KR20170057179A - Method for evaluating relation between keyword and asset value and Apparatus thereof - Google Patents

Abstract

An apparatus for evaluating relevance of keywords and asset values, according to one embodiment of the present invention, comprises: one or more processors; a memory which loads a computer program carried by the processor; a storage which stores date-basis price information on pre-registered assets, and a daily keyword generated by the performance of the computer program; and a network interface which transmits the daily keyword. The computer program comprises the operations of: collecting text content posted on a first date through the internet; extracting the keyword from each of the text content and generating a daily keyword of the first date; generating information on a date-basis exposure frequency for each keyword contained in the daily keyword of the first date; and comparing the information on the date-basis exposure frequency for each of the generated keywords with the date-basis price information on the pre-registered assets, then determining assets corresponding to each of the keywords among the pre-registered assets. The technical objective of the present invention is to provide a method and an apparatus which analyze the impacts of various issues on the asset price, and provide an investment guide for users by analyzing the impacts of keywords on the asset price.

Description

METHOD AND APPARATUS FOR EVALUATING KEYWORD AND ASSET PRICE RELEVANCE THEREOF

The present invention relates to a method and apparatus for evaluating keyword and asset price relevance. More particularly, the present invention relates to a method and apparatus for evaluating the relevance between an automatically generated daily keyword and an asset price using text contents.

Recently, according to the development of information communication, a variety of issues across the society are widely shared on the Internet. Economic issues such as corporate earnings announcements and development plans in specific regions affect individual stock prices and real estate price changes. In addition, not only economic issues but also political and social issues such as international relations and accidents have a wide impact on the economy. Certain social events, for example, have a negative impact on people's consumer sentiment, affecting the economy.

These various social issues are becoming more and more influential as they are expanded and reproduced in the Internet space. Accordingly, analysis of the impact of issues on asset value changes is required in establishing an individual asset investment or disposal plan.

However, among many issues, there is no service that predicts issues affecting the prices of assets held or assets being invested. In addition, there is no service that provides guidance to investors about the asset by analyzing the specific influence on the asset of each issue.

It is a technical object of the present invention to provide a method and apparatus for analyzing the impact of various issues on asset prices.

More specifically, the present invention provides a method and apparatus for extracting keywords representing the issue from text contents in which various issues are described.

It is another object of the present invention to provide a method and apparatus for automatically determining the influence of the keyword on an asset.

Another technical problem to be solved by the present invention is to provide a method and apparatus for providing an investment guide to a user by analyzing the effect of a keyword on the price of an asset.

Specifically, the present invention provides a method and apparatus for predicting the influence of keywords on assets based on the relevance of keywords and assets collected in the past.

Another technical problem to be solved by the present invention is to provide a method and apparatus for providing a keyword related to an asset that the user is possessing or is aiming at.

Specifically, the present invention provides a method and an apparatus that can provide a user with a chance to respond to a keyword that will become an issue later, by providing a keyword related to the asset being held or targeted.

The technical objects of the present invention are not limited to the above-mentioned technical problems, and other technical subjects not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a method for automatically generating a daily keyword using text content, the method comprising: collecting text contents posted on a first date through the Internet; extracting keywords from each of the text contents; Generating the daily keyword of the first date by using the keyword pool of the first date and the comparison result of the keyword pool of the first date and the keyword pool of the second date.

According to an aspect of the present invention, there is provided a keyword and asset price relevance evaluation method comprising the steps of: collecting text contents posted on a first date through the Internet; extracting keywords from each of the text contents; Generating impression frequency information for each keyword belonging to each keyword belonging to the first day of the month, generating impression frequency information for each of the generated keywords, And determining an asset corresponding to each of the keywords among the assets previously registered according to the comparison.

According to an aspect of the present invention, there is provided a keyword and asset price relevance evaluating apparatus comprising at least one processor, a memory for loading a computer program executed by the processor, pricing information for each registered date of the asset, A storage for storing a daily keyword generated by the user and a network interface for transmitting the daily keyword, the computer program comprising: an operation for collecting text content posted on a first date via the Internet; An operation for generating a daily keyword for the first date, an operation for generating date-specific exposure frequency information for each keyword belonging to the daily keyword of the first date, and an operation for generating an exposure frequency for each keyword Information and the above As the re-comparing the date pricing information of the registered assets, it includes an operation that of the pre-registered property, determining a property corresponding to the respective keywords.

According to another aspect of the present invention, there is provided a keyword and asset price relevance evaluating apparatus comprising: at least one processor; a memory for loading a computer program executed by the processor; Wherein the computer program further comprises: an operation for identifying an asset whose price change amount per date is greater than or equal to a threshold value during a first period of the preregistered asset, An operation for collecting the posted text content, an operation for extracting the keyword from each of the text contents to generate a daily keyword for the first date, a date for the second period for each keyword belonging to the daily keyword of the first date, Operator to detect star exposure frequency An operation for extracting a keyword having an exposure frequency by date for the second period from the daily keyword of the first date corresponding to a price change amount by date for the first period of the identified asset, And an operation of determining the keyword as a keyword corresponding to the identified asset.

According to the present invention, there is provided an apparatus and method for determining the influence of keywords collected on the Internet on assets.

According to another aspect of the present invention, there is provided a method and apparatus for predicting fluctuation of an asset price due to the influence of keywords collected on the Internet.

According to the present invention, there is provided an apparatus and method for predicting a period during which keywords collected on the Internet have an influence on an asset price.

According to the present invention, there is also provided an apparatus and method for predicting when a keyword collected on the Internet affects an asset price.

In addition, according to the present invention, it is possible to provide a method and apparatus for providing a user with an investment guide for an asset by predicting various influences of keywords collected on the Internet on asset prices.

Further, according to the present invention, since the keyword is provided to the user that affects the price of the asset being held or targeted, the user can secure the responsiveness to the keyword.

1 is a conceptual diagram illustrating the relationship between an asset price and a keyword referred to in some embodiments of the present invention.
2 illustrates an example of a keyword and asset price relevance evaluation system according to an embodiment of the present invention.
3 is a block diagram of a service server according to another embodiment of the present invention.
4 is a flowchart of a method for automatically generating keywords using text contents according to another embodiment of the present invention.
Figure 5 is an illustration of a keyword pool, which is referenced in some embodiments of the invention.
6 is an exemplary diagram for describing a daily keyword, referred to in some embodiments of the present invention.
7 is an exemplary diagram illustrating a first time window on a daily keyword, referred to in some embodiments of the present invention.
Figure 8 is an exemplary diagram for describing a second time window on a daily keyword, referred to in some embodiments of the present invention.
9 and 10 are illustrations for explaining a keyword refinement process referred to in some embodiments of the present invention.
11 is an exemplary diagram illustrating priorities of daily keywords according to sources referred to in some embodiments of the present invention.
Figure 12 is an exemplary diagram illustrating an asset matching a keyword referenced in some embodiments of the invention.
13 is a flow chart of a keyword and asset price relevance evaluation method according to another embodiment of the present invention.
FIG. 14 is an exemplary view for explaining an asset corresponding to a keyword referred to in some embodiments of the present invention. FIG.
FIG. 15 is another exemplary diagram illustrating an asset corresponding to a keyword referred to in some embodiments of the present invention. FIG.
Figure 16 is an exemplary diagram illustrating the impact of a keyword on an asset referenced in some embodiments of the invention.
FIG. 17 is an exemplary diagram illustrating a difference between a point of time when a keyword is generated and an asset price change point, which is referred to in some embodiments of the present invention.
18 is an exemplary view illustrating a period in which a keyword referred to in some embodiments of the present invention affects an asset.
FIG. 19 is an exemplary diagram of keywords affecting assets among a plurality of keywords referred to in some embodiments of the present invention. FIG.
20 is an illustration of an asset that is affected by a plurality of keywords referenced in some embodiments of the present invention.
Figure 21 is an illustration of the relevance information of keywords and assets referenced in some embodiments of the invention.
Figure 22 is an illustration of a relevance index for an asset of a keyword referenced in some embodiments of the invention.
Figure 23 is an illustration of a graphical user interface (GUI) that is referred to in another embodiment of the present invention.
Figure 24 is an illustration of an investment guide interface based on when a keyword affects asset pricing, referred to in some embodiments of the invention.
25 is an illustration of an investment guide interface based on the degree to which a keyword affects an asset price, referred to in some embodiments of the present invention.
FIG. 26 is an exemplary diagram illustrating a daily keyword corresponding to an asset according to another embodiment of the present invention. FIG.
27 is a flowchart of a daily keyword extraction method corresponding to a price change of an asset according to another embodiment of the present invention.
28 is an example of a service for recommending another asset when a change in the price of the asset occurs according to another embodiment of the present invention.
FIG. 29 is a conceptual diagram for explaining a matching relationship of text contents, keywords, and assets according to another embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

Unless defined otherwise, all terms (including technical and scientific terms) used herein may be used in a sense commonly understood by one of ordinary skill in the art to which this invention belongs. Also, commonly used predefined terms are not ideally or excessively interpreted unless explicitly defined otherwise. The terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification.

