KR100276694B1 - How to View Information Search Results - Google Patents

Abstract

The present invention relates to a method of viewing an information search result that can effectively view a large number of search results in an information search. After visualizing and showing an order view, another result view, a document summary view, and a state information view in a result lens, A first step of requesting selection, a rotation button selected according to the user's selection, a second step of rotating the spiral of the sequence view and matching the summary information of the document summary view, and a document summary view according to the user's selection When the current ranking button of is clicked, the third step of displaying the document of the corresponding URL using a web browser, closing the browser and returning to the first step and repeating the above process, and according to the user's selection, When the button is clicked, the fourth step of performing the user's menu, and when the detail button of the document summary view is clicked according to the user's selection, Close the information, the dialog shown after the dialog is characterized in that it returns to the first step consisting in a fifth step of repeatedly performing the above process.

Description

How to View Information Search Results

The present invention relates to a method of viewing information retrieval results in which a large amount of retrieval results can be effectively seen in an information retrieval.

Information retrieval is the retrieval of information that best suits the needs of the user. In recent years, with the rapid growth of the Internet, information retrieval has changed into a problem of how to find the desired information quickly, accurately and easily from the Internet, an infinite source of information. However, meeting the needs of users on the Internet, called the sea of information, is not easy. In particular, as the number of users on the web increases, the development of tools that can be obtained by simply manipulating desired information becomes important.

In general, the number of search results of an information retrieval system (including a search engine) varies by information retrieval system, and many systems do not search or present huge results. However, general users, of course, vary according to the search purpose or the user, but generally only a small number of search results (about 20 or so in my case) are reviewed. If another information retrieval system searches the same database (information), the index or search model of the information retrieval system is somewhat different. You don't know, but the documents you really need can be included in the overall search document.

Therefore, there is a need for a method of searching for desired information through various viewing of vast search results.

Almost all information search engines such as conventional information retrieval systems, Yahoo! Korea, Simmani, Information Investigator, Naver, Magpie, and Cosshik have a user-defined number of search results (usually 10) first. The ranking is displayed along with information such as importance, title, summary, and URL. Afterwards, if you want to see the search results continuously, click "Previous, 1, 2, 3, 4, ..., Next", etc. prepared at the bottom to continue to display as many as the user-defined number initially.

The following is the number of results of a general information search engine for the query "information retrieval system" executed on June 4, 1997.

Naver: 11,532 (1998.3.11.)

Web glider: 3,268

Information detective: 1,809 (1997.6.11.: 2,056 cases, 1998.3.11.: 2,688 cases)

Anime Search: 1,522

Uni Finder: 1,522

Zoom: 1,211 cases

Wakano: 1,507 cases

KOR-SEEK: 500

AltaVista: 200

IBot: 183

Yahoo Korea: 74 cases (1998.3.11.)

Miss Miss: 46

% MetaSearch Engine for AnySearch, Investigator, Simmani, Koshik, AltaVista

% Search engines in each of the top 10 searchable documents

Sim Manny: 35

Conventional systems have to view the search results in turn by a user-defined number of large search results.

Fig. 1 shows an example of Naver's results for a conventional natural language query " information retrieval system ", and Fig. 2 shows an example of Yahoo! Korea's result for a conventional natural language query " information retrieval system " Fig. 3 shows an example of the result of the information detection for the conventional natural language query "information retrieval system".

Meeting the needs of users on the Internet, called the sea of information, is not easy. In particular, as the number of users on the web has increased, many web-based information retrieval systems (search engines) have existed. In many cases, the user cannot obtain the desired information by the step-by-step method. In addition, although the indexing method, search model, and search ranking method may be different for each search system, the desired information of the user may be included in the entire search result.

Accordingly, the present invention provides a variety of views through search term adjustment, classification adjustment, mutual information, and search term merger for the searched results in order to search a document that satisfies the user's request for a large number of results more quickly and accurately. The purpose of the present invention is to provide a method of viewing information retrieval results that can improve retrieval efficiency of information retrieval in an easy and convenient way by refining results.

The present invention for achieving the above object is a first step for visualizing and showing the sequence view, the other results view, the document summary view and the status information view in the result lens, and then rotates according to the user's selection When the button is selected, the second step of rotating the spiral of the order view and matching the summary information of the document summary view, and if the current ranking button of the document summary view is clicked according to the user's selection, uses the web browser to After showing the document, the third step of closing the browser and returning to the first step to repeat the process, and if the button of the other result view is clicked according to the user's selection, the fourth step of performing the user's menu and When the detailed button of the document summary view is clicked according to the user's selection, the document summary information dialog is displayed, and then the dialog is closed to return to the first step. W, including a fifth step of repeatedly performing the above process is characterized in that is made.

