KR20220068793A - Method for providing news analysis service using robotic process automation monitoring - Google Patents

Abstract

Provided is a method for providing a news analysis service using robotic process automation (RPA) monitoring. The method comprises the steps of: performing news clipping on news data to correspond to a pre-input keyword file using RPA; performing natural language processing of the clipped news data and generating an issue keyword related to a main keyword of the news data using an association rule algorithm; collecting social media data and performing text mining using the main keyword and the issue keyword; and extracting a noun keyword from the text-mined social media data and visualizing the noun keyword as a keyword network.

Description

Method for providing news analysis service using RPA monitoring

The present invention relates to a method of providing a news analysis service using RPA monitoring, and a platform capable of collecting and analyzing news, extracting main keywords and related issue keywords, and analyzing contents and comments of related social media to visualize it as a keyword network provides

RPA (Robotic Process Automation) is a technology that reproduces a series of repetitive and routine actions performed by a human through a computer as it is. The business support system should be designed and implemented for each functional item such as database, screen design (User Interface), and data processing logic. It is a technology that defines in detail how it works and where data is input and implements it as it is. Therefore, due to the introduction of RPA, there is no need to modify or change the system, and since the unit processes to be developed are mutually independent, the development system and environment are simple and development is relatively easy. These characteristics make RPA a differentiated solution, and it has been excluded from the development of the work automation system, but the areas that can be automated include acquiring and inputting data that is difficult to interface with, or writing special reports for each user in each organization that is difficult to standardize. As the fact that it can be expanded is highlighted, related solutions are being actively applied.

At this time, a technology for building a database using RPA or automating collection and service using bots was developed. And in Korea Patent Publication No. 2019-0134874 (published on December 05, 2019), an accident simulation of the plant is performed according to the process conditions and accident scenario information of the plant to generate and store the accident simulation result, and the accident simulation operation and The configuration of a robotic process that automates and controls the storage process in the database unit, and automatically collects data in real time using a bot, analyzes it with an artificial intelligence algorithm, and calculates an index. A configuration for generating real-time responsive news content by using is disclosed.

However, although there is a configuration that generates news content with a robot, it is possible to collect news content from at least one medium using RSA, analyze competitors and SOV (Share Of Voice), keywords, related words, and further sentiment analysis. The platform does not exist. For competitor analysis and market analysis, people are still manually monitoring, and repetitive tasks such as monitoring and report writing are continuing. In addition, even if the news published in the media is collected and the issue is identified, analyzing the feed or response of social media related to this gives another load, so using natural language processing (NLP) and related rule analysis The skills to analyze comments related to news are also required. Accordingly, research and development of a method for visualizing comments by automatically clipping news using RPA and extracting related issue keywords from the collected news is required.

An embodiment of the present invention automates the monitoring task that was simply repeated by automatically clipping news articles using RPA (Robotic Process Automation), and after natural language processing of the news articles, the main keyword and Automatically generate related issue keywords, collect social media data by combining main keywords and issue keywords, conduct text mining, and use TF-IDF (Term Frequency-Inverse Document Frequency) and association rule algorithms to A keyword network visualized by extracting the comment as a response as a noun keyword, assigning the extracted noun keyword to each node, and placing the nouns closer as the relevance between them increases. It is possible to provide a method for providing a news analysis service using RPA monitoring, which can provide However, the technical task to be achieved by the present embodiment is not limited to the technical task as described above, and other technical tasks may exist.

As a technical means for achieving the above-described technical problem, an embodiment of the present invention includes a step of clipping news data to correspond to a previously input keyword file using RPA (Robotic Process Automation), clipped news The data is processed in natural language (Natural Language Processing) and the issue keywords related to the main keywords of the news data are generated using the association rule algorithm, social media data are collected using the main keywords and the issue keywords, and text mining is performed. and extracting a noun keyword from the text-mined social media data, and visualizing the noun keyword as a keyword network.

According to any one of the above-described problem solving means of the present invention, by automatically clipping news articles using RPA (Robotic Process Automation), the simple repetitive monitoring task is automated, and the news articles are processed in natural language and then related The rule algorithm automatically creates issue keywords related to the main keyword, collects social media data by combining the main keyword and the issue keyword, then conducts text mining, TF-IDF (Term Frequency-Inverse Document Frequency) and related rules Using an algorithm, the customer's response, a comment, is extracted as a noun keyword, and after allocating the extracted noun keyword to each node, the higher the relevance between each noun, the closer they are placed. can be increased to provide a visualized keyword network.