1 is a conceptual diagram illustrating the relationship between an asset price and a keyword referred to in some embodiments of the present invention. Texts are circulated through web pages on various internet sites such as blog, internet news, messenger, SNS. The content of each text includes various issues. These issues can affect the value of various types of assets.

Referring to FIG. 1, for example, if text 1 is a post of a stock information blog, the text content may include stock information. If Issue 1 contained in the text content is a stock view for Company A, Issue 1 may affect the value of the stock of that particular company. It is also assumed that text 2 is the text on the web page of Internet news, and the text content includes the domestic real estate market view as an issue. If Issue 2 is information on real estate prices in Region B, Issue 2 may affect real estate prices.

At this time, the issue can be distributed on the Internet in the form of a keyword indicating the issue. In the above example, issue 1 can be expressed by keywords such as "A "," A company ", "A company stock ", and the like. Issue 2 can be expressed by keywords such as "B "," B real estate ", "B quotes ", and the like. With reference to the textual content and keywords described,

Hereinafter, the configuration and operation of the keyword and asset price relevance evaluation system according to an embodiment of the present invention will be described in detail.

2 illustrates an example of a keyword and asset price relevance evaluation system according to an embodiment of the present invention. For convenience of explanation, the keyword and asset price relevance evaluation system will be referred to as a system. Referring to FIG. 2, the system may include a service server 100, a user terminal 200, and an external device 300.

Here, the service server 100, the user terminal 200, and the external device 300 are connected to each other via the Internet. The service server 100 may be a server that stores various information and stores one or more programs for implementing embodiments of the present invention. User terminal 200 may be any of a server device, a fixed computing device such as a desktop PC, a mobile computing device such as a notebook, a smart phone, or a tablet PC. Also, the external device 300 may be a server device in which text content on the Internet is stored. Also, the external device 300 may be a server device in which price information on assets and assets is stored. For example, the external device 300 may be a stock information server of a stock exchange providing stock quotation information.

In the system according to the embodiment of the present invention, the service server 100 may collect text content posted on the first date from the external device 300 via the Internet. To this end, the service server 100 may store a web crawler that automatically searches for a web page. For example, the first date may be the date the service server 100 performs the crawl. In this case, the service server 100 may collect text content posted on the Internet until a predetermined time on the same day.

The service server 100 can extract keywords from each of the text contents collected by the crawler of the web crawler. Various algorithms can be used here for keyword extraction. The service server 100 may store at least one program for an algorithm used for keyword extraction. For example, the service server 100 may use a Latent Dirichlet Allocation (LDA) algorithm. The service server 100 can determine the subject of the text content using the LDA algorithm and extract keywords having high relevance to the topic.

When the keyword is extracted from each of the collected text contents, the service server 100 can combine the extracted keywords from various sources. The service server 100 may construct a keyword pool of the first date including the extracted keywords. The service server 100 may construct a keyword pool of the second date in the same manner. Here, the second date means a date different from the first date, and may be a close date belonging to a predetermined range from the first date.

The service server 100 can generate the daily keyword of the first date by using the comparison result of the keyword pool of the first date and the keyword pool of the second date.

The service server 100 may provide the generated daily keyword to the user terminal 200. [ In addition, the service server 100 may provide various services using a daily keyword to each of the plurality of user terminals 200. For example, the service server 100 can provide a plurality of user terminals 200 with an investment guide service of an asset using a daily keyword.

According to another embodiment of the present invention, the service server 100 may generate the daily keyword of the first date from the text content collected from the external device 300 via the Internet. In addition, the service server 100 may generate date-specific exposure frequency information for each keyword belonging to the daily keyword of the first date.

The service server 100 may compare the date-specific frequency information for each generated keyword with the date-specific price information of the registered asset. Accordingly, the service server 100 can determine the asset corresponding to each keyword among the assets previously registered.

The service server 100 may transmit the daily keyword of the first date and information on the asset corresponding to the keyword to the user terminal 200. [

3 is a block diagram of a service server 100 according to another embodiment of the present invention.

Referring to FIG. 3, the service server 100 may include a processor 101, a network interface 102, a memory 103, and a storage 104.

The processor 101 controls the overall operation of each configuration of the service server 100. The processor 101 may be configured to include a Central Processing Unit (CPU), a Micro Processor Unit (MPU), a Micro Controller Unit (MCU), or any type of processor well known in the art. The processor 101 may also perform operations on at least one application or program to perform the method according to embodiments of the present invention. The service server 100 may include one or more processors.

The network interface 102 supports wired / wireless Internet communication of the service server 100. In addition, the network interface 102 may support various communication methods other than Internet communication. To this end, the network interface 102 may be configured to include various communication modules.

The network interface 102 may collect text content from the external device 300 via the Internet. In addition, the network interface 102 may transmit and receive information on keywords and assets to / from the user terminal 200.

The memory 103 stores various data, commands and / or information. The memory 103 may also store one or more programs 105 for loading the one or more programs 105 from the storage 104 to perform automatic keyword generation using the text content in accordance with embodiments of the present invention, The above program can be saved. RAM is shown as an example of the memory 103 in Fig.

The storage 104 may non-temporarily store data received from the external device 300 or the like. The storage unit 140 may be a nonvolatile memory such as a ROM (Read Only Memory), an EPROM (Erasable Programmable ROM), an EEPROM (Electrically Erasable Programmable ROM), a flash memory and the like, a hard disk, a removable disk, And may include any type of well-known computer-readable recording medium.

The storage 104 may store one or more programs 105 for performing the methods according to embodiments of the present invention. The asset management software is shown as an example of the program 105 in FIG.

The storage 104 may be provided with a database 106 for keyword pools and daily keywords. In addition, a database 107 may be provided for the assets previously registered in the storage 104 and the assets corresponding to the respective keywords.

Although not shown, the service server 100 may include an input unit for receiving various settings and information, and an output unit for displaying information, in addition to the above configuration. The input unit and the output unit may be configured to include input means and output means of any type well known in the art to which the present invention belongs.

In this specification, the service server 100 may be referred to as a keyword and an asset price evaluation device in that it carries out the keyword and asset price evaluation method. In addition, the service server 100 may be referred to as a daily keyword automatic generation device in that it performs a daily keyword automatic generation method using text contents. Alternatively, the service server 100 may be simply referred to as a device.

Hereinafter, it is assumed that the methods according to the embodiments of the present invention are performed by the service server 100.

Hereinafter, other embodiments of the present invention will be described according to the method performed by the service server 100, based on the description of FIG. 1 to FIG. The embodiments described below are not necessarily separately performed, but can be implemented in combination with each other. It should also be noted that the embodiments described below can be implemented in combination with the embodiments of the present invention described in the description of Figs.

text Content  Automatic generation of daily keywords using

According to an embodiment of the present invention, the automatic daily keyword generating apparatus 100 may execute a method of automatically generating a daily keyword using text contents. Hereinafter, a method for automatically generating a daily keyword using text content will be described in detail with reference to FIGS.

4 is a flowchart of a method for automatically generating keywords using text contents according to an embodiment of the present invention. Referring to FIG. 4, the apparatus 100 may collect the text content of the first date through the Internet (S10). The apparatus 100 may extract a keyword from each of the text contents and configure a keyword pool of a first date including the extracted keywords (S20). The collection of text contents and the keyword extraction method are replaced with the description of FIG. The device 100 may use the textual content collected up to a predetermined reference time of the first date, for daily keyword generation of the first date.

Figure 5 is an illustration of a keyword pool, which is referenced in some embodiments of the invention. 6 is an exemplary diagram for describing a daily keyword, referred to in some embodiments of the present invention.

Referring to FIG. 5, the apparatus 100 may extract a keyword from the text content posted on the date D1 to construct a keyword pool 501. FIG. Also, the apparatus 100 may extract keywords from the text contents posted on the dates D2 and D3 to configure the keyword pool 502 and the keyword pool 503, respectively. The keyword pool may include a predetermined number of keywords. In addition, the keyword pool may exist in a list format in which the extracted number of keywords is listed in the order of many keywords. The keyword pool may be stored in the storage 104 of the device 100.

The device 100 may compare the keyword pool of the first date with the keyword pool of the second date (S30). In the above example, the apparatus 100 can compare the keyword pool 501, the keyword pool 502, and the keyword pool 503. The apparatus 100 generates the daily keyword of the first date using the comparison result (S40).

Here, the daily keyword is a keyword showing the frequency of exposure different from other dates among the text contents collected on the same day. That is, the daily keyword of the first date means a keyword that distinguishes the first date from another date. For example, when a special issue occurs on the first date, a large number of Internet searches for the issue occur, and a large number of mentions are exposed on the web page, the text content collected by the apparatus 100 may include the issue A plurality of keywords may be included. Accordingly, the apparatus 100 may construct a keyword pool including a keyword indicating the issue. The device 100 may generate a keyword that exists at a high rate in the keyword pool and a daily keyword of the first date.

In FIG. 6, a graph 600 showing the frequency with which keyword 1 (KW1) is exposed temporally on the text content is shown. It is assumed that the keyword 1 (KW1) is the "North Korea nuclear test" included in the keyword pool of FIG. It is also assumed that the date D2 is t1 shown in Fig.

