It means that you are informed.

That is, the item used in the present invention refers to an object including information desired by a user, and includes information about an element itself or position information of an element in a multi-tree of a corresponding web page.

1 shows an internet configuration to which the present invention is applied.

When the plurality of clients 100 includes the information collecting agent 150 according to the present invention, each of the information providing servers 300 and the information collecting agent are provided through the Internet 200 and the DB server 450 is connected. It is connected to the management server 400.

The client 100 may include a computer system or a personal portable terminal device capable of general web browsing. The user may be registered as a member of the management server 400 to receive an information gathering agent or to provide other services.

The information gathering agent 150 may be a program for a general Internet user who uses Windows as an OS or an OS-based program applied to a wireless PDA or an IMT-2000 terminal device.

2, a block diagram illustrating a client information gathering agent in accordance with the present invention is shown.

As shown, the client information collection agent according to the present invention performs a general web browser function and at the same time provides the ability to specify elements in the web page of the information providing server and to access the multi- DOM type DOM mechanism for HTML documents. Web browser control module that can provide the contents of the items specified by the user.

153, an HTML tree generation module 155 for quickly and efficiently generating a multi-tree structure of an HTML document, an item generation module 156 for making a user-specified element 30 into one item 50, The item checking module 154 that checks whether one item 50 has a valid value in the multiple trees created by the HTML tree generating module 155, and the updated content when the content of the element pointed to by the item is updated. Add or delete items from the list with the item storage module 157 and the item list to keep them in the database 160, and download the HTML of the item to be checked through the timer over the Internet, The client item manager 152 manages load, transfer to the management server 400, production of content to be shown to the user.

The web browser control module 153 includes two components, one of which includes a web browser control and an HTML component.

The web browser controller functions to present information directly to the user and basically provide an interface for accepting the user's request through a mouse or keyboard.

To enable one application to surf the web by default, you need a web browser such as Internet Explorer or Netscape, which allows the user to visually understand what the HTML document is and allow you to specify the URL you want. In the present invention, the web browser control, which is one of the ActiveX controls provided by default in Microsoft's windows, is used to provide essential web browser functions.

When you run a browser, such as Internet Explorer, the web browser control is the part of the window that displays the web document except the menu or toolbar. In fact, when you run Internet Explorer, a very small program called Iexplore.exe is launched, which invokes the web browser control for most functions, and most of the functions are controlled by the web browser. In other words, using the web browser control means that most of the functions appearing in a browser such as Internet Explorer are possible, which is why the web browser control is used in the present invention.

In addition to the basic web browser functionality, the web browser control also provides the ability to handle events that occur when an HTML document is loaded.

The HTML component is one of the most basic components of Microsoft's Internet technology. It can be hosted automatically by the web browser ActiveX control, or it can be hosted through a common communication interface because it has Active Document server interfaces. In the present invention, the information collection agent may be included.

HTML components host objects that can be used by HTML scripts such as the ActiveX Script Engine, Java Applets, Plug-Ins, etc. It provides three basic functions: building and documenting a Document Object Model (Dynamic HTML Object Model) and rendering routines into Web browser controls.

By hosting HTML components, in addition to the above three functions, some additional controls are provided in the form of functions, including standard connection point interfaces, IDocHostUIHandler Interface, and ICustomDoc Interface.

(1) dynamic HTML object model interface

When a user designates an element of one of the web pages of a web browser as an item, in order to know which tag in the original HTML document specifically corresponds to the tag, the interface with the dynamic HTML object model must be connected. Information such as tag name, tag property, object included in tag, and text included in tag is obtained or modified.

(2) ICustomDoc interface

Basically, when you want to designate an element as a pinpoint item in a web browser, you can select the desired element with the mouse and specify it as an item through a menu outside of the web browser. The ICustomDoc interface allows you to customize the basic UI handlers used by the web browser controls.

(3) IDocHostUIHandler

With the defined name of the user interface handler just described above, the present invention provides an independent user interface handler. That is, as illustrated in FIG. 4, when the right mouse button is pressed in the web browser, the context window menu for the original web browser is modified to be designated as an item. The new user interface handler applied to the present invention is set through the above IcustomDoc interface.

