KR100568369B1 - Method for classifying and searching data - Google Patents

Abstract

The present invention relates to a method for automatically classifying data and a method for searching / analyzing classification data. The method for automatically classifying data according to the present invention includes: a first step of selecting a predetermined specific part from data data recorded on a recording medium; A second step of classifying data included in a specific item field among the selected data into two or more independent items based on the contents of the data; A third step of selecting respective identification codes corresponding to the contents of the classified data and adding them to the data of the classified items; And a fourth step of recording the classified and added data data on another recording medium, so that the search and analysis of the data data closest to the data data necessary for the user's desired problem solving can be performed among the vast data data. In this way, it is easy to find new technologies for solving new problems or new solutions that do not infringe the rights of others, and also provide information on problem solving by searching and analyzing the data classified automatically. By providing searched and analyzed data on the screen in various forms such as graph map, pie map, tree map, etc., it is possible to quickly find the desired data from the searched and analyzed data from the point of view of problem solving. It is a very useful invention that makes it easy to search for data and analyze analysis results.

Description

Automatic classification method of data and searching / analysis method of classification data {Method for classifying and searching data}             

1 is a flowchart of a conventional manual method for classifying data data;

2 is a flowchart of a conventional method of outputting a data screen of data;

3 is a schematic diagram illustrating an entire network configuration of an integrated information management system to which an embodiment of an automatic classification method of data and a search / analysis method of classification data according to the present invention is applied,

Figure 4 shows the flow of a preferred embodiment of the automatic classification method of data according to the present invention,

FIG. 5 is an exemplary diagram showing that the words and the identification codes of the solutions and solutions corresponding to the international patent classification codes (IPC Code) for each technical field are stored in association.

FIG. 6 is a diagram showing the content of one configuration data among data data accumulated and stored in a hard disk device (HDD) in FIG. 3;

Figure 7 shows the flow of one preferred embodiment of a method for searching / analyzing classified data according to the present invention,

8 shows a connection state of an integrated information management system through a web browser,

FIG. 9 illustrates an initial menu screen provided by an integrated information management system for retrieving user data data. FIG.

10 is a diagram illustrating an example of a screen output form of schematic information according to a search result of data data;

FIG. 11 illustrates a case in which data search results are outputted in the form of a tree map.

FIG. 12 illustrates a case in which the search result of the data data is output on a pie map form.

FIG. 13 shows a case where a search result of data data is output on a screen in the form of a graph map.

※ Explanation of code for main part of drawing

PC ₁ , PC ₂ , PC ₃ , ..: Search Terminal 10: Network Interface

20: ROM 21: Memory Bank

30: hard disk device (HDD) 40: central processing unit

41: classification program 42: search engine

43: DB manager 50: CD-ROM driver

100: integrated information management search server

The present invention relates to a method for automatically classifying data and a method for searching / analyzing classified data, and more particularly, based on a specified classification standard for solving a problem in which data is presented, such as a patent, a paper, or a technical report. The present invention relates to a method for automatically classifying data and a method for searching / analyzing classified data so as to classify the contents and search and analyze the classified data data by a problem solving task or means required by a user.

1 is a flowchart of a conventional method of manually classifying data data, and FIG. 2 is a flowchart of a method of outputting a conventional data data screen. Referring to this, a method of classifying and outputting conventional data data is as follows.

Here, data data means patents, articles, technical reports, parts information, and the like, which will be described using patent information as an example.

First, the manual classification method of patent information according to FIG. 1 will be described. The user selects data to be searched from the patent information (S1), and whether the data is in document form or CD-ROM (hereinafter, referred to as "CD-ROM"). 'Disk') or other forms are identified (S2).

After that, it is checked whether the data is recorded on a recording medium such as a disk or a tape (S3). If it is not recorded on the recording medium, the data is stored on a recording medium such as a disk so that a database (DB) can be made ( S4).