Referring to FIGS. 5 and 6, the keyword 1 (KW1) shows a high exposure frequency on the date D2 as compared with other dates. In FIG. 5, the exposure frequency of the keyword 1 (KW1) is 17th in the date D1, the exposure frequency is 1 in the date D2, and 25th in the date D3. That is, the keyword "North Korea Nuclear Test" corresponds to the keyword that differentiates date D2 from date D1 and date D3.

In step S30, the apparatus 100 may compare the frequency of exposure of the keyword belonging to the keyword pool of the first date with the frequency of exposure of the keyword belonging to the keyword pool of the second date. In particular, the device 100 may determine a keyword having an exposure frequency difference equal to or greater than a threshold value as a daily keyword of the first date. It is assumed that the frequency of the keyword 1 (KW1) corresponding to the seventeenth rank in the keyword pool 501 is b1 in FIG. It is also assumed that the difference between the exposure frequency a0 and the exposure frequency b1 in Fig. 6 is the above threshold value.

Referring to FIG. 6, in the date D2, the frequency of exposure of the keyword 1 (KW1) is a1. At this time, the difference between a1 and b1 has a value larger than the difference between the threshold values a0 and b1. Accordingly, the apparatus 100 can determine the keyword 1 (KW1) as the day keyword of the date D2.

In order to ensure the accuracy of the daily keyword, various embodiments other than the method of comparing the keyword pools of the first date and the second date in step S30 may be used.

For example, the device 100 may determine the frequency of exposure for a particular date interval for each keyword belonging to the keyword pool of the first date configured in step S30. To this end, the device 100 may determine a first time window based on a first date. That is, the device 100 may compare the keyword pool constructed from the text content collected on the first date with the keyword pool constructed from the text content collected during the time in the first time window. Wherein the first time window may be a plurality of date intervals including the second date.

7 is an exemplary diagram illustrating a first time window on a daily keyword, referred to in some embodiments of the present invention. In Fig. 7, it is assumed that the first date is t1.

Referring to FIG. 7, in a graph 710, the device 100 may determine an interval including a date before t1 as a first time window 711 based on t1. Alternatively, as shown in graph 720, the device 100 may determine a first time window 721 as a date interval that includes t1, based on t1. The size of the first time window 711, 721 may be determined by the user or manufacturer of the device 100.

Referring to graph 710, the device 100 may measure the frequency of exposure by date in the first time window 711 for the keyword 1 (KW1) belonging to the keyword pool. That is, when the device 100 compares the keyword pool of the date in the first time window 711, the device 100 can measure the frequency of exposure of repeated keywords. The apparatus 100 can determine the keyword 1 (KW1) as the daily keyword of t1 when the exposure frequency at t1 and t2 is equal to or greater than the threshold value. On the other hand, if the frequency of exposure at t0 and t2 is less than the threshold value, the device 100 may determine that keyword 1 (KW1) is not a daily keyword of t2.

Referring to graph 720, the device 100 may measure the frequency of exposure by date in the first time window 721 for the keyword 1 (KW1). Apparatus 100 may compare the frequency of exposure of keyword 1 (KW1) at t2 and t3 with the frequency of exposure at t1, respectively. As a result of comparison, when the difference in the exposure frequency is equal to or greater than the threshold value, the device 100 can determine the keyword 1 (KW1) as the daily keyword of t1.

According to another embodiment of the present invention, the apparatus 100 can determine the daily keyword without considering the exposure frequency of the keyword 1 (KW1) described above.

For example, it is assumed that the keyword pool of the first date includes a new keyword. Here, the new keyword may include a keyword that is not included in the daily keyword pool of at least one other date belonging to the first time window. Alternatively, the new keyword may include a keyword that is published in a very small number of times within the first time window.

The device 100 may determine whether a new keyword exists in the keyword pool of the first date. In addition, the device 100 may determine whether the keyword is posted on the first date with the number of times equal to or greater than the reference value.

As a result of the determination, if there is a new keyword posted above the reference value, the device 100 may generate the daily keyword of the first date including the new keyword. Accordingly, for a new keyword that has not been posted on another day, the device 100 need not perform a process of comparing a keyword pool of the first date with a keyword pool of a different date.

The device 100 may generate the daily keyword of the first date by performing the above-described process for each keyword belonging to the keyword pool. That is, the daily keyword of the first date may include one or more daily keywords.

Figure 8 is an exemplary diagram for describing a second time window on a daily keyword, referred to in some embodiments of the present invention. In FIG. 8, it is assumed that the first date is t1.

The device 100 may determine a second time window as well as a first time window for determining the frequency of exposure of the keyword 1 (KW1) based on the first date. The second time window may be a date interval that includes more dates than the first time window.

The device 100 can determine the date interval before t1 as the first and second time windows based on t1. In the graph 810, the first time window 711 is a date interval from t1 to t0, and the second time window 811 is a date interval from t1 to t4 as an example.

Alternatively, the apparatus 100 may determine a first and a second time window 721 as a date interval including t1 based on t1. In graph 820, the first time window 721 is a date interval between t2 and t3 that includes t1, and the second time window 821 is a date interval between t5 and t6 that includes t1 It is shown as an example.

The device 100 can determine the number of times of publication of the keyword 1 (KW1) on the text content collected during the second time window 811, 821 and the number of times of publication of the keyword 1 (KW1) in the first time window 711, 721 have. In addition, the apparatus 100 may calculate a ratio between the number of times of publication. The device 100 can determine whether to remove each keyword included in the keyword pool of the first date based on the calculation result.

The device 100 leaves the keyword 1 (KW1) in the keyword pool when the ratio of the number of times of publication of the keyword 1 (KW1) during the second time window and the number of times of publication of the keyword 1 (KW1) during the first window is equal to or greater than the threshold value , And if it is less than the threshold value, the keyword 1 (KW1) can be removed from the keyword pool.

It is assumed that the threshold value is 0.7. For example, it is assumed that keyword 1 (KW1) has been posted 100 times during the second time window, and keyword 1 (KW1) has been posted 80 times during the first time window. In this case, the ratio of the number of times of publication becomes 0.8. Since the ratio 0.8 is greater than or equal to 0.7, the device 100 may leave keyword 1 (KW1) in the keyword pool. Here, it can be seen that the keyword 1 (KW1) is a keyword intensively posted during the first time window.

As another example, it is assumed that keyword 1 (KW1) was posted 100 times during the second time window and keyword 1 (KW1) was posted 20 times during the first time window. In this case, the ratio of the number of times of publication is 0.2. Since ratio 0.2 is less than 0.7, device 100 may remove keyword 1 (KW1) from the keyword pool. Here, it can be seen that the keyword 1 (KW1) is posted more than the first window section during the time period other than the first window.

In the above description, the case where the first time window and the second time window are the date interval is mainly described. According to another embodiment of the present invention, the first time window and the second time window may be time intervals rather than a date interval. In this case, it is assumed that, in steps S10 to S20, the device 100 constructs a keyword pool from text content collected up to a preset reference time of the first date.

For example, in the system of FIG. 1, when the user terminal 200 accesses the service server 100 at 2:00 PM, the device 100 serving as a service server can generate a daily keyword based on the access time . At this time, the device 100 may determine the first time window to be a time period up to two hours before the connection time of the first date. In this case, the device 100 can compare the keyword pool constructed from the text contents collected up to 2 hours before the connection time based on the keyword pool and the connection time.

The sizes of the first time window and the second time window described above may be determined by the user or manufacturer of the apparatus 100. Alternatively, the sizes of the first time window and the second time window may be adjusted by the user setting of the user terminal 200 receiving the service according to the embodiments of the present invention from the apparatus 100. Accordingly, according to the user setting of the user terminal 200, the daily keyword of the first date may be different. To this end, the device 100 may provide the user terminal 200 with a user interface for adjusting the sizes of the first time window and the second time window.

FIG. 9 is an example of a keyword refinement process using a first time window, and FIG. 10 is an example of a keyword refinement process using a first time window and a second time window. The effects of the first time window and the second time window will be described in detail with reference to Figs. 9 and 10. Fig. In FIGS. 9 and 10, it is assumed that the first time window is a date interval including the first date.

Referring to FIG. 9, if the first date is a date D7, the device 100 may construct a keyword pool 901 of the first date. The keyword pool 901 may include keywords such as "FTA entry," " professional baseball ", and "semiconductor new technology" In addition, the keyword pool 901 may include "alternative holidays" having a low frequency of exposure as keywords.

On the other hand, the keyword pool 902 of the date D8 may include keywords such as "professional baseball" and "alternative holidays" In addition, the keyword pool 902 may include "FTA validation" with a small frequency of exposure as a keyword.

The device 100 may compare the keyword pool 902 of the date D8 belonging to the first time window with the keyword pool 901 of the date D7. Referring to FIG. 9, the keyword "FTA effective" has a very high frequency of exposure at date D7, while the frequency of exposure at date D8 is very low. In this case, the apparatus 100 may compare the frequency of exposure to allow the keyword " FTA validation "to remain in the keyword pool 901 of date D7. Further, when the difference in the exposure frequency is equal to or greater than the threshold value, the apparatus 100 may determine the keyword " FTA validation "as the daily keyword 910 of the date D7. In the same way, the device 100 may determine the "semiconductor new technology" keyword as the daily keyword 910 of date D7.