(4) standard connection point interfaces

This method is used when accessing multi-tree DOM mechanism (function as UI-less HTML parser) by loading HTML document without execution by using HTML component without web browser control, but it is stable in terms of speed and system allocation. There are disadvantages.

As described above, when a user designates a specific element in a specific web page, the user wants the content of the designated element to be continuously updated automatically, and a unit for collecting information necessary for this is called an item. 50 in FIG. 2 is an example in which an item is represented to the user. The bold text in the first line indicates the latest updated content of an element designated as 30 in FIG. 1, and the braces in the second line indicate the element. Indicates the title of the specified web page, and the content of the last third line shows information such as the interval of the search for the element or the time of the latest update.

The item basically stores the web page address of the specified element and the location information in the multi-tree structure of the corresponding HTML that allows you to locate the element on this web page. An HTML document has a multi-tree structure with HTML tags as one tree node, and when interpreted in DOM format, the units of the nodes are called elements instead of tags. In the present invention, when referring to the multi-tree structure to obtain the location information, the tree structure generated by the HTML tree generation module, which will be described later, rather than the form of the DOM, is used. The form and the tree structure of the DOM are the same. The node will use the name element node.

A process of generating an item by the item generating module 156 will be described with reference to FIG. 6.

First, as shown in FIG. 4, a user is asked to designate an HTML element selected on a specific web page as an item through the menu format 32 of the user interface handler of the present invention (step S10).

Subsequently, the tree structure of the web page is generated through the HTML tree generation module 155 (step S20), information of the HTML element designated by the user is obtained through the dynamic HTML object model, and then the target element corresponding to the HTML element is obtained. The node is found in the generated tree structure (step S30).

From the matched node to the top node of the tree structure, information on the relationship between the corresponding node feature and the parent node is stored in the item (step S40).

Here, a simple example of the location information will be described before explaining the location information storage method to be used in the present invention. Assume that an HTML document forms a multi-tree like the one in Figure 7, and that every node has a tag name, a size, and a background color (the characteristics of a tag are much more complicated than that). At this time, the position information of node 15 may be used as the "15th node". However, this method invalidates the position information of the original node 15 even if only one child node of node 4 is added. In other words, the target node cannot be found even with a small change of node 4 irrelevant to the original target node Node15. In order to avoid these disadvantages and to obtain efficient location information, the present invention stores the location information in the following manner.

(1) Store information of each node continuously from the target node to the highest node.

As above, the location information of the node 15 is stored, for example, in the order of node 15-node 6-node 3-node 1.

(2) Define whether each feature is valid according to tag type and save it.

This is used as a criterion to determine if the HTML document has changed, compared to before it has changed. For example, assuming that the color and width of a table in an HTML document has changed, it is used to determine whether or not it will actually change. Of course, it may or may not have changed. As more valid features are defined, smaller changes can be noticed, but the HTML provider determines that they have not changed the actual meaning. For example, in FIG. 7, when the TABLE name is a tag name, if the background color feature is defined as valid, node 4 and node 5 have different forms, and if it is defined as invalid, node 4 and node 5 have the same form. Judging.

(3) The number of the child nodes of the parent node, such as its own type, is stored to determine the location of itself.

After storing the location information in this manner, the URL of the web page to which the element is designated, the text of the element, the inquiry period of the item, and the like are stored in the item (step S50) to complete one item.

On the other hand, the item includes the location information on the HTML document of the user-specified element, and for this purpose, the multi-tree structure of the HTML document is required.

Unlike other executables written in machine language, HTML documents are scripting languages written in plain text mode that are converted into a computer-readable form when run through a browser.

When HTML is run in a browser, if you want to directly control through the HTML document to get the characteristics or change the content of something that is shown to the user, it becomes very difficult because HTML is just a sequence of textual information. So a mechanism has been devised that allows you to use these features programmatically, which is called the Document Object Model (DOM) mechanism.

