KR20020046558A - System and method for realtime based automatic market relationship analysis using text type information - Google Patents

Abstract

PURPOSE: A device and a method for analyzing an intelligent real time market relation are provided to supply script of market relation analysis information adapted to corresponding news by performing a morpheme analysis and a syntax analysis at a receiving point or a user requesting point with respect to a document information and using attributes of a word or a sentence in accordance with the analyses. CONSTITUTION: A user terminal(10) performs a wire/wireless data communication. A wire/wireless communication network(20) supports a communication connection the user terminal(10) to an external system. A router(30) is provided for performing a data conversion. A web server(40) supports an information input/output to the user terminal(10) and an internal device. A database(50) stores and manage various kinds of data. An internal and external data input system(60) inputs and transmits document information. An intelligent document information processing system(200) performs a market relation analysis with respect to corresponding document information in real time in accordance with document information received from the internal and external data input system(60) and user requesting contents received from the web server(40).

Description

System and method for realtime based automatic market relationship analysis using text type information}

The present invention performs a natural language process on document information in real time at the time of receiving the document information or at the time of user request for the document information received from the internal and external data input system. The present invention relates to a method and apparatus for extracting or generating the most suitable script analysis information (script), and simultaneously providing the result and the original text of the document information, thereby allowing users to easily spread qualitative information, not quantitative information. The advantage is that the structure and the extent of its impact can be assessed in advance.

For example, a number of information service companies at home and abroad are currently providing real-time news information service online. In the transmission method, the first user accesses the service homepage and retrieves the news content directly. News is sent to the user via email or a new pop-up window. In addition, the service method is that when the user selects a headline of a particular news, most of the past provides a news article related to the body of the selected content at the same time. However, according to this service method, the amount of news increases, and when a wider range of news is transmitted, the concentration and interpretation power of users about news information is low. In order to prevent this, the contents of the news content may be organized in detail, but in this case, as the file size increases, the concentration of the users may be reduced.

In order to overcome this problem, in the present invention, the document information transmission method maintains the existing system. However, the service method for the document information, unlike the existing method, includes the main document information and related document information. The market relation analysis information (script), which means the characteristics of the related ripple structure in the relevant market, is added, and through this, the user's real-time interpretation and concentration on the document information will be increased.

Accordingly, the present invention has been made to solve the above-mentioned conventional problems, the morphological analysis and parsing in real time at the time of receiving the document information or at the time of the user request, the attribute of the word or sentence extracted accordingly Provide market relation analysis information (script) that best matches the news information. Such a method of extracting the script includes extracting a static flow chart previously input / stored by the service provider according to the characteristics of all types of document information, or dynamically using the important words extracted by the parsing module. A method of directly generating a dynamic flow chart can be used.

To this end, the present invention uses the morpheme analysis module and the syntax analysis module to extract the key words for the headline or the text content of the document information, and to assign weights to the extracted words using the document term management module. In order to weight each word, the distance between words or the frequency of occurrence of words are used. In addition, the market structure analysis module extracts and generates market relationship analysis information that is optimally matched with the high weight words selected in the above process. This procedure is performed in real time when the document information is received or when the user selects a headline or a part of specific document information on the initial screen of the service, and the result is directly checked on the screen of the service or directly provided to the user's system. A pop-up approach is possible.

Through the present invention system, users can analyze in real time how the document information will affect the relevant market in real time, and as a result, it is possible to increase the real-time interpretation ability of the document information and the corresponding proactive measures. .

1 is a block diagram of a device capable of performing intelligent real-time market relationship analysis at the point of data reception by adding an intelligent document information processing system according to the present invention;

2 is a block diagram of a device capable of performing intelligent real-time market relationship analysis at the time of user request by adding an intelligent document information processing system according to the present invention;

3 is a configuration diagram of a detailed apparatus for the intelligent document information processing system shown in FIGS. 1 and 2;

4 is a flowchart illustrating a process of receiving a result when a user accesses the system and selects specific document information.

FIG. 5 is a flowchart illustrating a process of performing an intelligent market relationship analysis in real time when document information is received according to the device configuration diagram of FIG. 1.

FIG. 6 is a flowchart illustrating a process of performing intelligent market relationship analysis in real time when a user selects or inputs a specific word or sentence according to the device configuration of FIG.

7 is a screen configuration diagram showing the intelligent market relationship analysis information (script) provided to the end user through the device configuration diagram of FIG.