1 is a result example of Naver for a natural language query "information retrieval system" conventionally used.

2 is a result example of Yahoo! Korea for a natural language query " information retrieval system " used conventionally.

Fig. 3 is an example of the result of an information detective on a natural language query " information retrieval system "

Figure 4 is a block diagram showing an entire system applied to the present invention.

5 is a block diagram showing a resultant lens applied to the present invention.

6 is a block diagram showing a keyword adjustment component applied to the present invention.

7 is a view showing the initial screen of the resulting lens applied to the present invention.

8 is a diagram showing a document summary information dialog applied to the present invention.

Figure 9 shows a full text view of the document as applied to the present invention.

10 is a view showing another result view menu applied to the present invention.

11 is a view showing a search word adjustment screen applied to the present invention.

12 is a diagram illustrating a result of a search word adjustment applied to the present invention.

13 is a view showing a classification screen applied to the present invention.

14 is a diagram showing a mutual information screen applied to the present invention.

15 is a diagram illustrating a search word sum total initial screen applied to the present invention.

16 is a diagram showing a result of a search word sum aggregation applied to the present invention.

17 is a flow chart for viewing a large number of search results in accordance with the present invention.

<Description of Signs of Major Parts of Drawings>

1: server 2: client

3: result lens 4: browser

10: search term adjustment component 20: classification component

30: mutual information component 40: sum aggregation component

101: Query Box Module 102: Search X Module

103: weight control module 104: RL control unit

105: Order view 106: Document summary view

107: other result view 108: state information view

201: Query Contribution (Participation) 202: Query Box

203: Reorder button 204: Red hand

205: Classification view 206: Sub helix view

207: query set configuration button 208: recycling button

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

FIG. 4 is a block diagram showing the overall configuration of the system to which the present invention is applied, and includes a result lens (RL) 3 and a browser (RL) included in the server 1, the client 2, and the client 2. browser; 4).

The result lens 3 provides a variety of views for a large number of search results to help search the document more accurately. In addition, the resultant lens 3 is a plug-in program, which calls a hypertext transport protocol daemon (httpd) or a common gateway interface (CGI) on the server side through a browser 4 according to a user's request. Exchange data with (1).

The resultant lens 3 provides an environment in which the documents retrieved from the primary search can be viewed from various points of view through various views. Based on the first search result, the user may assign a user weight through a search word adjustment function. Targeting the primary search results (or search term adjustment results), the user can quickly find the desired information from the large search results through functions such as classification, mutual information, and aggregation.

FIG. 5 is a block diagram showing components of the resultant lens 3 applied to the present invention, which is composed of a search term adjustment component 10, a classification component 20, a mutual information component 30, and a sum aggregation component 40. do. These components react organically according to the user's request and communicate with the server 1.

The function of each component constituting the resultant lens 3 is as follows. The search term adjustment component 10 may arbitrarily assign a search term weight and provide a user search term adjustment function by excluding a search term using a search X function.

The classification component 20 shows a set of search documents classified by field by the classification algorithm of the system.

The mutual information component 30 reorders the documents using the mutual information of the system.

The aggregation aggregation component 40 defines the search terms as a term set and performs reordering for each term set.

FIG. 6 is a block diagram illustrating a keyword adjusting component 10 applied to the present invention. The function of the keyword adjusting component will be briefly described as follows.

The user 2 adjusts the search term weight in the search X module 102 and the weight adjusting module 103 in the search term box module 101 (c1). If the user presses the 'reorder' button after adjusting the search word, the user-defined query is formatted (c2), and the formatted query and all search term session information during the first search are transmitted to the server 1 (c3). The server 1 reorders according to the transmitted session and the formatted query and transmits the result to the RL control unit 104 (c4). The RL control unit 104 reflects the result in the order view 105, the document summary view 106 and the other result view 107 according to the transmission result of the server 1 (c5).

Next, a method of utilizing the information retrieval system having the above structure will be described.

1. Initial Screen

FIG. 7 shows the initial screen of the resultant lens 3 applied to the present invention, showing how the resultant lens 3 is executed on the retrieved document.