1 is a view for explaining a system for providing a news analysis service using RPA monitoring according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a news analysis service providing server included in the system of FIG. 1 .
3 is a diagram for explaining an embodiment in which a news analysis service using RPA monitoring is implemented according to an embodiment of the present invention.
4 is a diagram illustrating a program code for executing the news analysis service of FIG. 3 .
5 is a diagram illustrating a screen on which the news analysis service of FIG. 3 is provided.
6 is an operation flowchart illustrating a method of providing a news analysis service using RPA monitoring according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement them. However, the present invention may be embodied in several different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is "connected" with another part, this includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated, and one or more other features However, it is to be understood that the existence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded in advance.

The terms "about", "substantially", etc. to the extent used throughout the specification are used in or close to the numerical value when manufacturing and material tolerances inherent in the stated meaning are presented, and are intended to enhance the understanding of the present invention. To help, precise or absolute figures are used to prevent unfair use by unconscionable infringers of the stated disclosure. As used throughout the specification of the present invention, the term "step of (to)" or "step of" does not mean "step for".

In this specification, a "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be implemented using two or more hardware, and two or more units may be implemented by one hardware. Meanwhile, '~ unit' is not limited to software or hardware, and '~ unit' may be configured to be in an addressable storage medium or to reproduce one or more processors. Thus, as an example, '~' denotes components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functions provided in the components and '~ units' may be combined into a smaller number of components and '~ units' or further separated into additional components and '~ units'. In addition, components and '~ units' may be implemented to play one or more CPUs in a device or secure multimedia card.

Some of the operations or functions described as being performed by the terminal, apparatus, or device in the present specification may be performed instead of by a server connected to the terminal, apparatus, or device. Similarly, some of the operations or functions described as being performed by the server may also be performed in a terminal, apparatus, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal means mapping or matching the terminal's unique number or personal identification information, which is the identification data of the terminal. can be interpreted as

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

1 is a view for explaining a system for providing a news analysis service using RPA monitoring according to an embodiment of the present invention. Referring to FIG. 1 , the system 1 for providing a news analysis service using RPA monitoring may include at least one user terminal 100 , a news analysis service providing server 300 , and at least one manager terminal 400 . have. However, since the news analysis service providing system 1 using RPA monitoring of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1 .

At this time, each component of FIG. 1 is generally connected through a network 200 . For example, as shown in FIG. 1 , at least one user terminal 100 may be connected to the news analysis service providing server 300 through the network 200 . In addition, the news analysis service providing server 300 may be connected to at least one user terminal 100 and at least one manager terminal 400 through the network 200 . In addition, at least one manager terminal 400 may be connected to the news analysis service providing server 300 through the network 200 .

Here, the network refers to a connection structure in which information exchange is possible between each node, such as a plurality of terminals and servers, and an example of such a network includes a local area network (LAN), a wide area network (WAN: Wide Area Network), the Internet (WWW: World Wide Web), wired and wireless data communication networks, telephone networks, wired and wireless television networks, and the like. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), 5th Generation Partnership Project (5GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi (Wi-Fi) , Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth (Bluetooth) network, NFC ( Near-Field Communication) networks, satellite broadcast networks, analog broadcast networks, Digital Multimedia Broadcasting (DMB) networks, and the like are included, but are not limited thereto.

In the following, the term at least one is defined as a term including the singular and the plural, and even if the at least one term does not exist, each element may exist in the singular or plural, and may mean the singular or plural. it will be self-evident In addition, that each component is provided in singular or plural may be changed according to embodiments.

At least one user terminal 100 receives an insight report after analyzing news or social media for its products or services using a news analysis service-related web page, app page, program or application using RPA monitoring It may be a terminal of a customer who wants .

Here, the at least one user terminal 100 may be implemented as a computer that can access a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop. In this case, the at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one user terminal 100 is, for example, as a wireless communication device that guarantees portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) terminal, a smart phone, a smart pad, a tablet PC, etc. may include all kinds of handheld-based wireless communication devices.