<Explanation of symbols for the main parts of the drawings>

10: user terminal 20: wired and wireless communication network

30: router 40: web server

50: database 200: intelligent document information processing system

60: internal and external data entry system

Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings.

1 is a block diagram of a device capable of performing intelligent market relationship analysis in real time at the time of receiving data by adding the intelligent document information processing system 200 according to the present invention. Intelligent document capable of market relation analysis at the point of receipt of document information, including communication network 20, router 30, web server 40, database 50, internal and external data entry system 60 Information processing system 200 is configured to be added between the database 50 and the internal and external data entry system 60.

The user terminal 10 is composed of at least one communication and computer equipment including a communication interface, such as a modem and appropriate communication software to enable data transmission and reception with other systems, such as a computer terminal, the wired and wireless communication network 20 is the Internet It is composed of communication networks that can be used at home and abroad as well as communication networks.

The router 30 serves to support different communication protocols between communication systems, the web server 40 supports interfaces and information transmission and reception between various users and internal systems, and the database 50 is internal and external. It stores and manages the data transmitted from the data input system 60 or intelligent document information processing system 200 of the. In addition, the internal and external data input system 60 is a system provided with a data input and transmission device is a system capable of transmitting data to the intelligent document information processing system 200 and the database 50 of the present invention. The intelligent document information processing system 200 performs morphological analysis and syntax analysis on the received document information, extracts key words and sentences that can represent the document information, and best matches the result. Extracts and generates analysis information (script).

An embodiment of an intelligent real-time market relationship analysis apparatus and method using document information according to the present invention configured as described above will be described with reference to FIGS. 2 to 7.

FIG. 2 is a block diagram of an apparatus for performing intelligent market relationship analysis in real time at the time of user request by adding the intelligent document information processing system 200 according to the present invention. The difference from FIG. In the case of FIG. 1, the natural language processing process for the document information is performed at the time of receiving the document information, while FIG. 2 has a difference in analyzing the document information through the natural language processing process at the user's request point, not at the point of receiving the data. . In this case, when the document information is received from the internal and external data input system 60, it is simply stored in the database 50 without any manipulation on the corresponding received information.

3 is a block diagram of a detailed apparatus for the intelligent document information processing system 200 shown in FIG. 1 and FIG. 2, and shows contents transmitted through an internal and external data input system 60 or a web server 40. Receivable data receiving module 210, the morpheme analysis module 220 to analyze the morphemes for the received data to separate / extract meaningful words, and at the same time find a standard word for the abbreviation, extracted from the morphological analysis module Document term management module 230 for assigning a weight to each word by using the occurrence frequency of the words and the distance (distance) between the words, syntax analysis to extract the second important word using the weighted words The module 240, the document term management module 230 for recalculating the weights of the second extracted words, and the document term management module 230 are optimal time values for the final weighted words. Market relationship analysis module 250 for extracting and generating relationship analysis information (script), and a data input module that allows the internal or external service provider to directly create / enter market relationship analysis information (script) for a specific issue ( And a data transmission module 270 for transmitting market relationship analysis information (script), which is a result of the market relationship analysis module 250, to the web server 40 and the database 50, including 260. do.

4 is a flowchart illustrating a process of receiving a result when a user accesses the system and selects specific document information, where S denotes a step. In particular, this flowchart corresponds to a case where an intelligent market relation analysis of the document information is performed at the time of receiving the document information as shown in FIG. 1, and the user first accesses the service contents initial screen in a specific execution order (S10). ), The user selecting or inputting a specific sentence or word (S11), calling the input / output process of the database in the web server using the selection and input (S12), and the related content (text, script in the database). ) And transmitting the extracted contents to the user terminal, including the step of extracting (S13).

FIG. 5 is a flowchart illustrating a process of performing intelligent market relationship analysis in real time when document information is received by the device configuration diagram of FIG. 1. First, document information is received from an internal and external data input system 60. Receiving step (S21), the step of extracting the morpheme (word) for the received data through the morphological analysis module 220 (S22), to give the first weight in the document term management module 230 for the extracted words Step S23, re-extracting the word by the parsing module 240 (S24), giving a second weight to the re-extracted word (S25), and index for each weighted word by meaning Storing the words, scripts, and related contents in the database 50, including the step (S26) and comparing the scripts related to the words and extracting and generating an optimal script (S27). That S28) is added It shall be.