On the other hand, in the case of the "professional baseball" keyword, the difference in the exposure frequency of the date D7 and the date D8 is not large. Thus, the device 100 may not include the "professional baseball" keyword as the day keyword of the date D7 by removing the "professional baseball" keyword from the keyword pool 901 of date D7.

If the first date is date D8, then device 100 may configure a keyword pool 902 for the first date. The device 100 may compare the keyword pool 902 of the date D7 belonging to the first time window with the keyword pool 902 of the date D8. Referring to FIG. 9, the keyword "alternative holidays" The device 100 may compare the frequency of exposure to allow the keyword "alternative holidays" to remain in the keyword pool 902 of date D8. , The device 100 may determine the keyword " FTA Effective "as the daily keyword 910 of date D8 if the difference in the exposure frequency is equal to or greater than the threshold value. On the other hand, The device 100 may not include the "professional baseball" keyword as the day-to-day keyword of date D8 by removing the "professional baseball" keyword from the keyword pool 902 of date D8. have.

In the above example, only the keyword pool of the date D7, which is the first date, and the date D8, which belongs to the first time window, are compared. However, the device 100 may store the keyword pool of the plurality of dates belonging to the first time window and the keyword pool of the first date May be compared.

The device 100 may determine a second time window, including the first time window referenced in the description of FIG. In FIG. 10, it is assumed that the second time window is determined as a date interval including the date D2 and the date D12.

If the first date is a date of D7, the device 100 will have a plurality of keyword pools including a keyword pool 1001 of date D2, a keyword pool 902 of date D8, and a keyword pool 1002 of D12, The keyword pool 901 can be compared.

The device 100 may determine the number of times the keyword " FTA Effective "and the" semiconductor new technology "keyword were posted during the first time window including date D7 and date D8. In addition, the device 100 may determine the number of times the keyword "FTA Enforcement" and "Semiconductor New Technology" keywords were posted during the second time window except for the first time window. Referring to FIG. 10, The device 100 determines the number of times of publication to remove the keyword " FTA effective "and the" semiconductor new technology "keyword from the keyword pool of date D7. Accordingly, the daily keyword 1010 of date D7 may not include the keyword " FTA effective "and the keyword" semiconductor new technology ".

On the other hand, in the case of the keyword "professional baseball ", a large number of times was posted during the first time window including the date D7 and the date D8, but the keyword was not included in the keyword pool. Therefore, the device 100 can determine the ratio of the number of times of posting so that the "professional baseball" keyword remains in the keyword pool of the date D8. Accordingly, the daily keyword 1010 may include a "professional baseball" keyword.

If the first date is date D8, results similar to the above example are derived. That is, the daily keywords of different days 1010 may include the same keywords. In addition, the daily keywords of successive dates may include the same keyword. For example, there may be cases where a particular issue continues to affect society for a significant period of time. In this case, the device 100 can extract the same keyword even on different dates using the second time window.

9 and 10, although the same keyword pools 901 and 902 are configured on the same date D7 and D8, the apparatus 100 may use different time keywords 910 and 1010 using the second time window, Can be generated.

11 is an exemplary diagram illustrating priorities of daily keywords according to sources referred to in some embodiments of the present invention. The device 100 may identify the source of the textual content collected on the first date. The device 100 may only identify the source of the textual content including the daily keyword of the first date.

Accordingly, the device 100 can determine the priority of the keywords belonging to the daily keyword of the first date, based on the identified source. The device 100 may determine the priority according to the attributes of the specifically identified source. Here, the attributes of the source may be determined according to the nature of the provider of the source, the media type of the source, the channel type of the source, and the like. For example, the provider's personality may be information about whether the source is a public agency, a private agency, or an individual. The source media type may be information about whether the source is economic media, sports media, or the like. In addition, the channel type of the source may be information about whether the source is Internet news, blog, SNS, or the like. Further, the attribute of the source may be determined based on the content of the text content identified by the keyword extraction algorithm described above.

In addition, the device 100 may identify different sections of the same source as different sources. For example, the device 100 may identify the entertainment news section on the Internet news and the political news section as different sources provided by the A press.

The apparatus 100 may assign different weights to the keywords according to the attributes. Referring to FIG. 11, the device 100 may store daily keyword information 1100 including weight information according to attributes of the source and the source, with respect to the daily keyword generated in step S30. Based on the keyword information 1100, the apparatus 100 may generate priority information 1110 of the keyword.

Referring to the keyword information 1100, for example, it is assumed that the weight A has a weight of 1, the weight B has a weight of 0.5, and the weight C has a weight of 0.3. At this time, the keyword 2 (KW2) may have a priority score of (25 * 1) + (20 * 0.5) = 35. The keyword 1 (KW1) has the number of posts 34 as a priority score. Also, the keyword 3 (KW3) has a priority score of DMS 50 * 0.3 = 15.

Accordingly, the device 100 may generate the priority information 1110. Referring to the priority 1110, the device 100 can determine the highest priority of the keyword 2 (KW2) having the highest priority score.

On the other hand, keywords obtained from different sources may have different impacts on the price of an asset even with the same keyword. Accordingly, the device 100 may need to identify the same keywords obtained from different sources as different keywords.

According to another embodiment of the present invention, when the keyword of one of the daily keywords of the first date is a keyword having a different source, the apparatus 100 may recognize the same keyword as a different keyword. Referring to FIG. 11, the apparatus 100 can identify that the source of the keyword 2 (KW2) is different from the 'addition day report' and the 'added sports'. In addition, since the media type of the source is different, the device 100 may determine that the attributes of the source are different. In this example, the device 100 may determine that the ' The keywords 2 (KW2) and the keyword 2 (KW2) of 'added sports' can be determined as different keywords.

Thus, even when the homonym is simultaneously generated as a daily keyword of the first date, the device 100 can identify the homonym as a different keyword.

In some embodiments of the invention, the asset to be matched with each keyword may be limited to the asset that is pre-matched to the source of the keyword. For example, entertainment-related sources (eg, entertainment on the Internet) could be pre-matched with entertainment-related stocks. At this time, keywords with entertainment-related sources will only be judged by the existence of correlation with entertainment-related stocks.

Figure 12 is an exemplary diagram illustrating an asset matching a keyword referenced in some embodiments of the invention. As described above, the device 100 may identify the source of the textual content that includes the daily keyword of the first date. In addition, the device 100 may match the asset corresponding to the daily keyword according to the source of the text content

In FIG. 12, a keyword 1 (KW1) 1201, a keyword 2 (KW2) 1202, and a keyword 3 (KW3) 1203 are shown as an example of a keyword belonging to a daily keyword of the first date.

It is assumed that the source of the text content including the keyword 1 (KW1) 1201 is a market forecast blog of IT stock ticker. In this case, the device 100 may match the keyword 1 (KW1) 1201 to the asset 1210. The asset 1210 may be a security of an individual company or a group of stocks belonging to a specific category.

It is assumed that the source of the text content including the keyword 2 (KW2) 1202 is an Internet magazine in which a new car marketizer is described. In this case, the device 100 may match the keyword 2 (KW2) 1202 to the asset 1220. Asset 1220 may be a securities of an automotive enterprise. Further, it is assumed that the source of the text content including the keyword 3 (KW3) 1203 is Internet-related caffeine related to real estate. In this case, the device 100 may match the keyword 3 (KW3) 1203 to the asset 1230. Asset 1230 may be a reconstruction pre-sale right for a particular area.

Up to now, embodiments of the present invention relating to a method for automatically generating a daily keyword using text contents performed by the automatic daily keyword generating apparatus 100 have been described. Hereinafter, embodiments related to a method of using the generated daily keyword will be described.

How to evaluate keyword and asset price relevance

In order to confirm the influence of the daily keyword generated in the above embodiment on the price of the asset, it is necessary to determine which asset corresponds to each daily keyword. Next, the impact of daily keywords on the corresponding assets should be analyzed. The method of determining the asset corresponding to the keyword and the method of analyzing the influence of the keyword will be clarified by the embodiments described below.

According to another embodiment of the present invention, the keyword and asset price relevance evaluation apparatus 100 can execute the keyword and asset price relevance evaluation method. Hereinafter, a keyword and an asset price relevance evaluation method performed by the keyword and asset price relevance evaluation apparatus 100 will be described in detail with reference to FIGS. 13 to 20. FIG.

13 is a flow chart of a keyword and asset price relevance evaluation method according to another embodiment of the present invention. 14 is an exemplary view for explaining an asset corresponding to a keyword referred to in some embodiments of the present invention.

Referring to FIG. 13, the apparatus 100 may collect text content posted on the first date via the Internet (S1301). The device 100 may extract a keyword from each text content to generate a daily keyword for the first date (S1302). As a specific method by which the device 100 generates the daily keyword of the first date, a method of automatically generating a daily keyword using the above-described text content can be used.