Upon completion of the above series of preparations, each data is read sequentially from the disc to determine whether there is a field to be searched, i.e., the data of the required item (e.g., the purpose of the invention, the abstract, etc.). If it is present (S5), if present, the user directly retrieves the data and sorts it manually (S7). If it is not present, the search target field is randomly selected (S6).

Thereafter, the reliability of the classified data is checked through the above process (S8), that is, whether the classification is normally performed, and if the condition is satisfied, the classified data is stored in a hard disk device (HDD) or the like, and the screen or printer Output and exit (S9).

According to FIG. 2, the screen output method of the patent information classified as described above is as follows.

In the state where the patent information is classified and stored as described above, the user selects the material of the object to be searched (S11), and inputs a keyword related to the searched material (S12) to check whether the corresponding searched material is detected. (S13).

When the corresponding search target data is detected, the list output screen of the detected data is selected (S14) and then displayed on the screen in the form of a predetermined search result (S15), so that the user checks whether the data on the output displayed screen is appropriate. (S16).

As a result of the check, if the screen data is not appropriate, the above search process is repeated until the appropriate data is obtained, and if it is determined that appropriate, the data is stored and output (S17).

However, in the conventional method of classifying and displaying data data as described above, a large amount of information is directly classified manually, and general information is used as an object of construction without considering the characteristics of the business fields of the enterprise, thereby inefficiently functioning the database. There is a problem to utilize, and in particular, since searching the database by keyword and displaying the search results on the screen only in the form of a list, it is impossible to provide various types of search results required by the user, and also to search the database. Since input elements are limited to keywords and the like, the amount of information to be searched is enormous, and accordingly, the user has to search for a number of times and check the vast search result information displayed on the screen one by one, and has a troublesome problem of finding the desired data.

Therefore, the present invention has been created to solve the above problems, so that various data data are automatically classified based on a predetermined classification criteria according to the retrieval purpose of the data, and the user needs the classified data data as described above. The purpose of the present invention is to make it easy to search and analyze data by a problem solving task or means, so that the search or analysis of data data can be easily obtained.

According to an aspect of the present invention, there is provided a method for automatically classifying data, comprising: a first step of selecting a predetermined specific portion from data data recorded on a recording medium; A second step of classifying data included in a specific item field among the selected data into two or more independent items based on the contents of the data; A third step of selecting respective identification codes corresponding to the contents of the classified data and adding them to the data of the classified items; And a fourth step of recording the classified and added data data on another recording medium.

In addition, a method for retrieving automatically classified data according to the present invention may include: a first step of retrieving a classification code related to problem solving, which is input from the outside, on a predetermined database (DB); And a second step of outputting and displaying, on the screen, the schematic information of the data data stored in association with the retrieved classification code.

Hereinafter, an embodiment of an automatic classification method of data and a search / analysis method of classification data according to the present invention will be described in detail with reference to the accompanying drawings.

3 is a diagram illustrating an entire network configuration of an integrated information management system to which an embodiment of an automatic classification method of data and a search / analysis method of classification data according to the present invention is applied, and a plurality of search terminals through a predetermined network (LAN). A network interface 10 for exchanging data with (PC ₁ , PC ₂ , ..); A ROM (ROM) 20 in which representative words of a problem and a means for solving the corresponding problem and an identification code corresponding to the international patent classification code (IPC Code) for each technical field are stored in association; A memory bank 21 for program loading; A central processing unit 40 executing a program loaded in the memory bank 21 to perform classification, storage, retrieval and analysis of data data; A hard disk device (HDD) 30 for storing and storing classified data data and storing a program for sorting / searching; And a CD-ROM driver 50 into which a CD-ROM containing patent data data is inserted.

4 is a flowchart illustrating a preferred embodiment of the automatic classification method of data according to the present invention. Hereinafter, the automatic classification method of FIG. 4 according to the present invention will be described in detail with reference to the configuration of FIG. 3.