The news analysis service providing server 300 may be a server that provides a news analysis service web page, app page, program, or application using RPA monitoring. In addition, the news analysis service providing server 300 receives an analysis request from the user terminal 100 and receives a file including a central keyword from the manager terminal 400 , and at least one medium in at least one medium based on the central keyword. It may be a server that does news clipping as a way to collect news of In addition, the news analysis service providing server 300 may be a server that extracts a main keyword from the collected news data and extracts an issue keyword related to the main keyword. In addition, the news analysis service providing server 300 searches for at least one social media data including a main keyword and an issue keyword, analyzes content and comments in the social media data, extracts a noun keyword, and each noun keyword It may be a server that creates a keyword network in such a way that nodes are allocated to , and the size of the node increases as the number of each noun keyword increases, and the distance between the noun keywords increases as the correlation between the noun keywords increases. In addition, the news analysis service providing server 300 may be a server that generates an Insight Report based on the collected and analyzed data in a report format and then transmits the collected and analyzed data to the user terminal 100 by using RPA.

Here, the news analysis service providing server 300 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop.

The at least one manager terminal 400 may be a terminal of a doctor using a web page, an app page, a program, or an application related to a news analysis service using RPA monitoring. And, when the analysis request of the user terminal 100 is received via the news analysis service providing server 300 , the at least one manager terminal 400 collects news to be analyzed and creates a keyword file for analysis. It may be a terminal for uploading to the news analysis service providing server 300 . In addition, the at least one manager terminal 400 may be a terminal for uploading a final inspection result to the news analysis service providing server 300 with respect to the result of the analysis completed by the news analysis service providing server 300 .

Here, the at least one manager terminal 400 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop. In this case, the at least one manager terminal 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one manager terminal 400, for example, as a wireless communication device that guarantees portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) terminal, a smart phone, a smart pad, a tablet PC, etc. may include all kinds of handheld-based wireless communication devices.

FIG. 2 is a block diagram illustrating a news analysis service providing server included in the system of FIG. 1 , and FIG. 3 describes an embodiment in which a news analysis service using RPA monitoring is implemented according to an embodiment of the present invention. 4 is a diagram illustrating program code for executing the news analysis service of FIG. 3, and FIG. 5 is a diagram illustrating a screen on which the news analysis service of FIG. 3 is provided.

Referring to FIG. 2 , the news analysis service providing server 300 includes a collecting unit 310 , a generating unit 320 , a performing unit 330 , a visualization unit 340 , a reporting unit 350 , and a sending unit 360 . ) may be included.

The news analysis service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one user terminal 100 and at least one manager terminal 400 uses RPA monitoring. When transmitting a news analysis service application, program, app page, web page, etc., at least one user terminal 100 and at least one manager terminal 400, the news analysis service application, program, app page using RPA monitoring , can install or open web pages, etc. In addition, the service program may be driven in at least one user terminal 100 and at least one manager terminal 400 by using a script executed in a web browser. Here, the web browser is a program that enables the use of a web (WWW: World Wide Web) service, and refers to a program that receives and displays hypertext written in HTML (Hyper Text Mark-up Language), for example, Netscape. , Explorer, Chrome, and the like. In addition, the application means an application on the terminal, for example, includes an app (App) executed in a mobile terminal (smartphone).

Referring to FIG. 2 , the collection unit 310 may clip news data to correspond to a keyword file previously input using RPA (Robotic Process Automation). When the collection unit 310 clips news data to correspond to a previously input keyword file using RPA, the input keyword file is a first Excel file containing main keyword data and data collected by company, competitor, and industry It may include a second Excel file including the collection keyword data for, negative keyword data, which is a keyword to be excluded from collection, and media data for collecting news data, and the media data may be stored as a hashtag.

The generating unit 320 may perform natural language processing on the clipped news data and generate an issue keyword related to the main keyword of the news data using an association rule algorithm. When the generating unit 320 performs natural language processing on the clipped news data and generates an issue keyword related to the main keyword of the news data using the association rule algorithm, the clipped news data and the clipped news data included in the Comment data can be tokenized and stop words can be removed with TF-IDF (Term Frequency-Inverse Document Frequency).