The result lens 3 basically shows a ranking view 105 for ranking the search documents, ranking, importance of search documents, search query participation, title, summary, URL, etc. A document summary view 106, another result view 107 for viewing different results, and a status information view 108 for indicating the number of documents retrieved, and the like. The order view 105 and the document summary view 106 correspond to each other's information 1: 1. For example, it is reflected 1: 1 based on the highest ranking position (corresponding to the first position, not just the first document).

The ordinal view 105 basically follows the spiral model and high ranking documents are located outward, and the arrows at the bottom are page up / down and forward / backward, respectively. The function rotates by clicking, and the contents of the document summary view 106 are automatically updated according to the rotation.

In the document summary view 106, the change of the current rank searched by changing the previous rank and method of the result document (searched document) is shown centering on the finger icon. Highlights the current ranking of documents that have changed significantly (more than five levels up).

The importance calculated by the built-in result formula of the result document is a result calculated by the search system to show how accurate the document is. The search term participation rate indicates a contribution (participation degree) for each term for the searched document as a graph 201. Each search term is identified by color. In addition, title, summary, and URL information can be viewed simply, and detailed information can be viewed again, such as a document summary information dialog shown in FIG. 8 by clicking a detailed radio button.

The retrieved information can also be viewed through another browser, as shown in FIG. 9, by double-clicking the button of the current ranking.

FIG. 10 illustrates a pull-down menu of another result view applied to the present invention, and includes a search term adjustment, classification, mutual information, and search term sum collection menu.

2. Refine query

11 illustrates a search word adjustment screen applied to the present invention.

Search term adjustment is an interface that assigns a user weight to a used search term. The search word received in the search X item on the left side of the search box 202 means that the document including the search word is "never come out" as a search result. The search term weights range from 0 to 1 and are adjusted for the slide bar. The weight of the query term is used to calculate the built-in algorithm of the search system. After completing the search word adjustment, if the reorder button 203 is pressed, the search word adjustment result is displayed.

12 is a view illustrating a search word adjustment result screen applied to the present invention, where a red hand shape between a previous rank and a current rank can be seen. The red hand shape 204 is displayed when it is increased by 5 or more compared with the previous ranking, and it can be seen that the ranking has increased a lot by weighting the user.

3. Classification

FIG. 13 shows a classification screen applied to the present invention, which shows that it starts by selecting a classification in another result view 107.

The classification is represented by using a spiral model of a collection of search documents classified by a classification algorithm of a system for documents of a first search or a search term adjustment result. Each time the icon of the left sorted view 205 is clicked, the right sub-helical view 206 displays corresponding information. The sub helix view 206 also has a 1: 1 correspondence with the document summary view 106.

4. Mutual Information

FIG. 14 illustrates a mutual information screen applied to the present invention, wherein mutual information is executed by selecting mutual information in another result view 107, and mutual information of a system targeting documents of a primary search or a search term adjustment result. Shows the results of automatic reordering using. As shown in FIG. 14, it can be seen that there are many differences from when the user weight is applied.

5. Search term sum total

FIG. 15 illustrates a search word sum collection initial screen that appears when the search word sum aggregation is selected in another result view 107 applied to the present invention.

In the search term sum aggregation, search terms are defined as a term set (TS), and TS is reordered. After selecting the search words targeted for a group and pressing the TS configuration button 207, one TS is generated. Each time the TS configuration button 207 is pressed, TS1, TS2, etc. are automatically created and the search box is automatically cleared. After repeating this series of steps, the reorder button 208 is pressed.

FIG. 16 is a cross-sectional view showing the result of the search word sum aggregation applied to the present invention, and is a screen displayed when the reordering button 208 is pressed. The search term summation view shows the results reordered by TS in a spiral model, and the interface of the view of the search term summation is the same as the classification.

17 (a) and 17 (b) are flowcharts showing a large amount of search results according to the present invention.

Referring to FIG. 17A, the result lens visualizes the order view, the other result view, the document summary view, and the state information view (302), and then requests the user's selection (303). When the rotation button is selected according to the user's selection, the spiral of the sequence view is rotated, the summary information of the document summary view is matched, and the browser is closed and terminated (304 and 305). When the current ranking button of the document summary view is clicked according to the user's selection, after showing the document of the corresponding URL by using a web browser, the browser is closed, and the process returns to step B 301 to repeat the above process (306, 307). And 308). If a button of another result view is clicked according to the user's selection, the process proceeds to step A to perform a continued command (309). When the detailed button of the document summary view is clicked according to the user's selection, after showing the document summary information dialog, the dialog is closed and the process returns to step B 301 to repeat the process (310, 311, and 312).