First, as in the conventional method described above, it is assumed that the data data as patent information, and in the ROM 20, as shown in FIG. 5, an international patent classification code (IPC) corresponding to each technical field and a solution corresponding thereto. In the state that the representative words of the task and the solution and the identification code for the related data are stored together, a series of preparation processes such as selection of the data to be analyzed and the identification of the data and the storage of the data on the disk are completed. Afterwards (S21 to S24), when a CD-ROM (disk) containing patent information (data data) is inserted into the CD-ROM driver 50, it is loaded from the hard disk device 30 into the memory bank 21 and is subjected to the central processing. The classification program 41 executed by the device 40 requests transmission of data data in a disc inserted into a control unit (not shown) in the CD-ROM driver 50, and thus the CD-ROM. The control unit in the driver 50 causes the data data in the inserted disc to be read sequentially.

Accordingly, the classification program 41 performs a classification operation on the data data read out from the disk, and the classification program 41 reads from the data data read out and inputted from the inserted disk. As described above, the analysis target field is selected according to the existence of the specific analysis target field (S25 ~ S26), and when the data of the field of the analysis target, ie, the required item (summary of the patent data) is selected, the classification is performed. The program 41 detects a word or a sentence having regularity in the selected analysis target field and classifies the sentence into a solution and a solution which are basic data for analysis based on this (S27). .

As an example, in the abstract (summary) part of Japanese patent data data, since the object field of the invention is composed of a TO sentence and a BY sentence, the TO sentence is a task and the BY sentence is a solution. The purpose field of the invention will be classified.

In addition, in the US patent data data, the first sentence of the entire sentence of the abstract is classified as a solution, and the remaining sentences are classified as a solution. In the case of domestic patent data, the data with the summary portion resolves the summary portion. The claim is classified as a solution, and if there is no summary, the first sentence of the claim is classified as a solution and the remaining sentence as a solution, so that it can be used as basic data for analysis. .

Thereafter, the classification program 41 checks the reliability of the classified data as described above, that is, confirms whether it is normally classified (S28), and when the condition is satisfied, classifies the identification code corresponding to the classified data content. The central processing unit 40 detects the international patent classification code specified in the classified patent data from the corresponding data data, and selects a code of a higher concept of the detected classification code. do. That is, when the detected classification code is G11B7 / 00, the higher concept code of this classification code is G11B.

The classification program 41 searches the ROM 20 for the selected higher concept code in the ROM 20 to identify representative words of the problem and the solution stored in association with the higher concept code (S29). It is searched whether the identified representative word is included in the data of the classified problem and solution item (S30). The representative words of the solution and solution stored in association with the international patent classification code are related to the technical field. It can be assigned to words based on the core technical terms commonly used in the word or words described in the international patent classification code.

When the search result includes the words 'simplification' and 'stabilization' which are pre-designated representative words in the solution sentence, and the word 'aggregation' is included in the solution means sentence, the classification program 41 ) Reads out the identification code stored in association with the corresponding representative word from the ROM 20, and adds C and E codes to the solution sentence according to the grant code of FIG. E code is added (S31).

As a result, items of the problem and the solution are extracted and classified, and an identification code representative of the content is added to the patent data, and the patent data data whose contents are reconstructed in FIG. 6 is stored in the hard disk device 30. It becomes (S32).

The classification program 41 executes the classification process and the code addition process according to the above-described problems with respect to all the data data read out from the inserted disc, whereby all the patent data data contained in the inserted disc is shown. By reprocessing in the form of 6, to be stored separately in the storage area of the hard disk device 30, the database for the integrated information management system is constructed.

On the other hand, since the representative words for the above-mentioned technical fields and solutions for the technical fields are different for each technical field (for example, the agricultural field may have a representative word such as 'increase', but in the electric field, Although the code assigned to the problem and the solution means the same code, the representative word represented by the code may be designated differently. That is, in G11B, A corresponds to the representative word of 'error prevention', but in A01B field, A may correspond to the representative word of 'proliferation'.

FIG. 7 illustrates a flow of a preferred embodiment of a method for searching / analyzing automatically classified data according to the present invention. Hereinafter, the searching / analyzing method of FIG. 7 according to the present invention will be described in detail with reference to FIG. 3. do.