Here, TF-IDF is a weight used in information retrieval and text mining, and is a statistical value indicating how important a word is in a specific document when there is a document group consisting of several documents. The term frequency is a value indicating how often a specific word appears in a document, and the inverse document frequency refers to a word that appears frequently in the document but not many in other documents. The value of TF-IDF is used as the product of word frequency and inverse document frequency.

Basically, TF can be calculated using the total frequency of words in the document, but if the number of words increases, the value can continuously increase. can do. Alternatively, the size of the document may be resolved and the actual frequency may also be reflected by taking a logarithm of the TF value using a logarithmically scaled frequency. Finally, by using the augmented frequency, the maximum scale may not exceed 1 in a way that indicates the relative frequency of words according to the document length. As a result, the TF-IDF increases as a word has a high frequency within a specific document and there are fewer documents including the word among all documents. This allows you to filter out words that appear frequently in all documents.

In addition, the issue keyword becomes a related word of the main keyword, that is, a related keyword, which uses big data processing technology to generate issue keywords and related data that are highly semantically related to the central keyword for a specific time point or the entire period. An efficient extraction method can be used. Using Hadoop, one of the big data platforms, and HBase, which is widely used as an open source among NoSQL, a large amount of data for a specific central keyword is collected, and then the collected contents are divided into sentences to perform morphological analysis. After morpheme analysis, the frequency of each extracted word is calculated and stored by date, and in order to automatically collect data, the word extracted as an issue keyword candidate is sent back to the central keyword to collect data recursively. Finally, when the desired information is obtained through a search, issue keywords and related data with high semantic relevance for a specific point in time or for the entire period are provided. You can use this to find issue keywords.

First, there is a module that collects data from Internet data for central keywords, and morphological analysis is performed through the collected data. The frequency can be calculated through the MapReduce operation for the morpheme-analyzed data, and the data can be stored according to the HBase schema configured for the data extracted from the data storage module. In order to automatically collect a large amount of data, it is possible to collect data recursively by excluding previously searched words among extracted words and then sending the extracted words as a central keyword. In the data extraction module, you can acquire issue keywords and related data for the finally searched central keyword by using the HBase filter function. At this time, the data collection module collects Internet data searched for by the central keyword. When extracting content from numerous Internet data, crawling is performed to extract the content of the main text, and for the collected Internet data, not only the content of the main text, but also the date and Internet address of the Internet data can be collected. Thereafter, each content may be divided into sentences in order to perform morphological analysis on the collected Internet data content.

In the morpheme analysis module, morpheme analysis is performed to extract only common nouns and proper nouns for the contents of Internet data acquired through the data collection module. After determining each part-of-speech by morpheme analysis, it is possible to remove parts of speech such as articles, prepositions, surveys, and conjunctions that will be stop words, and extract common nouns and proper nouns that will become issue keyword candidates. In the frequency extraction module, the frequency of each word can be calculated using the common nouns and proper nouns received from the morpheme analysis module. After calculating the frequency of extracted words when a central keyword is input by performing a MapReduce operation, the secondary sort provided by Hadoop can be used to sort in the order of highest frequency. In the data storage module, the data extracted from each module described above is stored in HBase. When storing data in HBase, the central keyword is stored as a row key, and the column family is divided into three, Data is stored in a qualifier for each column family. The first column family stores the word extracted through the central keyword, the second column family stores the address of the most recent Internet data with related words, and the third column family name stores the creation date. A value may store the frequency of each extracted word.

In order to automatically collect data on issue keywords, it is possible to collect data in a recursive manner by sending a word extracted as a central keyword to the central keyword. When collecting data recursively, if a word with a central keyword is stored as a low key in HBase, it is possible to prevent data duplication by setting it to skip Internet data without collecting it. When collecting data recursively, data is collected by setting the number of search steps so as not to fall into an infinite loop, and the data extraction module collects issue keywords and related data through the filters for rows, qualifiers, and values provided by HBase. extract

After searching the word to be searched with the RowFilter, use the ValueFilter to search the list of high-relevant words, and modify the TF (Term Frequency) to correlate the central keyword (main keyword) with the issue keyword. measure After calculating the average value using the frequency of extracted words extracted through the central keyword, the correlation between the central keyword and the extracted word is calculated. Thus, you can find each data and provide it with issue keywords. Of course, the method of searching for and databaseizing the issue keyword is not limited to the above-described method.