In general, the most common browsers are Internet Explorer and Netscape, and these browsers provide their own DOM services, with the two DOM types being mostly identical but some different. Objects that can be accessed through the DOM are called elements, which correspond to the tags in an HTML document, and all elements exist in a single, multi-tree structure, and each DOM can get or transform information about each Element. .

For example, using a web browser used by Internet Explorer, you can access elements through a DOM called the dynamic HTML object model, which you can use to create items, but the DOM provided by the web browser provides a rich feature set. Instead, it is not suitable for the characteristics of the present invention, which creates and updates many items due to the drawback of using a lot of resources of the system and being slow.

Therefore, the present invention applies the HTML tree generation module that can perform only the functions required for the present invention among the functions of the general DOM, but uses the system resources that are quick to generate and execute and use relatively little system resources.

The function of each node of the tree structure generated by the HTML tree generation module of the present invention is the same as the element function in the general DOM, but in the process of generating the tree, the node means one tag of the HTML document. For example, when there is an HTML sentence <A href=xxx> AAA </A>, the original node means the whole sentence, but the HTML tree generation module of the present invention uses <A href=xxx> and </A>. After the two nodes are created and interpreted in their original meaning, the node that ends the tag, such as </A>, is excluded from the tree structure. This node is defined as the termination node.

The features of the HTML tree generation module of the present invention are as follows.

(1) Create a tree structure without parsing HTML documents.

That is, the tree structure is generated quickly because the tree structure is created by finding only the start (<) and end (>) characters of the tag in the HTML document.

The DOM of a generic browser is parsed because all HTML content is parsed for the purpose of showing all the content to the user, but the present invention uses this feature to speed up the creation because the tree structure is primarily needed.

(2) The configuration of the nodes forming the tree structure is simple.

A node is a tag start position where a tag corresponding to one node starts in an HTML document, a tag end position where a tag corresponding to one node ends in an HTML document, an end node position and tree which are positions of an end node for a node. In the structure, it has only the parent node position which is the position of the parent node of one node. Thus, this simple structure contributes to an improvement in tree generation speed and memory usage.

(3) Parse only the contents of the node associated with the item.

Since only the destination node specified in the item is used in the present invention, only the content of the node related to the destination node needs to be analyzed.

Referring to the process of generating a tree by the HTML tree generation module of the present invention.

When the generation process starts, we do the following with values from the first character in the HTML document to the last: The current value refers to the text of the HTML document currently being inspected. The current parent node is used to specify the position of the parent node of the node. The initial value is 0.

(1) If the current value is "<", specify the current value position at the tag start position of the current node.

(2) If the current value is ">", the position of current value is designated at tag end position of current node, and if current node is end node, if current node is end node, current node is paired with current node. Find the parent node position of the parent node by reference and assign the current node to the end node position of the found node, specify the parent node position of the node found in the current parent node, and if the current node is not the end node, the parent node of the current node Create a new node by specifying the current parent node in the location, make it the current node,

(3) Move the current value to the next character of HTML.

In this way, a node list is generated.

The method of interpreting the generated node list is to use the information of the tag corresponding to one node, and the tag end position is the contents of the tag from the node start position. There is also a way to refer to a child node, where the first child node of a node is next to that node in the node list, and the second is after the end node position of the previous child node in the node list. In this way, you can find the end node position of the node. In addition, there is a method of obtaining the contents of a node, and the portion of the contents of the node from the end tag position of one node to the start tag position of the node corresponding to the end node position is the content of the node.

The created item is periodically checked for validity according to the inspection interval recorded in the item information, and if it is valid, it is checked whether the content of the element specified in the item in the current HTML of the web page recorded in the current item is updated. 154 does this.

The validity of an item means that there is a web page with a stored URL and that there is a node that matches the destination node specified in the multi-tree structure of that web page.

The method of finding the destination node finds nodes from the top node to the destination node that match the tag name and valid features in the multi-tree of the new web page based on the tag names and valid features of the nodes consecutively stored by the item generating module. will be.

If the item is valid, the text information is taken from a node in the tree structure corresponding to the object node of the item, and compared with the text information in the item, it is determined that the item is updated.

