KR20180066726A - Apparatus and method for estimating stock price - Google Patents

Abstract

Disclosed are an apparatus and a method for predicting a stock price. According to an embodiment of the present invention, the apparatus comprises: a keyword extraction unit for extracting one or more keywords from news generated in relation to a stock market; a keyword dictionary construction unit for constructing a keyword dictionary indicating the correlation between the extracted keyword and a stock price of a corresponding stock item changed after the occurrence of the news; an interest item selection unit for selecting an interest item by using one or more stock price trend prediction indices calculated from each stock index from a plurality of stock items; and a stock price up and down prediction unit for predicting up and down of a stock price of an interested item by using keywords extracted from new news and the previously constructed keyword dictionary when the new news is generated in relation to the selected interested item.

Description

[0001] APPARATUS AND METHOD FOR ESTIMATING STOCK PRICE [0002]

Embodiments of the present invention relate to techniques for predicting stock prices.

Modern society is welcoming the age of 100 because of the development of medical technology and the development of social welfare infrastructure. In fact, the age of 100 has become a reality, as insurance companies are releasing products that guarantee up to 110 years of age. However, due to rapid industrialization, no new jobs are created, and it is a reality that does not fit the age of retirement and the age of 100. If the social infrastructure can not keep pace with the development of science and technology in the future, this phenomenon will become more sophisticated.

Individuals who are living in a recent society should prepare themselves for the age of 100, and various tech-tech products are being launched to meet these demands. In the case of stocks, which are one of the most popular financial technologies, the number of individual investors is increasing due to online transactions like HTS and MTS. However, stocks tend to be negative investors because they are slower in information, lack funding power, and have less investment technology than institutions such as institutions.

Korean Patent Registration No. 10-0836066 (2008.06.02)

Embodiments of the present invention are intended to accurately predict fluctuations in stock prices.

According to an exemplary embodiment of the present invention, a keyword extracting unit extracts one or more keywords from news generated in connection with a stock market. A keyword dictionary construction unit for constructing a keyword dictionary indicating a relation between the extracted keyword and a stock price of a corresponding stock item changed after the occurrence of the news; An interest item selecting an item of interest using one or more stock price trend indicators calculated from each stock price index from a plurality of stock items; And a stock price fluctuation predicting unit for predicting fluctuation of the stock price of the item of interest using the keyword extracted from the new news and the pre-established keyword dictionary when new news is generated in relation to the selected item of interest, A prediction device is provided.

The keyword dictionary construction unit may update the keyword dictionary based on the keyword extracted from the new news and the stock price of the interest item changed by the generation of the new news.

The keyword extracting unit may extract the keyword from morpheme analysis from the news.

The keyword dictionary may include an up / down result of a stock price corresponding to each combination of one or more keywords.

The interest item selection unit may calculate different types of stock trend indicators for the stock items, and may select a stock item that indicates a rise in stock price index with a predetermined number or more of the stock trend indicators as an item of interest.

The stock price fluctuation predicting unit predicts the fluctuation of the stock price according to the fluctuation of stock price corresponding to the combination of the keywords extracted from the new news in the keyword dictionary.

The stock fluctuation predicting unit may predict the fluctuation of the stock price using a SVM (support vector machine) from the combination of the keyword dictionary and the extracted second keyword.

According to another exemplary embodiment of the present invention there is provided a method performed in a computing device having one or more processors and a memory storing one or more programs executed by the one or more processors, Extracting one or more keywords from the news; Constructing a keyword dictionary indicating a relation between the extracted keyword and a stock price of a corresponding stock item changed after the news is generated; Selecting an item of interest using at least one stock price trend index calculated from each stock index from a plurality of stock indexes; And a system for predicting a change in the stock price of the item of interest using the keyword extracted from the new news and the pre-established keyword dictionary when new news occurs in relation to the selected item of interest / RTI >

The constructing step may include updating the keyword dictionary based on the keyword extracted from the new news and the stock price of the item of interest fluctuated by the occurrence of the new news.