The generator 320 may automatically generate an issue keyword related to the main keyword by applying the association rule algorithm to the news data and the comment data. In this case, the association rule algorithm may be an Apriori Algorithm. This is an algorithm that finds a rule that can best explain the main relationship observed in data variables, such as association rules, preferences, and information filtration, and can be used to find important associations between data having high multidimensional or complex relationships.

In the Apriori algorithm, the relationship is expressed in two forms: Frequent Item Sets or Association Rules. The frequent item set is a collection of items that occur frequently together, and the association rule suggests that there is strength in the relationship between items. In a specific data set, using the above two methods, it is possible to determine whether there is a relationship to a set of items. Since the formula for the degree of support for the reliability of a specific item is known, it will not be described in detail. Support is the ratio of a data set that includes specific data in a data group, and reliability is defined as an association rule and is a type of clustering that groups highly correlated data. Of course, it will be apparent that the analysis can be performed by various methods other than the above-described method, and that various methods are available.

The performing unit 330 may collect social media data using the main keyword and the issue keyword and perform text mining. When performing text mining and collecting social media data using the main keyword and the issue keyword, the performing unit 330 may search for social media data related to the issue keyword among the social media data, Tokenization of content and comments included in social media data related to keywords, removal of stop words with TF-IDF (Term Frequency-Inverse Document Frequency), and application of association rule algorithm to content and comments can do.

In this case, text mining is a process of extracting relational data between data refined by processing natural language from unstructured data composed of characters or text, rather than numerical structured data, and refined using morpheme analysis technology. In this case, a topic modeling algorithm may be used during text mining, which is a statistical method for inferring a topic by clustering related words in atypical documents such as online comments or messages on SNS. Latent Semantic Analysis (LSA), pLSA (Probabilistic Latent Semantic Analysis) and LDA (Latent Dirichlet Allocation) may be used.

In this case, in an embodiment of the present invention, since the LDA algorithm does not secure a document-level probabilistic model, probabilistic topic modeling based on Dirichlet Allocation may be used. It is possible to predict which topic the document is dealing with by analyzing the probability distribution of co-occurrence of given words. Gibbs Sampling used in LDA topic analysis can be used. way.

In this case, a method of increasing the accuracy of review analysis by selectively removing words based on neutrality may be further used as a preprocessing method. First, sparse words can be removed using Delete Sparse terms. The purpose of this method is to remove words extracted due to typos or words that appear very few. A threshold set to remove rare words may be a method of removing rare words in which the number of occurrences of words relative to the number of documents is less than or equal to a preset percentage. Second, words are removed based on neutrality using Delete Neutral terms.

The purpose of this method is to remove words that appear frequently according to the properties of each type of data, not words that affect classification. Neutrality-based word removal is basically removing rare words based on sparsity and then removing words belonging to both sets based on neutrality, the degree to which a word appears in both useful and non-useful review sets. is defined as the Neutrality Index. Of course, since various pre-treatment methods may exist in addition to the above-described pre-treatment method, the present invention is not limited to the above-described method.

The visualization unit 340 may extract a noun keyword from the text-mined social media data and visualize the noun keyword as a keyword network. When the visualization unit 340 extracts a noun keyword from the text-mined social media data and visualizes the noun keyword as a keyword network, TF-IDF (Term Frequency-Inverse Document Frequency) and association rule algorithm are applied to find the noun keyword. can be extracted. In addition, the visualization unit 340 extracts a noun keyword from the text-mined social media data, and when visualizing the noun keyword as a keyword network, assigns a node to the noun keyword, and closes the distance when the correlation between the noun keywords is high The larger the number of mentions of noun keywords, the larger the node size can be formed and visualized as a keyword network.

The reporting unit 350 extracts a noun keyword from the text-mined social media data in the visualization unit 340 , and visualizes the noun keyword as a keyword network, followed by news data, main keywords, issue keywords, social media data and nouns. You can create an Insight Report using keywords.

The sending unit 360 extracts a noun keyword from the text-mined social media data in the visualization unit 340, visualizes the noun keyword as a keyword network, and then sends the keyword network to a pre-stored email address using RPA. can

Hereinafter, an operation process according to the configuration of the news analysis service providing server of FIG. 2 will be described in detail with reference to FIG. 3 as an example. However, it will be apparent that the embodiment is only one of various embodiments of the present invention and is not limited thereto.