The device 100 may generate the daily frequency of exposure information for each keyword belonging to the daily keyword of the first date (S1303). For example, the exposure frequency information may be information representing the frequency of exposure of each keyword belonging to the generated daily keyword as a histogram. A graph 1400 is shown in FIG. 14 as an example of exposure frequency information. The exposure frequency information may include exposure frequency information by date of a predetermined date interval. Referring to Fig. 14, keyword 1 (KW1) has an exposure frequency N1 at date t1 and an exposure frequency N2 at date t2. In addition, the keyword 1 (KW1) has the exposure frequency N11 at the date t11 located between the dates t1 and t2.

The device 100 may compare the date-specific frequency information for each generated keyword with the price information for each day of the registered property (S1304). The device 100 can receive price information on the time and date of the assets and assets from the external device 300 of FIG. The device 100 may register the provided information in the storage 104. 14, price information by date of the assets 1 (ASSET1), 2 (ASSET2) and 3 (ASSET3) is shown as graphs 1401, 1402 and 1403, respectively, as an example of price information by date .

Referring to the graph 1401, Asset 1 (ASSET1) has price information P0 at date t1 and price P1 at date t2. Referring to graph 1402, Asset 2 (ASSET2) has price P1 at date t1 and price P2 at date t2. Also, Asset 2 (ASSET2) has a price P0 at date t11 located between dates t1 and t2. Referring to graph 1403, Asset 3 (ASSET3) has price P1 at date t1 and price P2 at date t2. Asset 3 (ASSET3) has a price P01 at date t11 located between dates t1 and t2.

The device 100 can determine the asset corresponding to each keyword among the previously registered assets by comparing the date-specific frequency information for each keyword with the price information for each day of the registered asset (S1305). That is, the device 100 may determine which asset is the asset corresponding to a particular keyword. At this time, the device 100 may identify an asset having a price change amount per date greater than or equal to a threshold value during the second period, corresponding to the frequency of exposure for each day during the first period for each keyword among the assets previously registered. The device 100 may determine the identified asset as an asset corresponding to each of the keywords.

Here, the first period is a period for measuring the frequency of exposure information of the keyword, and is a preset date interval. The second period may be a period during which the influence of the keyword is expressed in the asset, and the second period may be a period beginning with a predetermined time difference from the first period. This is because certain keywords may not have an immediate effect on changes in the price information of an asset. For example, if keyword A is generated as a daily keyword and keyword A is a keyword that affects asset A, the price of asset A may change after two days. Or the second period may be a period including the first period. If the particular keyword has an immediate effect on the change in the price information of the asset, the second period may be the same period as the first period. The length or starting point of the second period relative to the first period may be determined according to the setting of the user or manufacturer of the apparatus 100. [

The frequency-by-date frequency information of the keyword 1 (KW1) is compared with the price information by date of the asset 1 (ASSET1).

Referring to the graph 1400, the frequency of the keyword 1 (KW1) increases from N0 to N1 during t1 from a predetermined initial date which is the first period. Asset 1 (ASSET1) in graph 1401 continuously decreases in price from price P0 for a predetermined date after t1, which is the second period. Here, Asset 1 (ASSET1) may be an asset whose price decreases due to the influence of Keyword 1 (KW1). Again in the graph 1400, keyword 1 (KW1) decreases the frequency of exposure from N1 to N11 for dates from t1 to t11. For Asset 1 (ASSET1), prices continue to decline over the same date. The apparatus 100 can detect that the price of the asset 1 (ASSET1) is continuously falling while the frequency of exposure of the keyword 1 (KW1) increases or decreases. Accordingly, it can be determined that the asset 1 (ASSET1) is not influenced by the keyword 1 (KW1).

The frequency-by-date frequency information of the keyword 1 (KW1) is compared with the price information of the asset 2 (ASSET2) and the asset 3 (ASSET3) by date.

The trend of the frequency of the keyword 1 (KW1) by the date coincides with the trend of the price change of the asset 2 (ASSET2) and the asset 3 (ASSET3) by date when the graph 1400 and the graphs 1402 and 1403 are compared with each other. Accordingly, the apparatus 100 can determine that the asset 2 (ASSET2) and the asset 3 (ASSET3) have a price change amount by date corresponding to the frequency of exposure of the keyword 1 (KW1) by date. At this time, the apparatus 100 can determine the asset 3 (ASSET3) having the price change amount equal to or larger than the threshold value among the assets 2 (ASSET2) and 3 (ASSET3) as the asset corresponding to the keyword 1 (KW1). Since the price change of the asset may also be caused by an influence other than the keyword, the device 100 may determine that the asset 2 (ASSET2) having a price change less than the threshold value is an asset that is not affected by the keyword 1 (KW1).

Referring to the graph 1403, the amount of price change of the asset 3 (ASSET3) is similar to the exposure frequency of the keyword 1 (KW1). That is, when the frequency of exposure of the keyword 1 (KW1) increases, the price of the asset 3 (ASSET3) also increases. On the other hand, there may be an asset whose price decreases with increasing frequency of keyword 1 (KW1).

FIG. 15 is another exemplary diagram illustrating an asset corresponding to a keyword referred to in some embodiments of the present invention. FIG. Referring to FIG. 15, the apparatus 100 can identify the asset 3 (ASSET3) and the asset 4 (ASSET4) as an asset having a price change amount corresponding to the exposure frequency of the keyword 1 (KW1). At this time, the price change amount includes the absolute value of the price change amount. That is, when the graph 1400 and the graph 1501 are compared, the price change amount of the asset 4 (ASSET4) is negative while the exposure frequency of the keyword 1 (KW1) is positive. In this case as well, the apparatus 100 can determine the asset 4 (ASSET4) as an asset corresponding to the keyword 1 (KW1).

A plurality of keywords corresponding to one asset may exist. And may include a plurality of keywords. At this time, a method of determining a keyword having a high influence on a corresponding asset among a plurality of keywords will be described with reference to Fig.

Figure 16 is an exemplary diagram illustrating the impact of a keyword on an asset referenced in some embodiments of the invention. Here, it is assumed that the first date is t1 and the second date is t2. The second date is the date after the first date.

The device 100 can determine the keyword 1 (KW1) as the daily keyword of the first date t1 in step S1305. 16, the apparatus 100 compares the exposure frequency information 1400 of the keyword 1 (KW1) with the pricing information of the previously registered asset, and stores the asset 5 (ASSET5) as an asset corresponding to the daily keyword You can decide. Specifically, the apparatus 100 is an asset having a price change of more than a threshold value (P0 to P1) during the second period, corresponding to the increase in the exposure frequency of the keyword 1 (KW1) from NO to N1 during the first period Asset 5 (ASSET5) can be identified. In FIG. 16, it is shown as an example that the first period and the second period are both periods from a predetermined initial date to t1.

Thereafter, the device 100 may generate a daily keyword of the second date t2. The device 100 can determine whether the same keyword as any one of the keywords belonging to the daily keyword of the first date exists in the daily keyword of the second date. For example, if keyword 1 (KW1) of the first date is a "interest rate hike ", device 100 may determine whether there is a" rate hike "

As such, if the same keyword as any one of the daily keywords of the first date is present in the daily keyword of the second date, the apparatus 100 can monitor the price information for each date of the asset determined to correspond to the same keyword. At this time, the device 100 may monitor price information by date for a predetermined period based on the second date. In Fig. 16, the case where the predetermined period is the period from t1 to t2 is shown as an example.

In the above example, the apparatus 100 can monitor the price information of the asset 5 (ASSET 5) corresponding to the "interest rate raise" by date. A graph 1601 is shown as an example of price-by-date information of asset 5 (ASSET5) corresponding to "interest rate hike " in Fig. The device 100 determines that the price by date of the asset 5 (ASSET5) is equal to the price shown in the graph 1601, corresponding to the change in the frequency of exposure of the "interest rate hike" from N1 to N2 as shown in the graph 1400 The change from P1 to P2 can be monitored.

Based on the monitoring results, the device 100 may determine relevance information for asset 5 (ASSET5) of the "interest rate hike" keyword. Here, the relevance information may include whether the keyword "interest rate hike" affects the price change of the asset 5 (ASSET5) and the influence index on the asset 5 (ASSET5) price of the "interest rate hike" keyword.

The device 100 generates a rate of price change of the asset 5 (ASSET5) generated against the frequency of exposure of the daily keyword of the first date ("interest rate hike"), (ASSET 5) of the asset 5 can be measured. Based on the measured ratio, the device 100 can confirm the impact on the asset 5 (ASSET5) of the keyword 1 (KW1).

If the price change amount from t1 to t2 of the asset 5 (ASSET5) is equal to or larger than the threshold value as a result of the monitoring, the apparatus 100 updates the relevance information on the asset 5 (ASSET5) of the keyword " You may. In this example, if the difference between P1 and P2 in graph 1601 is greater than or equal to the threshold value, device 100 may upgrade the impact indicator for asset 5 (ASSET5) of the "interest rate hike" keyword. The relationship between the keyword "interest rate increase" determined on the basis of the first date t1 and the asset 5 (ASSET5) has been confirmed again on the basis of the second date t2.