The client item manager 152 properly inspects the items specified by the user and is responsible for outputting the results. Some of the designated item list are provided with updated contents through the management server 400. The item manager 152 is also responsible for processing information input and output with the management server.

Hereinafter, a method of collecting client information will be described with reference to FIG. 8.

When the client item manager is driven, the item information of the client is transmitted to the management server 400 to distinguish items that receive information from the management server 400 (step S100).

It is determined whether the item to be inspected exists in the client 100 (step S110), and if the item exists, the HTML string of the URL specified for the item is downloaded through the Internet (step S120), and downloaded through the HTML tree generation module. After generating the tree structure of the received HTML (step S130), the item inspection module checks whether the item is valid (step S140).

It is then determined whether the element has been invalidated (step S150), and outputs and notifies the user that the invalidation has been made (step S155). (Step S165).

In addition, it is determined whether there is item information received from the management server (step S170), and if there is item information, the updated contents of the corresponding item are output to the user (step S175).

It is determined whether there is an end request (step S180). If there is no end request, the above steps are repeated. If there is an end request, the process server is notified of the end of the process (step S190).

The most visually and functionally focused part of the client information gathering agent is that it is configured in a form similar to Internet Explorer so that users can easily learn how to operate the agent. To do this, in addition to visually using Microsoft's web browser controls as a browser, it provides toolbar interfaces such as forward, backward, stop, and refresh, as shown in Figure 4. The client item manager process, which handles most of the functions, runs as a background process and hides it. On the surface, it mainly deals with the functions of a normal web browser.

In addition, when a user designates an element in a web page according to the purpose of the client information gathering agent, it may be implemented in the form of an open service or a closed service by restricting the web page that can be specified.

An open service is an unlimited form of web page that can be specified, and moves directly to another web page without restriction by moving the web page through the input window that receives the web page address from the user or by clicking the anchor tag within the web page. Say the format. That means the same function as a normal browser.

Restrictive service means that the operator can designate a user-specified web page without input box, so that only the designated page can be specified. When moving through anchor tag in the web page, the address of the page to be moved through the event provided by the web browser control After getting, we can compare this address with pages designated by the operator. 4 and 5 show an example of the implementation of the limited service, in which the part indicated as the channel is the web page addresses provided by the operator, and when clicked, a list of addresses corresponding to the channel is shown.

By providing a limited service rather than an open service that functions as a general browser, the user has a function that can serve as a guide to the valid ones among a large number of websites according to the purpose of the information gathering agent, and the operator can specify the desired purpose. To make it possible. For example, in your website, you can collect and view the information you want, and you want to know what users prefer from your website. Only web pages are specified.

3, the configuration of the management server system 400 applied to the present invention is shown.

The management server system 400 stores a network input / output control module 403 which is responsible for direct information input / output with the client information collecting agent 150 through a network, and the item information transmitted from the clients in the DB server 450. The server item storage module 406, the server HTML tree generation module 405 for quickly and efficiently generating the multi-tree structure of the HTML document as in the client information collection agent, the server item inspection module 404 and the management server directly The server item manager 402 is configured to download an HTML document of an item to be double-checked through the Internet with a list of items to be inspected and to manage production of content to be sent to a client.

The management server system 400 has two functions. First, it provides information useful to a server operator by accumulating information of a specified item from a client, and second, updates information of an item in a client information collecting agent. It is augmented to provide better service.

In the first function, when a client gathering agent is distributed to a user, the user informs the server of information about his or her identity, and when the agent is started or shut down, items are added or deleted by the user. When the information of the item is notified to the server, the server accumulates this information on the DB server. Obtain information such as gender, location, age, and taste from data about your personal information, and from information about items delivered in real time, the time, website, web page, element of the web page, and click rate It is an anchor element, and on the screen showing the updated contents of the item to the user, when the user clicks on the contents of the item, if the function of displaying the connected anchor is built), the operator has a low error rate and obtains very useful information.

For example of the information that can be obtained

-The most viewed pages of a specific website by user time zone

-The most enjoyed webpage among users of Gyeonggi-do

-6 months of user fluctuations in preferences across the Internet.