For a specific embodiment, consider the sentence {the insurance industry M & A is rapidly progressing}. When the sentence is received through the data reception module 210 of the present invention, the morpheme analysis module 220 finds morphemes (words) such as {insurance industry, M & A, rapidity, progress}. In order to detect such morphemes, the morpheme analysis module 220 refers to a preprocessing process for a sentence, a functional word dictionary reference, a central word dictionary reference, and a user term dictionary reference.

First of all, if a preprocessing process is performed to check a space between words and special symbols (except an English letter), the sentences are separated as in the insurance industry, M & A, and rapidly. The difference from the initial sentence is that the period is omitted.

Next, using the built-in functional dictionary, the morphological analysis module separates surveys, telephony suffixes, verbal suffixes, adverbative suffixes, auxiliary words, new endings, and endings attached to words. To give. That is, {insurance industry, M & A, rapid, abbreviated (adverbized suffix), progressed, (investigated),}

Next, the morpheme module finds the core words such as nouns, pronouns, adjectives, adjectives, verbs, adverbs, adverbs, adjectives, and interjections using the built-in dictionary of core words, and grammars about the results. Index of the award. That is, {insurance industry, M & A, rapid (noun), hagi (adverbized suffix), progress (noun), and (verb)}

Next, the morphological analysis module separates and organizes foreign words, standard words, and the like for the separated words using a built-in user glossary, and assigns grammatical indexes to the results. That is, {insurance industry (anti-life insurance industry; nouns), M & A (mergers and acquisitions; nouns), rapid (nouns), hagi (adverbized suffixes), progressive (nouns), (investigative), and (verb)} Using the results, the morphological analysis module finally extracts the morphemes (important words) for the sentence. As a result, the sentence is arranged as {insurance industry, M & A, rapidity, progression}.

Afterwards, the first weight of the word is assigned to the extracted morpheme (word) using the morpheme dictionary and the document term management module 230 (S23). In other words, by using the database built in the past to analyze the frequency of appearance of each morpheme to give weight to the morphemes, for example, as a result of examining the database, 'insurance industry' 500 times, 'M & A' 700 times, If the 'speed' is 200 times, and the 'progress' is 300 times, the weight can be simply given as 0.5, 0.7, 0.2, 0.3. Eventually, when the morphemes are arranged in the order of weight, it becomes like {M & A, insurance industry, progress, rapid}. In addition, the morpheme type is {noun + noun + verb + noun}, and whether there is a previously written syntax type matching the type is examined using a syntax type dictionary built in the parsing module 240. If such a type is found, the stem is reextracted to match the type (S24). As a result, the weight arrangement is rearranged as {insurance industry, M & A, progress, rapid}.

After that, the second weights are assigned to each morpheme (word) (S25). If there is a matching syntax type in the step, weights are assigned, but otherwise, the weights are all assigned a value of zero. And by comparing the syntax type extracted from the syntax type dictionary and the morpheme (word) type to give an index for each word meaning (S26), through this process in the morpheme {M & A, M & A, progress } Is the final morpheme. After that, the market relation analysis information (script) related to the morpheme is extracted using the market relation analysis module 250 (S27), and the words, scripts, and related contents are stored in a database (S28). If there is no related syntax type in the parsing step (S24), based on the weights given in the previous step (S23), that is, the script is found based on the morpheme having a high weight.

FIG. 6 is a flowchart illustrating a process of performing an intelligent market relationship analysis when a user selects or inputs a specific word or sentence according to the device configuration diagram of FIG. 2. First, the user accesses the service initial screen ( S31), the user selecting or inputting a specific sentence or word (S32), extracting the morpheme (word) for the received data through the morphological analysis module 220 (S33), the document for the extracted word Giving a first weight in the term management module 230 (S34), re-extracting words by the syntax analysis module 240 (S35), and giving a second weight for the extracted words ( S36), the step of assigning an index for each of the weighted words by meaning (S37), and comparing the script associated with the word and extracting and generating the optimal script (S38), including the words, script, Use related content Characterized in that the step (S39) of transmitting to the terminal 10 is added.

FIG. 7 is a full screen diagram including intelligent market relationship analysis information (script) provided to an end user through the apparatus diagram of FIG. 1 or FIG. This full screen diagram is again divided into a number of detailed screens, which are classified and sorted according to a specific criteria industry classification screen (1000), a news group screen (1100) showing the real-time news grouped for each industry by group, A script screen showing market relation analysis information (script) for each news, including a graph screen 1200 showing changes in the indexes classified by industry, in real time graph form, and a news content culture page 1300 showing the main content of individual news. 1400 has been added. Comparing the service method that is currently generally implemented and the service method of the present invention system using FIG. 7, only the newsgroup screen 1100 and the news book culture page 1300 are serviced so far. As described above, a script screen 1400 is added, including an industry classification screen 1000 and a graph screen 1200.