Meanwhile. A case in which the keyword 2 (KW2) is determined as the daily keyword of the first date t1 in the same manner as the above example will be described. The device 100 may determine asset 5 (ASSET5) as an asset corresponding to keyword 2 (KW2). It is assumed that the keyword 2 (KW2) is "inflation." The device 100 may generate a daily keyword of the second date t2 and may identify it if the daily keyword of the second date includes "inflation. &Quot;

Next, the device 100 may monitor price-by-date information of the asset 5 (ASSET5) corresponding to the corresponding daily keyword of the first date. A graph 1602 is shown as an example of price-by-date information of asset 5 (ASSET5) corresponding to "price rise" in Fig. The device 100 will determine that the price by date of asset 5 (ASSET5) is equal to the price shown in graph 1602, corresponding to the change in the frequency of exposure of "inflation" from N1 to N2 as shown in graph 1600 It is also possible to monitor the change from P1 to P2. Accordingly, the apparatus 100 may confirm the influence of the keyword 2 (KW2) on the asset 5 (ASSET5).

The device 100 can determine the rank relationship for the asset 5 (ASSET5) of the keyword 1 (KW1) and the keyword 2 (KW2) based on the identified influence.

In the above, the influence of the keyword on the asset means the influence of each keyword belonging to the daily keyword on the price change of the corresponding asset. That is, the device 100 can determine how much the price of the asset is increased or decreased due to the influence of the keyword. The device 100 may store the determination result as relevance information.

The device 100 may estimate a change in price of an asset of another date based on the stored relevance information.

As described above, the device 100 may determine relevance information for an asset of an individual keyword belonging to a daily keyword. Hereinafter, various indexes of relevance information will be described.

FIG. 17 is an exemplary diagram illustrating a difference between a point of time when a keyword is generated and an asset price change point, which is referred to in some embodiments of the present invention. 18 is an exemplary view illustrating a period in which a keyword referred to in some embodiments of the present invention affects an asset.

17, it is assumed that the keyword 1 (KW1) is determined as a daily keyword on the first date t1 and the second date t2 as shown in the graph 1400. [ It is also assumed that the assets corresponding to the keyword 1 (KW1) are Assets 6 (ASSET6) and Assets 7 (ASSET7).

While the frequency of exposure of the keyword 1 (KW1) increases during the first period (t1 for a given initial date), the price of the asset 6 (ASSET6) in the graph 1701 is increased during the second period (t01 Respectively.

The apparatus 100 may determine a subsequent relationship between the first period and the second period. The device 100 may store the determination result as relevance information on the asset 6 (ASSET6) of the keyword 1 (KW1) which is the daily keyword of the first date. Referring to the graph 1701, the change in the price of the asset 6 (ASSET6) has a characteristic that precedes the occurrence date (first date) of the keyword 1 (KW1).

Thereafter, when the keyword 1 (KW1) is determined as the daily keyword of the second date t2, the device 100 determines the property that the price change of the asset 6 (ASSET6) precedes the second date based on the stored relevancy information .

In graph 1702, the price of asset 7 (ASSET7) has increased during the second period (t11 for a given initial date). In this case as well, the device 100 can determine the next-after relationship between the first period and the second period and store it as relevance information for the asset 7 (ASSET7) of the keyword 1 (KW1). Referring to the graph 1702, the change in the price of the asset 7 (ASSET7) has a characteristic that follows the generation of the keyword 1 (KW1).

Thereafter, when the keyword 1 (KW1) is determined as the daily keyword of the second date t2, the device 100 calculates the characteristic that the price change of the asset 7 (ASSET7) follows the second date based on the stored relevance information .

In addition, the apparatus 100 may measure the time interval between the first period and the second period. The apparatus 100 may store the measurement result as relevance information on the asset 7 (ASSET7) of the keyword 1 (KW1). Referring to the graph 1702, if the keyword 1 (KW1) is generated as a daily keyword in the first date t1, the price change of the asset 7 (ASSET7) affected by the keyword is shown at t11. The device 100 can determine that the influence of the keyword 1 (KW1) on the asset 7 (ASSET7) appears after the time interval t11-t1.

Thereafter, when the keyword 1 (KW1) is determined as the daily keyword of the second date (t2), the apparatus 100 calculates the price change of the asset 7 (ASSET7) based on the stored relevance information after the time interval (t21-t2) And the like.

In FIG. 18, it is assumed that the keyword 1 (KW1) is determined as a daily keyword on the first date t1 and the second date t2 as shown in the graph 1400.

During the first period (during t1 for a given initial date), the price of the asset in graph 1800 increased while the frequency of exposure of keyword 1 (KW1) increased. Also, the increased price was maintained during the second period E1.

The device 100 may store the second period in which the influence of the keyword 1 (KW1) on the price of the asset is maintained as the relevance information.

Thereafter, when the keyword 1 (KW1) is determined as the daily keyword of the second date t2, the apparatus 100 predicts that the asset price will have the property of being maintained for the second period E2 based on the stored relevancy information .

On the other hand, the daily keyword of the first date generated in step S1302 may include a plurality of keywords. It is assumed that the price of a specific asset rises from the first date corresponding to the exposure frequency information of a plurality of keywords. In this case, all of the plurality of keywords may affect the price of the asset. Or one of the plurality of keywords may not affect the price of the asset. Such a case will be described with reference to Figs. 19 and 20. Fig.

FIG. 19 is an exemplary diagram of keywords affecting assets among a plurality of keywords referred to in some embodiments of the present invention. FIG.

19, it is assumed that the keyword 1 (KW1) and the keyword 2 (KW2) of the graph 1900 are included in the daily keyword of the first date. Also, a graph 1910 is shown as an example of an asset determined to correspond to the keyword 1 (KW1) and the keyword 2 (KW2) in Fig.

In step S1305, the device 100 may determine that both the keyword 1 (KW1) and the keyword 2 (KW2) affect the asset based on the first date. At this time, when the keyword 1 (KW1) is determined as the daily keyword of the second date t2, the device 100 determines that one of the daily keywords of the first date is determined again as the day keyword of the second date t2 Can be identified.

The device 100 may monitor the price information by date of the asset determined during a predetermined period based on the second date t2.

In FIG. 19, a graph 1911 is shown as an example of price-by-date information of the asset. The apparatus 100 may determine relevance information on the property of the keyword 1 (KW1) based on the monitoring result. By comparing the graph 1910 and the graph 1911, the device 100 can determine that the relevance of the keyword 1 (KW1) to the asset is high.

On the other hand, when the keyword 2 (KW2) is determined as the daily keyword of the second date t2, the device 100 determines that the keyword of one of the daily keywords of the first date is determined again as the daily keyword of the second date t2 Can be identified. The device 100 may monitor the price information by date of the asset determined during a predetermined period based on the second date t2.

In FIG. 19, a graph 1912 is shown as an example of price-by-date information of the asset. The device 100 may determine the relevance information for the asset of the keyword 2 (KW2) based on the monitoring result. By comparing the graph 1910 and the graph 1912, the device 100 can determine the keyword 2 (KW2) as a keyword irrelevant to the asset. In this case, the device 100 may correct the result of determining that the keyword 1 (KW1) and the keyword 2 (KW2) all affect the asset based on the first date. That is, the device 100 may modify the daily keyword registered on the first date since the keyword 2 (KW2) is not related to the asset. At this time, the device 100 judges that there is an error in the keyword generation of the first date, and adjusts the size of the first window and the second window described in the embodiment of the method of automatically generating a daily keyword using the text content It is possible.

20 is an illustration of an asset that is affected by a plurality of keywords referenced in some embodiments of the present invention. The description of FIG. 19 and the repeated description thereof will be omitted.

In step S1305, the device 100 can determine that the keyword 1 (KW1) and the keyword 2 (KW2) both affect the asset based on the first date. The device 100 may then generate a daily keyword for the second date. If there is a keyword 1 (KW1) and a keyword 2 (KW2) again in the generated second date keyword, the apparatus 100 generates a keyword corresponding to the plurality of keywords (keyword 1 (KW1) and keyword 2 (KW2) And to monitor the price information by date of the determined asset.

The device 100 may determine relevance information for the determined assets of the plurality of keywords based on the monitoring results.

For example, assume that the daily keyword keyword 1 (KW1) and the keyword 2 (KW2) of the first date (t1) all affect the asset as shown in the graph 1910, as in the graph 1900. [

The device 100 may then generate the keyword 1 (KW1) as a daily keyword at the second date t2, as in the graph 1901. [ If the second day's day-to-day keyword does not affect the asset, such as graph 2001, the device 100 may determine that the asset is unrelated to the keyword 1 (KW1).

The device 100 can also determine that the keyword 2 (KW2) is also unrelated to the asset if the keyword 2 (KW2) does not affect the asset, such as the graph 2002. [

It is assumed that the price of the asset is increased by a threshold value or more when the keyword 1 (KW1) and the keyword 2 (KW2) are all included in another day as in the case of the daily keyword of the first date t1.

The device 100 determines whether or not the asset has a plurality of keywords, that is, keyword 1 (KW1) and keyword 2 (KW2) based on the price information of the assets of the first date (t1), the second date ), And it can be judged that they are not affected by individual keywords.