The extracting may extract the keyword from morpheme analysis from the news.

The keyword dictionary may include an up / down result of a stock price corresponding to each combination of one or more keywords.

The selecting step may calculate different types of stock trend indicators for the stock items, and may select a stock item showing an increase in the stock price index with a predetermined number or more of the stock trend indexes as the items of interest.

The predicting step may predict the fluctuation of the stock price in accordance with the result of the fluctuation of the stock price corresponding to the combination of the keywords extracted from the new news in the keyword dictionary.

The predicting step may predict the fluctuation of the stock price using a SVM (support vector machine) from the combination of the keyword dictionary and the extracted second keyword.

According to the embodiments of the present invention, it is possible to improve the accuracy of stock price prediction by primarily selecting the stocks of interest and analyzing the selected stocks in detail. Also, by selecting the stocks of interest first among the stocks of interest, the stock selection department of interest can save resources and time consumed in stock price prediction by stocks.

1 is a block diagram showing a detailed configuration of a stock price prediction apparatus according to an embodiment of the present invention;
2 is a flowchart for explaining a stock price prediction method according to an embodiment of the present invention.
3 is a block diagram illustrating and illustrating a computing environment including a computing device suitable for use in the exemplary embodiments.

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to provide a comprehensive understanding of the methods, apparatus, and / or systems described herein. However, this is merely an example and the present invention is not limited thereto.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. The following terms are defined in consideration of the functions of the present invention, and may be changed according to the intention or custom of the user, the operator, and the like. Therefore, the definition should be based on the contents throughout this specification. The terms used in the detailed description are intended only to describe embodiments of the invention and should in no way be limiting. Unless specifically stated otherwise, the singular form of a term includes plural forms of meaning. In this description, the expressions "comprising" or "comprising" are intended to indicate certain features, numbers, steps, operations, elements, parts or combinations thereof, Should not be construed to preclude the presence or possibility of other features, numbers, steps, operations, elements, portions or combinations thereof.

1 is a block diagram showing a detailed configuration of a stock price prediction apparatus 100 according to an embodiment of the present invention. 1, a stock price prediction apparatus 100 according to an embodiment of the present invention includes a keyword extracting unit 102, a keyword dictionary construction unit 104, an interest item selection unit 106, (108).

The keyword extracting unit 102 is a module for extracting keywords from news related to fluctuation of stock price or the like. According to one embodiment, the keyword extracting unit 102 may extract one or more keywords from the generated news related to the stock market. News is data collected from the stock market web, etc., and means all the information that can affect the stock price of stocks. Web sites related to the stock market refer to web pages provided by media companies, over-the-air broadcasting companies, cable broadcasting companies, portal site news, financial companies, and stock market related organizations. Stock items are items subject to stock trading. For example, news may relate to the merger, acquisition, or launch of a new company, but the type of company is not particularly limited. Also, the keyword is a key word extracted from the news, and it means a word that is important enough to affect the stock price fluctuation of the corresponding stock item. On the other hand, news refers to a wide range of information that can have a large and small impact on the stock market.

Specifically, the keyword extracting unit 102 may crawl news that has been started on a social media, a stock market related web, and the like for a set period (for example, the last 10 years). Crawling is a technology that collects documents that are stored on a large number of computers and includes them as an index to be searched. In addition, the keyword extracting unit 102 can extract a keyword from the crawled news.

According to one embodiment, the keyword extracting unit 102 can extract a keyword from news through morphological analysis. Specifically, the keyword extracting unit 102 can distinguish parts of words constituting a sentence included in news and detect nouns among the words. In addition, the keyword extracting unit 102 can extract a keyword related to the change in the actual stock price among the nouns. The keyword extracting unit 102 according to an embodiment can determine whether the compound noun and the compound noun constitute a keyword for each of the plurality of nouns constituting the compound noun when a compound noun is detected. For example, if the news includes the word 'Korea', the keyword extraction unit 102 may extract not only 'Korea' but also 'us' and 'country' by keywords.