Since the program code corresponding to FIG. 3 has to be described with reference to FIG. 3 , this will be described with reference to FIG. 4 . The analysis service according to an embodiment of the present invention may be divided into two stages. One is news clipping shown in FIG. 3A , and the other is social media analysis shown in FIG. 3B . We will mainly explain two steps.

<News Clipping>

Referring to FIGS. 3A and 4A , when a new news article is checked, a number of days of news articles to be collected is set as of today. Then, the search list is checked in the search target.xlxs, that is, the Excel file of FIG. 4B, and it is checked whether the folder structure is in the order of year < month < day < keyword, and then the folder is created or skipped. Referring to FIG. 4C , in the first Excel file, the name of the Excel file containing each central keyword information is called, and in the second Excel file, keywords, negative keywords, and media information for data collection by company/competitor/industry are stored as hashtags. configured to be And, referring to FIG. 4D , creating a list to contain the news data (news articles) to be collected and checking the news 10 per page through the for statement, turning the page with the While statement, until the page no longer exists will repeat

Referring to Figure 4e, since the date information of the collected articles has a media name and unnecessary data, only the date is extracted with a regular expression. Since it is not marked as , it is calculated and changed to the date format. Referring to FIG. 4G , the collected results are saved as an Excel file, the respective files are called again, and the collected results are saved as a sheet within one Excel file.

<Social Media Analysis>

3b and 4h, the collected news articles are tokenized, stopwords are removed, and issue keywords related to the main keyword are automatically generated with the Apriori algorithm. Together, you can collect cafe name/title/date/view count/posting content/comment information from Naver Cafe. In the case of Naver, since only up to 1000 cafe posts are displayed in the search window, to solve this problem, searches are performed on a daily basis. Repeat the for statement until the maximum desired date and the accumulated number in time_manage match. Because the computer is faster than the Internet reaction speed, it recognizes that there is no specific data and can terminate the program. Selenium's EC function starts collecting data after all data is normally present on the web page.

As shown in Fig. 4j, only title/body/comments are retrieved and text mining is started, and three classes are integrated into one column as shown in Fig. 4k. Referring to Fig. 4l, TF-IDF used in the above process And Apriori algorithm is used again to extract what customers think of keywords and issues on SNS as nouns. Referring to FIG. 4M , when the extracted data is visualized, the correlation between nodes with a close distance is strong, and the larger the size of the node, the more frequently mentioned keywords. The result is shown as shown in FIG. 4N.

Referring to FIGS. 5A and 5B , it is possible to automate tasks such as monitoring, report writing, competitor analysis, and market analysis that were performed manually by humans by using RPA and text mining technology. In this case, the technology used may be RPA, R, or Python.

The matters not described with respect to the method of providing a news analysis service using RPA monitoring of FIGS. 2 to 5 are the same as those described above for the method of providing a news analysis service using RPA monitoring through FIG. 1 or from the described contents. Since it can be easily inferred, the following description will be omitted.

6 is a diagram illustrating a process in which data is transmitted/received between components included in the system for providing a news analysis service using RPA monitoring of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process in which data is transmitted and received between each component will be described with reference to FIG. 6 , but the present application is not limited to such an embodiment, and the example shown in FIG. 6 according to the various embodiments described above will be described. It is apparent to those skilled in the art that the data transmission/reception process may be changed.

Referring to FIG. 6 , the news analysis service providing server clips news data so as to correspond to a previously input keyword file using RPA (Robotic Process Automation) (S6100).

Then, the news analysis service providing server performs natural language processing on the clipped news data and generates an issue keyword related to the main keyword of the news data using the association rule algorithm (S6200), and uses the main keyword and the issue keyword. Collect social media data and perform text mining (S6300).

In addition, the news analysis service providing server extracts a noun keyword from the text-mined social media data, and visualizes the noun keyword as a keyword network (S6400).

The order between the above-described steps S6100 to S6400 is merely an example, and is not limited thereto. That is, the order between the above-described steps ( S6100 to S6400 ) may be mutually changed, and some of these steps may be simultaneously executed or deleted.

The matters not described with respect to the method for providing a news analysis service using RPA monitoring of FIG. 6 are the same as those described for the method of providing a news analysis service using RPA monitoring through FIGS. 1 to 5 or from the contents described above. Since it can be easily inferred, the following description will be omitted.