Referring to FIG. 17 (b), when another result view button of the result lens is clicked (314), a user may request a menu selection among a search term adjustment, classification, mutual information, and search term sum menu (314). ). If the search term adjustment is clicked according to the user's selection, the search term adjustment view is shown and the user is asked to select among reordering buttons, query box adjustment, and other result view buttons (315 and 316). If the user selects the reorder button, the process returns to step B 301 and repeats the above process (317). If the user selects the query box adjustment, the user adjusts and displays the search word (318). If you select to return to step A (313) to repeat the process (319). When the mutual information is clicked according to the user's selection, after displaying the mutual information view (320), the process returns to step B 301 and repeats the process (321). If the classification is clicked according to the user's selection, after showing the classification view (327), the user is asked to select between the classification type node and other result view buttons (329). When the user selects the classification type node, the user shows the sub-helix view (330), and if another user selects another result view button, the process returns to step A (313) and the process is repeated (331). When the search term summation view is clicked according to the user's selection, the search term summation view is displayed (322) and the user is asked to select among the search term set configuration, another result view button, and the reordering button (323). When the user selects the query set configuration, the user configures and displays the query set (324). If another user selects a view button, the process returns to step A (313), and the process is repeated (325). Proceeding to step C 328, the continued command is performed (326).

As described above, the present invention can easily access the information retrieval system through refinement of the results of various views through the reconciliation of search terms, the adjustment by classification, the mutual information, and the aggregation of search terms for a large amount of information retrieval results. In addition, the search performance of the information retrieval system can be improved, and the meta search engine is often used rather than the general search engine.

In addition, in order to be applied to a web-based search engine and information retrieval system, it is composed of a plug-in based on Netscape. The present invention is to take the plug-in form to the web browser dedicated to the communication process between the server and the client, it is possible to reduce the volume of the client due to the communication module and to easily distribute the plug-in by mime type Has an advantage.

Claims (6)

A first step of visualizing and showing the sequence view, the other result view, the document summary view, and the status information view in the result lens, and requesting the user's selection; If the rotate button is selected according to the user's selection, the second step of rotating the spiral of the order view and matching the summary information of the document summary view, When the current ranking button of the document summary view is clicked according to the user's selection, the third step of displaying the document of the URL using a web browser, closing the browser and returning to the first step and repeating the above process Wow, A fourth step of performing a user's menu when a button of another result view is clicked according to a user's selection; When the detailed button of the document summary view is clicked according to the user's selection, after the document summary information dialog is displayed, a fifth step of repeating the process by closing the dialog and returning to the first step may be performed. Viewing method of information retrieval result. The method of claim 1, The fourth step of performing the user's menu is A method of viewing information retrieval results comprising requesting a user to select one of the user's search term adjustment, classification, mutual information, and search term sum total menu. The method of claim 2, The query reconciliation step Showing a query reconciliation view and asking the user to choose between a reorder button, reconcile a query box, and other results view buttons; If the user selects the reordering button, the process returns to the step of visualizing and showing the order view, the other result view, the document summary view, and the status information view in the result lens, and repeating the above steps; If the user chooses to refine the search box, the steps to adjust and show the query, If the user selects a different result view button, the method returns to the step of requesting selection of a menu of search term adjustment, classification, mutual information, and search term sum total, and repeating the above process. How to view the results. The method of claim 2, The mutual information step And displaying the mutual information view and returning to visualizing and displaying the order view, another result view, document summary view, and status information view in the result lens, and repeating the above process. . The method of claim 2, The classification step Showing the classification view and asking the user to choose between a classification type node and other result view buttons; When the user selects a classification type node, the steps shown in the sub helix view, When selecting another result view button, the method returns to the step of requesting selection of a menu of search term adjustment, classification, mutual information, and search term sum total, and repeating the above process. Viewing method. The method of claim 2, The search term aggregation step Showing the query aggregation view and asking the user to choose between constructing a query set, a different results view button, and a reordering button; When the user chooses to construct a query set, the steps to configure and show the query set, Selecting another result view button, repeating the above process by returning to a step of requesting a selection from a menu of search term adjustment, classification, mutual information, and search term sum collection; And when the reordering button is selected, proceeding to requesting the user to select from among a classification type node and another result view button, and performing a continued command.