The sector classification method is as follows. First, the sector classification system classified by the securities company is used. Second, based on the settlement data submitted by the individual companies to the supervisory authority, the sectors are classified according to the business content of each company and the main production items. It is possible to classify. In addition, as a method for fitting document information generated in real time to the classified business type, first, a classification code is attached to each document information at the time of document information transmission, and at the time of document information reception, the corresponding information is automatically applied. Second, extract the important words contained in the received document information through the natural language processing process for the received document information, and compares with the industry word defined in each industry in advance based on this word It is also possible to combine them with the industry.

As described above, according to the present invention, since individuals can be provided with market relation analysis information (script) in real time with respect to document information transmitted in real time through a wired or wireless communication network, specific interpretation ability of the received document information is possible. Do. In addition, since the present invention supports a real-time service, it can play an important function as a part of decision support means for the general public who needs real-time decision making. Furthermore, through market relation analysis information (script) provided through the present invention system, users can grasp the scope of influence of the document information or the document information related to the market in advance, and also through this function, the preliminary response to the problem. There is an effect that becomes possible.

Claims (6)

User terminal capable of wired / wireless data communication, wired / wireless communication network supporting communication connection between user terminal and external system, router for data conversion, web server supporting information input / output between the user terminal and internal device, storage of various data And an internal and external data input system capable of inputting and transmitting document information, including a manageable database, comprising: document information received from the internal and external data input system or received from a web server Intelligent real-time market relationship analysis apparatus using document information, characterized in that the intelligent document information processing system for performing a market relationship analysis on the document information in real time according to the user request. The method of claim 1, The intelligent document information processing system includes a data receiving module, a morphological analysis module for morphological analysis of received data, a document term management module for assigning weights to classified morphemes, a syntax analysis module for parsing classified words, Data transmission that can transmit information to other systems, including a market relationship analysis module that extracts and generates an optimal market relationship analysis script using the final selected words, and a data input module that can communicate with data input. Intelligent real-time market relationship analysis apparatus using document information, characterized in that the module is added. User terminal capable of wired / wireless data communication, wired / wireless communication network supporting communication connection between user terminal and external system, router for data conversion, web server supporting information input / output between the user terminal and internal device, storage of various data And intelligent real-time market relationship analysis device using document information added with intelligent document information processing system that can analyze market relations of relevant document information in real time, including database that can be managed and database, internal and external data input system In the method, the method for analyzing the market relationship using the document information in real time at the time of receiving data from the internal and external data input system receiving the document information, extracting the morpheme (word) by the morphological analysis module Performing a first weight on the extracted words Granting the word, re-extracting the word through syntax analysis, granting the second weight to the word, assigning the index for each word according to the meaning, comparing the script related to the word, and then selecting the optimal script. And extracting and generating, and storing the words, scripts, and related contents in a database. User terminal capable of wired / wireless data communication, wired / wireless communication network supporting communication connection between user terminal and external system, router for data conversion, web server supporting information input / output between the user terminal and internal device, storage of various data And intelligent real-time market relationship analysis device using document information added with intelligent document information processing system that can analyze market relations of relevant document information in real time, including database that can be managed and database, internal and external data input system In the method, the method for analyzing the market relation using the document information in real time at the time of user request may include accessing a service initial screen, selecting or inputting a sentence or a word by a user, and morpheme (word) by a morphological analysis module. ), The extracted word Comparing the steps of assigning a primary weight, re-extracting words through syntax analysis, assigning secondary weights to words, indexing each word according to meaning, and scripts related to words. And extracting and generating an optimal script, and transmitting the word, script, and related information to a user terminal. The method according to claim 3 or 4, The method of extracting and generating the optimal script after comparing the scripts related to the word is to prepare and store the market relation analysis information (script) corresponding to the document information in the form of a file in advance and compare the related scripts upon request. After that, a dynamic real-time market using document information is added, which includes a static method of extracting the optimal script and a dynamic method for automatically generating and providing a script corresponding to the document information step by step at the time of request. Relationship Analysis Method. The method of claim 5, The extracted optimal market relationship analysis information (script) is an intelligent real-time market relationship analysis apparatus and method using the document information, characterized in that the form of a summary document is further added, including the form of a flow chart (flow chart).