In the case of searching or analyzing patent data in the database of the integrated information management system constructed as described above (S41), the user first executes a web browser installed in a search terminal (PC), and thereby By accessing the search server 100 in which the integrated information management database is established (S42), the contents of the homepage as shown in FIG. 8 can be received. For this purpose, the search server 100 performs data search and analysis of the search terminal. This is done by loading a search engine (Search Engine) and a DB management program for the memory bank 21 from the hard disk device 30.

The search engine 42, which is an execution program in the search server 100, downloads the JABA APPLET constituting the integrated information management system to the corresponding search terminal through the network interface 10 when a connection is made from the search terminal. Done.

As described above, when the access to the integrated information management search server 100 is completed and becomes a searchable state, the user wants to search and analyze the initial menu screen as shown in FIG. 9 provided from the integrated information management search server 100. Set the search and analysis conditions of the data data to (S43). That is, the search and analysis of patent data data to be searched and analyzed (for example, US patent, Japanese patent, Korean patent, etc.), IPC code (for example, G01B *, etc.), keywords (for example, video) From among input elements such as a problem code (eg, A) and a solution code (eg, B), the value is selectively set.

JABA APPLET in the downloaded search terminal converts the condition values set by the user into a query for a desired search condition or analysis type and delivers them to the search engine 42 used as a user and a Java interface. The search engine 42 receives the query again, and transmits the query to the DB manager 43 which is previously read and loaded into the memory bank 21 to be driven, thereby causing the DB manager 43 to input the query. The data data corresponding to the set query is retrieved from the hard disk device 30.

Accordingly, the DB management program 43 searches for the patent data data corresponding to the received query from the hard disk device 30 (S44). , Korean patent data, 'G01B *' as the IPC code, 'A' as the problem identification code (corresponding to the representative word of 'error prevention' in Fig. 5), and 'B' as the solution identification code (Fig. 5). In the case of a representative word of 'substitution', the DB manager 43 performs all of the input and set conditions among the vast patent data data stored in the hard disk device 30 in the form of FIG. It is checked whether or not satisfactory data data exists in the hard disk device 30.

As a result of the search, it has patent data that satisfies all the conditions set by the user, that is, the IPC code starts with G01B, the problem identification code is 'A', and the solution identification code is 'B'. When the Korean patent data is detected from the hard disk device 30 (S45), the DB management program 43, in the searched data data, outlines information such as the title of the invention, the applicant, the patent number, and the like. It extracts only the data constituting the data, and delivers it to the search engine 42, the search engine 42 constitutes the schematic information to be the content of the form of FIG.

By transmitting the outlined information thus configured to the corresponding search terminal PC through the network interface 10, the outline information on patent data data that the user wants to find is displayed on the screen of the search terminal PC in the form of FIG. The output is displayed (S46).

Through the schematic information output as described above, the user can select the patent data of the user's desired data and search and analyze it more closely. When the specific patent data is selected and requested to view the data through the schematic information, In operation S47, the search engine 42 recognizes the query data from the query data through the network interface 10, and then searches for the specific patent data data requested from the hard disk device 30 and reads it out. In operation S48, the specific patent data data read as described above is transmitted to the corresponding search terminal PC through the network interface 10 so as to be displayed on the screen. By solving the problem and the solution means to be displayed separately (S49), the user to display the detailed patent Whether the patent data of the destination you're looking for before you read all of the data makes it easy to see the challenges focused.

In addition, when outputting the searched document data, the corresponding code for the problem solving item (solution and solution) and the words represented by the code are also output together, and the identification code for the problem solving item is found in the searched material. If there are a plurality of, all of them are provided to the search terminal.

As such, when a plurality of troubleshooting codes are provided, the searching users determine which of the representative words related to the plurality of troubleshooting are appropriate as the corresponding technical data, and provide feedback to the operator of the search server 100. In the case of a feed-back, the order of providing a plurality of representative words is provided in a different order, and if necessary, the user can distinguish between the words searched by the first representative word and the data searched by the first and subsequent representative words. It can also be provided to.