The device 100 may store the keyword 1 (KW1) and the keyword 2 (KW2) as pairing information for the property.

Specific embodiments of the keyword and asset price relevance evaluation device

According to the embodiments described above, the device 100 may determine the asset corresponding to the daily keyword and analyze the impact of the daily keyword on the asset. In particular, the device 100 may predict the impact of the same daily keyword on the second date on the price of the asset based on the relevance information of the asset corresponding to the daily keyword of the first date. Based on such a prediction, the device 100 may provide an investment guide service for the asset to the user terminal 200.

In order to provide the investment guide service, the device 100 may store information on the daily keywords and corresponding assets in the storage 104. The device may also store an analysis of the impact of the daily keyword on the corresponding asset. For example, the device 100 may store the relevancy information described above in the description of Figures 16-18.

Figure 21 is an illustration of the relevance information of keywords and assets referenced in some embodiments of the invention. 22 is an example of a relevance index for an asset of a keyword referred to in some embodiments of the present invention.

In Fig. 21, the relevance information (CR) data of the asset A corresponding to the daily keyword KW is shown as an example. The storage 104 of the device 100 may store the data. Referring to FIG. 21, the data may include daily-specific keyword (KW) information of a specific date and corresponding asset (A) information. In this case, the device 100 may store the relevance information CR based on each date on which the daily keyword KW was generated. Alternatively, the apparatus 100 may store the relevance information CR of the corresponding daily keywords KW based on the type of the asset A. In addition, the data may include priority information based on the origin of the keyword.

Each relevancy information CR in FIG. 21 may include information on a relevance index. The relevance index is information on the specific effect of the daily keyword on the corresponding asset.

Referring to FIG. 22, the relevance information CR includes information on the following relevance indexes.

The relevance information (CR) may include information about the impact of the daily keyword on the price increase and decrease of the asset. The relevance information CR is information on whether a specific daily keyword is generated and the price of the asset rises or falls in response to the daily keyword.

The relevance information (CR) may include information on the time difference in which the daily keyword affects the price change of the asset. The relevance information CR is information on whether a daily keyword is generated and the price of the asset fluctuates at a time interval corresponding to the daily keyword.

The relevance information (CR) may include information about the period during which the daily keyword affects the price change of the asset. The relevance information CR is information on a period in which the daily keyword is generated and the price of the asset continues to fluctuate corresponding to the daily keyword.

The relevance information (CR) may include influence information on the asset price of the daily keyword. The relevance information CR is information on how a daily keyword is generated and how the price of the asset increases or decreases according to the daily keyword.

The relevance information (CR) may include reliability information of the relevance index. The relevance information (CR) is information on the prediction accuracy of the asset price according to the relevance index. It is assumed that the stored relevance information exists for the daily keyword generated on the first date. At this time, when the same keyword as the daily keyword is generated on the second date after the first date, the apparatus 100 can determine whether the price change of the asset follows the same relevance information of the first date. The apparatus 100 may store the determination result in the relevance information CR of Fig. 22 as a relevance index.

Hereinafter, the investment guide service provided by the device 100 to the user terminal 200 using the relevance information will be described with reference to FIG. 23 to FIG.

Figure 23 is an illustration of a graphical user interface (GUI) that is referred to in another embodiment of the present invention.

Referring to FIG. 23, the device 100 may provide a graphical user interface 2300 for an investment guide service to a user terminal 200. The interface 2300 may include daily keyword information 2301 of the first date. The keywords generated from the text content collected on that day as an example of the daily keyword of the first date in the interface 2300 are shown.

When any one of the daily keywords 2302 is selected through the user terminal 200, the apparatus 100 can generate the interface 2310 corresponding to the selected keyword. In addition, the device 100 may provide an interface 2310 to the user terminal 200.

The interface 2310 may include the asset information 2311, 2312, and 2313 corresponding to the selected keyword. The interface 2310 may include one or more asset information 2311, 2312, and 2313 corresponding to the selected keyword. The interface 2310 may also include an interface 2314 for selecting asset information by type.

The device 100 can provide the investment guide service using the information on the relevance information and the relevance index of the keywords described in Figs. 21 and 22.

Figure 24 is an illustration of an investment guide interface based on when a keyword affects asset pricing, referred to in some embodiments of the invention.

Referring to FIG. 24, the apparatus 100 may provide the interface 2400 to the user terminal 200 when the keyword 2302 of the daily keyword is selected. 24, an example in which the interface 2400 includes stock quotation information 2401 of one or more companies corresponding to the keyword 2302 as asset information corresponding to the keyword 2302 is shown as an example.

If company A is selected through the user terminal 200, the device 100 may provide the interface 2410 to the user terminal 200. The interface 2410 may include the price information 2411 of the A corporation stock fluctuated corresponding to the keyword 2302. [ For example, the price information may be price change information of a company A stock for a predetermined period.

Interface 2410 may also include keyword 2302 and relevance information for A corporate stock. For example, the keyword 2302 may include information 2412 on how the stock price of the A company is affected. The interface 2410 may also include time difference information 2412, 2413 where the keyword 2302 affects the stock price of the A company. The interface 2410 may also include information 2413 on how long the keyword 2302 affects the stock price of the A company.

The device 100 may send the message 2414 to the user terminal 200 by providing the relevance information and providing an investment guide for the A corporate stock.

The message 2414 may include a recommendation for purchase or sale of A corporate stock. In addition, the message 2414 may include a guide to the purchase or sale of the A-memory stock, and the holding period based on the time difference information and the influencing period information.

25 is an illustration of an investment guide interface based on the degree to which a keyword affects an asset price, referred to in some embodiments of the present invention.

The device 100 can receive the keyword from the user through the user terminal 200 when the keyword is input. The device 100 may generate the interface 2500 corresponding to the keyword input from the user. The interface 2500 may include information 2502 about an asset corresponding to the input keyword 2501.

At this time, the device 100 can determine whether the input keyword coincides with any other day of the day or the day-of-the-day keyword. That is, when the keyword belonging to the daily keyword is inputted, the device 100 can identify the asset information corresponding to the input keyword. The device 100 may generate the interface 2500 based on the keyword and the identified asset information.

If any one of the information on the asset 2502 is selected, the device 100 may generate the interface 2510. [ In addition, the device 100 may send the generated interface 2510 to the user terminal 200.

The interface 2510 may include price information 2511 of the changed asset corresponding to the input keyword 2501 and relevance information. For example, information 2512 on whether the influence of the keyword 2501 precedes or follows the fluctuation of the asset price may be included. The interface 2510 may also include time difference information 2513 in which the keyword 2501 affects the price of the asset. The interface 2510 may also include information 2514 on the influence of the keyword 2501 on the price of the asset, that is, the influence of the keyword 2501 on the price fluctuation of the asset.

The device 100 may transmit the message 2514 to the user terminal 200 by providing the relevance information and providing an investment guide for the A corporation stock.

The message 2514 may include a recommendation for purchase or sale of an asset. The device 100 may generate target revenue information that is anticipated at the time of asset investment based on the influence information. In this case, the message 2514 may also include the target revenue information.

In the above description, the investment guide service provided from the device 100 when the keyword is selected or input from the user has been described. According to another embodiment of the present invention, the device 100 may provide keyword information to the user terminal 200 for assets the user is holding or is interested in. That is, the keyword is provided to the user for an asset that he / she is holding or is interested in so that the user can provide an opportunity to respond if the keyword is included in the daily keyword.

FIG. 26 is an exemplary diagram illustrating a daily keyword corresponding to an asset according to another embodiment of the present invention. FIG. Referring to FIG. 26, the device 100 may receive, via the user terminal 200, a selection input for an asset. To this end, the device 100 may provide the interface 2600 to the user terminal 200. [ The interface 2600 may include an asset list 2601. The asset list 2601 may include one or more assets 2602. If a user selection is received for any of the assets, the device 100 may generate an interface 2610 accordingly. In addition, the device 100 may provide the interface 2610 to the user terminal 200. The interface 2610 may include selected asset information 2602 and a corresponding keyword list 2611.

Meanwhile, the device 100 may store information on the user-selected asset. The device 100 can identify the keyword corresponding to the user-selected asset from the daily keyword of the first date. The device 100 may then generate a daily keyword of a second date. At this time, if there is a keyword corresponding to the user-selected asset in the day-time keyword of the second date, the apparatus 100 senses the keyword and transmits the keyword corresponding to the user-selected asset to the user terminal 200. Through this, the user can recognize that he or she is holding or that the price of the asset of interest may fluctuate. The device 100 may determine the likelihood of a change in the user-selected asset price based on the fact that the keyword is included in the daily keyword of the second date. The apparatus 100 may transmit an investment guide message to the user terminal 200 based on the determination.

27 is a flowchart of a daily keyword extraction method corresponding to a price change of an asset according to another embodiment of the present invention. 28 is an example of a service for recommending another asset based on the change in the price of the asset according to another embodiment of the present invention.

As described above, the keyword may not necessarily affect the corresponding asset in advance. That is, after the price change of the asset occurs, the keyword corresponding to the asset may be included in the daily keyword. Hereinafter, a description will be given of a method of identifying a keyword corresponding to a change in price of an asset when the change occurs in advance of the keyword. A method of identifying another asset whose price fluctuation is expected to follow the identified keyword will be described.