In the second function, the utilization of the client information gathering agent is better as the number of items designated, the shorter the inquiry period of the designated item is, but is limited by the network speed owned by the user, and the user is in addition to the information gathering agent of the present invention. Consider that other applications related to the Internet can be used at the same time. In particular, the present invention downloads an HTML document of a web page designated by a user, and the size of the document is relatively large as about 1K bytes at about 30K except for an image. For example, assuming that the web page being downloaded is 60K bytes, and the user's internet speed is only 2% of 2M bits per second, which is the speed of a private home line, only 50% is used in consideration of other internet related applications, this means that the item has a 5 second view period. This means that up to 10 can be specified. It is difficult to say that this result is sufficient in terms of inquiry period or the number that can be specified. In order to reinforce this restriction, the management server system directly inspects some of the registered items and performs a service that sends only the results when updated to the client. The network speed of the server system and the system can be used as needed by the commercial user, and many clients can receive new information by inspecting one item.

The server item manager 402 in the management server system 400 properly inspects the items provided from the client information gathering agent and is responsible for outputting these results.

Roughly select the item to be inspected-> download the web page-> create the HTML tree-> inspect the item-> the results of the server item manager in the order of the output of the result is almost the same as the operation of the client item manager, There is a difference.

First, when outputting the inspection result of the item, the client item manager outputs the output content to the network input / output control module 403 as compared with the output on the monitor screen for the user to check.

In addition, in the information included in the item, the server system includes the total number of clients referencing the item and the information of the clients in addition to the information of the original item, which refers to the address when sending the updated contents of the item to the clients. To do this.

In addition, in selecting an item to be inspected, the client item manager selects the item to be inspected based on the inquiry period specified for the item, but the management server selects the item in the order specified by the client.

Meanwhile, the item checking module and the HTML tree generating module used in the server item manager should be the same as the modules used in the client.

As described above, one of the server functions of the present invention is to improve the performance of the client information gathering agent by performing some of the inspection of items to be performed by the client gathering agent on the server instead.

The server system delivers the information of specified items from multiple clients. It is not an efficient way to simply select and inspect these items in turn without any special rules. The server system of the present invention uses a method of selecting and inspecting only a certain number of items in order of the items most specified by the user for efficiently supporting the client.

The number of items to be inspected and the lookup cycle can be adjusted to the operator's system or purpose, for example, the number of items to be inspected may simply be a constant value, and the selected sum of the items to be inspected is a percentage of the total number of items. Until the occupancy may be a reference, the inquiry period of the item to be inspected may be shorter as the number of times selected, or may be a constant value.

The server system has several additional elements related to the selection of items for this method: an item list, a user list and a selected list. The item list holds information of all items delivered from the client. One node of the list contains the item's information, the number of users who assigned the item, and the location information in the user list for those users. It is arranged to add, delete, and find easily. The user list is information about users who have sent items to the server. A node contains information of the user's address, the number of items selected by the user, and the location in the item list. Also, the selected list is actually a pointer list of items that the server item manager should inspect. Each node in the item list records whether the node is in the selected list.

Hereinafter, a description will be given of how new items are registered in the server system or specific items are removed from the client.

(1) When a new item is registered from the client

Get the node corresponding to the new item from the item list. If not, create a new node and add it to it.

Get the node corresponding to the user who sent the new item from the user list. If not, create a new node and add it to it.

-Increase the number of users specified in the corresponding node of the item list and the number of items held in the corresponding node of the user list, and then record the position of each other

-Check whether the node of the item list can be added to the selected list and record the information that it is in the selected list if added. If there is a node that is removed from the selected list due to this addition, the information that is in the selected list of this node is removed.

(2) when a particular item is removed from the client

-Get the node corresponding to the item to be removed from the item list and make it the node.

-Get the node holding the item to be removed from the user list and call it the node.

-Reduce the number of users specified in the corresponding node of the item list and the number of items held in the corresponding node of the user list, and then remove the recorded positions of each other.

If the number of users specified in the node of the item list is 0, the node is removed. If the number of items retained in the node of the user list is 0, the node is removed.