In the above embodiment, the integrated information management retrieval server 100 allows the user to select a screen output form of the IPC classification or problem-oriented analysis results of the retrieved data data, according to the selection, the tree as shown in FIG. Tree Map type, Pie Map type as shown in Fig. 12 (for data retrieved when no task is specified), and Graph Map type as shown in Fig. 13 (also in Fig. 13). Graphic data can be provided in various forms such as IPC and or keyword only, if it is specified, it can be represented in graph form as a solution or a solution code.

In addition, in the above embodiment, the technical classification or search classification of the data based on the international patent classification code as a classification standard, but also the data data can be classified based on the independent classification system, wherein the classification system is structural It can be selected as Structure Classification Code by Function or Function Classification Code by Functional Property, where functional classification code is again used as Effect Code and Subject Code. After subdividing into, it can also be used in combination.
The above-described preferred embodiments of the present invention are disclosed for purposes of illustration, and those skilled in the art can improve, change, and substitute various other embodiments within the technical spirit and scope of the present invention disclosed in the appended claims below. Or addition may be possible.

The automatic classification method of the data and the search / analysis method of the classification data according to the present invention as described above are automatically classified based on the problem and the solution suggested by the data data, and the classified data can be searched and analyzed. Representing and storing the words as the solution, the search and analysis of the data data that is closest to the data data necessary for the user's problem solving, among the vast data data, so that the new technology or the rights of others to solve the new problem Make it easy to find new solutions that do not infringe.

In addition, it provides search results of automatically classified data data by classifying the contents of problem solving, and provides search and analysis data on the screen in various forms such as graph map, pie map, tree map, etc. It is a very useful invention that makes it possible to find the desired data quickly from the point of view of problem solving, and also to make it easier for users to search and analyze the data data.

Claims (17)

A first step of selecting a predetermined specific portion from the data data recorded on the recording medium; A second step of classifying data included in a specific item field among the selected data into two or more independent items based on the contents of the data; A third step of selecting respective identification codes corresponding to the contents of the classified data and adding them to the data of the classified items; And And a fourth step of recording the classified and added data data on another recording medium. The method of claim 1, The two or more items to be classified include a problem item and a solution item item. The method of claim 2, The problem item is the automatic classification method of data, characterized in that the data appearing after the target phrase among the data included in the specific item field. The method of claim 2, And the solution means item is data that appears after the means phrase among the data included in the specific item field. The method of claim 2, The problem item is the automatic classification method of data, characterized in that the data corresponding to the first sentence of the green portion of the data contained in the specific item field. The method of claim 2, And the solution means item is data corresponding to a sentence excluding the first sentence of the green portion of the data included in the specific item field. The method of claim 2, The problem item is an automatic classification method of data, characterized in that the data corresponding to the summary portion of the data contained in the specific item field. The method of claim 2, And the solution means item is data corresponding to a claim portion of data included in the specific item field. delete The method of claim 1, In the third step, a predetermined word is searched for in the classified item data, and an identification code corresponding to the searched word is selected and added to the classified item data. . The method of claim 10, In the third step, a plurality of corresponding identification codes are added. The method of claim 10, The identification code is an automatic classification method of data, characterized in that the same value corresponds to a word having a different meaning between different technical fields. The method of claim 1, On the basis of the correlation between the classified data, the classification information of the data is displayed on the screen in any one of the form of tree, pie, graph, the automatic classification method of the data. The method of claim 1, The classification into two or more independent items is performed based on an International Patent Classification Code (IPC Code). In the database search / analysis method, A first step of retrieving a classification code related to problem solving from a predetermined database (DB) input from the outside; And And a second step of outputting and displaying, on a screen, the schematic information of the data data stored in association with the searched classification code. The method of claim 15, The classification code for solving the problem is a code for one of the predetermined words included in the content of the problem or solution item. The method of claim 15, The first step is a search / analysis method of the classified data, characterized in that the search based on the technical field.

Images