Referring to FIG. 27, the device 100 can identify an asset whose price change amount per date is equal to or greater than a threshold value during a first period of the pre-registered assets (S2701). In addition, the device 100 may collect the text content of the first date to generate a daily keyword of the first date (S2702). At this time, the step S2702 may not be executed after the step S2701. That is, the apparatus 100 may perform step S2702 separately from step S2701. Here, the first date may be the same day. That is, the device 100 may extract a keyword from the text content collected every day, and generate a daily keyword of the first date based on the extracted keyword.

The device 100 may detect an exposure frequency by date for a second period for each keyword belonging to the daily keyword of the first date (S2703). The device 100 may extract a keyword having an exposure frequency by date for a second period corresponding to the amount of price change per date during the first period of the asset (S2704). The device 100 may then extract the keyword from the daily keyword of the first date.

The apparatus 100 may determine the extracted keyword as a keyword corresponding to the asset (S2705).

Next, the device 100 may detect a price change over a threshold value for a third period of time for the asset. In addition, the device 100 can identify an asset corresponding to the determined keyword among the assets previously registered. The device 100 may send information about the identified asset to the user terminal 200. Accordingly, the apparatus 100 can predict the price fluctuation of the specific asset and the other asset based on the change in the price of the specific asset. In addition, the device 100 may provide an investment guide based on the prediction to the user terminal 200.

Referring to FIG. 28, the apparatus 100 may transmit an interface 2800 to the user terminal 200, which displays price change information of an asset registered in advance. The price change information 2801 of the asset may include price change information of the asset during the first period.

The device 100 can identify a keyword corresponding to any one of the assets based on the price change amount of any one of the assets included in the price change amount information 2801 of the asset.

Referring to the price change amount information 2801, the case where the price change amount of the corporate stock A is 20% and the price change amount of the corporate stock B is 5% is shown as an example in FIG. For example, if the threshold of the price change amount is 15%, the device 100 can identify a keyword having an exposure frequency corresponding to the A corporation stock.

The device 100 may store daily keyword information in advance for keyword identification. That is, the device 100 can identify the keyword among the previously stored daily keyword information. The device 100 may determine the identified keyword as a keyword corresponding to the asset.

The device 100 may generate the interface 2810 as it identifies the keyword and provide it to the user terminal 200. The interface 2810 may include an asset 2811 whose price has changed over a threshold value and keyword information 2812 judged to correspond thereto. The keyword information 2812 may include one or more keywords. The device 100 may determine a priority based on the source of the keyword and the interface 2810 may include priority information of the keywords 2813, 2814, and 2815. [

Next, the apparatus 100 may identify information about an asset corresponding to any one of the keywords 2813, 2814, and 2815 on the keyword information 2812. [ The identified asset may include an asset different from the asset 2811 whose price has changed beyond the threshold value. The device 100 may generate an interface 2820 that includes any one of the keywords 2813, 2814, and 2815 and the corresponding asset information 2821. The device 100 may send the interface 2820 to the user terminal 200.

Accordingly, when the price of a specific asset has a change amount greater than or equal to a threshold value, the apparatus 100 can provide the investment guide service for the other asset for which the price change is expected to the terminal 200 of the user.

text In content Matched  How to display asset information

In the above, embodiments of the present invention using relevance information between keywords and corresponding assets have been described. The relevance between the keyword and the asset may be extended to text content including the keyword. Hereinafter, this will be described in detail with reference to FIG.

FIG. 29 is a conceptual diagram for explaining a matching relationship of text contents, keywords, and assets according to another embodiment of the present invention. In FIG. 29, an Internet news is shown as an example of text content. According to the embodiments of the present invention described above, it is assumed that the keyword 2901 and the corresponding asset 2903 are matched and stored.

The user terminal 200 can display the Internet news 2905 on the display unit. The news 2905 may include at least one keyword. In FIG. 29, news 2905 is shown as an example in which keyword 1 (KW1), keyword 2 (KW2) and keyword 3 (KW3) are included.

The user terminal 200 can detect the keyword 1 (KW1), the keyword 2 (KW2) and the keyword 3 (KW3) on the news 2905 and extract it 2911. As the keyword is extracted, The terminal 200 can extract the assets 2913, 2923, and 2933 matched to each of the extracted at least one keyword 2911 among the assets previously registered.

29, Asset 1, Asset 2 and Asset 3 are shown as an example of an asset 2913 matched with keyword 1 (KW1) among the assets registered in advance. In addition, Asset 1, Asset 3 and Asset 4 are shown as an example with the keyword 2 (KW2) matched asset 2923. The assets 2933 matched to the keyword 3 (KW3) are Assets 3 and 5.

The user terminal 200 can identify the matched asset and extract it from previously registered assets. In addition, if there are assets extracted from the extracted assets more than a preset number of times, the user terminal 200 may match the extracted assets to the text content.

Referring again to FIG. 29, in the case of asset 3, extracted three times. For example, if the preset number of times is three, asset 3 may be matched to news 2905.

The user terminal 200 may display information on the extracted assets over a predetermined number of times in a second area different from the first area. That is, the information about the asset 3 may be displayed in an area different from the area where the news 2905 is displayed.

At this time, the asset matched to the at least one keyword includes an asset having a price change amount per date of more than a threshold value for a second period corresponding to the exposure frequency for each of the at least one keyword for each of the first periods . That is, the matched asset may be an asset matched to the keyword according to the embodiments of the present invention described above. The information on the asset may include price prediction information of the asset extracted over the predetermined number of times determined based on at least one keyword and relevance information between the assets extracted over the predetermined number of times. That is, the information on the asset here may include a result of prediction of fluctuation of the asset price of the asset according to the embodiments of the present invention described above.

As described above, if the text content and the asset are matched, the user may inquire related text content based on the specific asset through the user terminal 200. [

To this end, the user terminal 200 may display information about the asset in the first area on the display.

The information on the asset may be information on the price of the asset, price change, and the like provided through the web page according to a user's request.

The user terminal 200 may display a list of text content matched to the asset in a second area different from the first area. Here, the text content matched to the asset may include at least one keyword matched to the asset. For example, the list of textual content may be an Internet news article that includes keywords matched to the asset. Also, the list may be a list of the Internet news articles.

The user terminal 200 may display any one of the text contents when a selection input for one of the text contents on the displayed list is received. In this example, the user terminal 200 may display an Internet news article containing the keyword.

The methods according to embodiments of the present invention described above with reference to the accompanying drawings can be performed by the execution of a computer program embodied in computer readable code. The computer program may be transmitted from a first computing device to a second computing device via a network, such as the Internet, and installed in the second computing device, thereby enabling it to be used in the second computing device. The first computing device and the second computing device all include mobile computing devices such as a server device, a fixed computing device such as a desktop PC, a notebook, a smart phone, and a tablet PC.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

Claims (5)

In a method performed by a keyword and asset price relevance evaluation device,
Determining a subject for text content posted on the Internet and extracting a plurality of keywords related to the subject of the text content;
Extracting an asset matched to each of the plurality of keywords based on previously stored matching information between an asset and a keyword for each of the plurality of extracted keywords and calculating an extraction frequency of the extracted asset;
Associating the at least one asset with the text content when there is at least one asset that is commonly matched to each of the plurality of keywords and the calculated extraction count is equal to or greater than a predetermined value; And
Sending a list of textual content corresponding to the selected asset to the user terminal in response to receiving the selection of any one of the corresponding at least one asset from the user terminal,
A method of displaying asset information matched to textual content. The method according to claim 1,
Wherein the step of determining a subject for the text content and extracting a plurality of keywords related to the subject of the text content comprises:
Determining a subject of text content posted on the Internet using a Latent Dirichlet Allocation (LDA) algorithm, and extracting the plurality of keywords from the textual content;
A method of displaying an asset matched to textual content. The method according to claim 1,
Responsive to the selection of the textual content, transmitting information about the at least one asset corresponding to the selected textual content to the user terminal when a selection of the textual content from the user terminal is received.
A method of displaying asset information matched to textual content. In a method performed by a keyword and asset price relevance evaluation device,
Determining a subject for text content posted on the Internet and extracting a plurality of keywords related to the subject;
Extracting an asset matched to each of the plurality of keywords based on previously stored matching information between an asset and a keyword for each of the plurality of extracted keywords and calculating an extraction frequency of the extracted asset;
Associating at least one asset having the calculated number of times of extraction with a predetermined value or more with the text content; And
Transmitting the terminal screen to the user terminal in which the text content is displayed in a first area on a user terminal and information on an asset corresponding to the text content is displayed in a second area different from the first area; ,
A method of displaying asset information matched to textual content. 5. The method of claim 4,
Wherein the asset matched to each of the plurality of keywords includes:
An asset having an amount of price change per date greater than or equal to a threshold value for a second period corresponding to an exposure frequency for each of the plurality of keywords during a first period,
Wherein the information about the asset comprises price prediction information of the at least one asset determined based on relevance information between each of the plurality of keywords and the at least one asset,
A method of displaying asset information matched to textual content.

Images