According to the present invention, when an HTML element indicating information desired by a user is designated as an item from a web page of an information providing server through an information collection agent provided to each client, the contents of these items are collectively shown to the user. At the same time, it performs a normal web browser function, and continuously updates the contents of designated items, and continuously stores updated contents of the items.

Therefore, the user is provided with the convenience of being able to collectively view the contents continuously updated without any effort by selecting only the desired information from the various web pages of the Internet.

Moreover, the management server collects user-specified item information via the Internet, maintains / manages this information through a database, updates items designated by the administrator, and transmits the contents directly to the client information collecting agent.

As a result, the administrator can have accurate data about the information of interest to the user to assist in the development of marketing and site operations.

Claims (6)

A plurality of clients, a management server that registers and manages the clients as members, and a plurality of information providing servers are interconnected through the Internet, and each of the clients includes only information required by the user from the information providing servers. Selective and periodic information collection agent for collecting is built in, the information collection agent In addition to performing general web browser functions, the user can specify HTML elements, which are the unit information of the web page of the information providing server, and access a multi-tree DOM (Document Object Model) mechanism for HTML documents. A web browser control module for providing contents of items corresponding to an element designated by a user; An HTML tree generation module for quickly and efficiently generating a multi-tree structure of an HTML document; An item generating module for making the element designated by the user into one item; An item checking module for checking whether one item has a valid value in multiple trees created by the HTML tree generating module; An item storage module for continuously storing the updated contents in a database when the contents of the element indicated by the item is updated; And Add or remove items from the list with the item list, download the HTML of the item to be checked through the timer via the Internet, manage the storage and loading of the item list, transfer it to the management server, and create the content to be displayed to the user. Information collection system on the Internet, characterized in that it comprises a client item manager. The method of claim 1, wherein the management server A network input / output control module that is responsible for direct information input and output with the client information collecting agent through the Internet; A server item storage module for storing item information transmitted from the clients in a DB server; A server HTML tree generation module for quickly and efficiently generating a multi-tree structure of an HTML document; An item checking module that checks whether one item has a valid value in multiple trees created by the server HTML tree generating module; And And a server item manager which has a list of items to be directly inspected by the management server, and downloads an HTML document of items to be inspected from the Internet and manages production of contents to be transmitted to the client. Information collection system on the Internet. The information collection system of claim 1, wherein the item stores location information in a corresponding web page, a URL of a web page to which a corresponding element is designated, text of an element, and an inquiry period of the item. The method of claim 3, wherein the location information of the item includes information on the relationship between the node characteristics and the parent node from the node corresponding to the HTML element of the item to the top node of the tree structure of the web page. Information gathering system on the Internet. A plurality of clients, a management server that registers and manages the clients as members, and a plurality of information providing servers are interconnected through the Internet, and each of the clients includes only information required by the user from the information providing servers. It is applied to the information gathering system on the Internet where the information gathering agent is collected for selective and periodic collection. Designating, by the user, an HTML element selected from a specific web page of the information providing server as an item; Generating a tree structure of the web page through the HTML tree generation module of the information collection agent; After obtaining information of the HTML element designated by the user through a dynamic HTML object model, finding a target element node corresponding to the HTML element in the generated tree structure; Storing information on a relationship between a corresponding node feature and a parent node from a matched node to a top node of the tree structure in the item; And And storing the URL of the web page designated by the HTML element, the text of the element, and the inquiry period of the item in the item. The method of claim 5, Transmitting item information of the client to the management server to distinguish items that receive information from the management server; It is determined whether an item to be inspected exists in the client, and if present, the HTML string of the URL designated to the item is downloaded through the Internet, and the tree structure of HTML downloaded through the HTML tree generation module is generated. Confirming whether an item is valid with an item inspecting module of the information collecting agent; Determining whether the element is invalidated, outputting a notification to the user if the element is invalidated, and determining whether the element has been changed or not, and outputting updated item contents to the user if the element is changed; And Determining whether there is item information received from the management server, and if there is item information, outputting updated contents of the corresponding item to the user.