The method of providing a news analysis service using RPA monitoring according to an embodiment described with reference to FIG. 6 may also be implemented in the form of a recording medium including instructions executable by a computer, such as an application or program module executed by a computer. can Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. Also, computer-readable media may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

In the method for providing a news analysis service using RPA monitoring according to an embodiment of the present invention described above, an application basically installed in a terminal (which may include a program included in a platform or operating system, etc. basically installed in the terminal) may be executed, and may be executed by an application (ie, a program) directly installed in the master terminal by a user through an application providing server such as an application store server, an application, or a web server related to the corresponding service. In this sense, the method for providing a news analysis service using RPA monitoring according to an embodiment of the present invention described above is implemented as an application (ie, program) installed basically in a terminal or directly installed by a user, and is transmitted to a computer such as a terminal. It may be recorded on a readable recording medium.

The above description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a dispersed form, and likewise components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be interpreted as being included in the scope of the present invention. do.

Claims (10)

In the news analysis service providing method executed in the news analysis service providing server,
Clipping news data to correspond to a keyword file previously input using RPA (Robotic Process Automation);
processing the clipped news data in natural language and generating an issue keyword associated with the main keyword of the news data using an association rule algorithm;
collecting social media data using the main keyword and the issue keyword and performing text mining; and
extracting a noun keyword from the text-mined social media data, and visualizing the noun keyword as a keyword network;
A method of providing a news analysis service using RPA monitoring comprising a.
The method of claim 1,
In the step of clipping the news data to correspond to the previously input keyword file using the RPA,
The previously input keyword file may include: a first Excel file including main keyword data; and
a second Excel file including collected keyword data for data collection by company, competitor, and industry, negative keyword data that is a keyword to be excluded from collection, and media data for collecting the news data;
including,
The method of providing a news analysis service using RPA monitoring, wherein the media data is stored as a hashtag.
The method of claim 1,
The step of natural language processing of the clipped news data and generating an issue keyword related to the main keyword of the news data by an association rule algorithm comprises:
Tokenizing the clipped news data and comment data included in the clipped news data, and removing stop words to TF-IDF (Term Frequency-Inverse Document Frequency);
automatically generating an issue keyword related to the main keyword by applying the association rule algorithm to the news data and comment data;
A method of providing a news analysis service using RPA monitoring, which is executed while performing.
The method of claim 1,
Extracting a noun keyword from the text-mined social media data, and visualizing the noun keyword as a keyword network,
extracting the noun keyword by applying Term Frequency-Inverse Document Frequency (TF-IDF) and the association rule algorithm;
A method of providing a news analysis service using RPA monitoring, which is executed while performing.
The method of claim 1,
Extracting a noun keyword from the text-mined social media data, and visualizing the noun keyword as a keyword network,
allocating nodes to the noun keywords, arranging a distance close to each other when the relevance between the noun keywords is high, and forming a size of a node larger as the number of mentions of the noun keywords increases to visualize the keyword network;
A method of providing a news analysis service using RPA monitoring, which is executed while performing.
The method of claim 1,
The association rule algorithm is an Apriori Algorithm, a method of providing a news analysis service using RPA monitoring.
The method of claim 1,
After extracting a noun keyword from the text-mined social media data and visualizing the noun keyword as a keyword network,
generating an Insight Report using the news data, main keywords, issue keywords, social media data, and noun keywords;
The method of providing a news analysis service using RPA monitoring further comprising a.
The method of claim 1,
Collecting social media data using the main keyword and the issue keyword and performing text mining includes:
searching for social media data related to the issue keyword among the social media data;
tokenizing content and comments included in social media data associated with the issue keyword;
removing stop words with TF-IDF (Term Frequency-Inverse Document Frequency); and
applying the association rule algorithm to the content and comments;
A method of providing a news analysis service using RPA monitoring, which is executed while performing.
The method of claim 1,
After extracting a noun keyword from the text-mined social media data and visualizing the noun keyword as a keyword network,
pre-sending the keyword network to a pre-stored e-mail address using the RPA;
The method of providing a news analysis service using RPA monitoring further comprising a.
A computer-readable recording medium recording a program for executing the method of any one of claims 1 to 9.

Images