The keyword extracting unit 102 can obtain not only the keyword but also the result of the fluctuation of the stock price after the news including the keyword is generated. That is, the keyword extracting unit 102 can grasp the effect of the news on the stock item. Specifically, the keyword extracting unit 102 may obtain information on how much the stock price of the news item related to the news has increased, decreased, or increased or decreased after the news is generated. For example, the keyword extracting unit 102 may collect market prices, stock prices, high prices, low prices, close prices, quotations, contracts, transaction amounts, transaction prices, transaction sources, upper limit prices, lower limit prices, have. Meanwhile, the keyword extracting unit 102 may extract a keyword using a regression analysis normalization model (for example, Elastic Net, Lasso normalization method, rugged normalization method, or the like).

The keyword dictionary construction unit 104 is a module for building a keyword dictionary from news that has occurred in the past. The keyword dictionary is intended to indicate the relationship between the keyword extracted from the news and the stock price of the corresponding stock item changed by the occurrence of the news. In other words, the keyword dictionary lists the impact of these keywords on the stock market if the keywords are included in the news. According to one embodiment, the keyword dictionary may include an up / down result of a stock price corresponding to each combination of one or more of the keywords. Table 1 shows contents included in the keyword dictionary according to one embodiment.

[Table 1]

As shown in Table 1, the keyword dictionary shows the result of the fluctuation of the stock prices according to a combination of a plurality of keywords. For example, when news containing keyword 1, keyword 3, and keyword 5 is generated, it indicates that the stock price of the stock related to the news has increased by one. According to one embodiment, the keyword dictionary construction unit 104 may assign a weight to each keyword in accordance with the degree of contribution to the stock fluctuation result. In addition, since the effect of the news that occurred in the past on the stock price may appear on the same day but may appear the next day, the keyword dictionary construction unit 104 constructs a keyword dictionary considering both the stock price of the day after the news occurs and the stock price of the next day can do. For example, the keyword dictionary construction unit 104 may assign the weighted values of 0.3 and 0.7 to the stock price of the same day and the next day, respectively, so as to indicate the up / down results of the stock price.

The keyword dictionary construction unit 104 may update the keyword dictionary based on the keyword extracted from the new news and the stock price of the interest item changed by the generation of the new news when new news occurs. The method for updating the keyword dictionary by the keyword dictionary update unit may be the same as the above-described construction process. That is, the keyword dictionary construction unit 104 can add the newly collected information to the keyword dictionary that is constructed in advance. In addition, the keyword dictionary construction unit may delete the keyword dictionary constructed in advance and build a new keyword dictionary based on the newly collected news and keywords.

The interest item selection section 106 is a module for primarily selecting stock items having a high probability of rising stock prices among a large number of stock items. According to one embodiment, the interested item selection unit 106 can select the item of interest using one or more (for example, four) stock price trend indicators calculated from each stock index from a plurality of stock items have. The stock price index can be an index that is created to understand the overall movement of the stock price. The stock price trend indicator can be the stock moving average, MACD (moving average convergence and divergence), Signal 9, or stochastic.

The stock price moving average line is a line connecting the stock moving average which is the arithmetic average of the stock price during a certain period. The MACD is the difference between short and long term exponential moving averages, using 12-day exponential moving average and 26-day exponential moving average. That is, MACD = 12-day exponential moving average ?? The 26-day exponential moving average is generally regarded as a rising signal when the MACD line crosses zero. Signal 9 is the 9-day exponential moving average of MACD, which shows trend trends better than MACD because it calculates the weighted MACD recently. The stochastic is a percentage indicator to see where the closing price is located in the stock price fluctuations over a period of time. When the share price is in an upward trend, the closing price is likely to be formed near the peak price, and when the price falls, the closing price is likely to be formed near the lowest price. In general, if the stochastic is above 80, the stock price is expected to rise.

According to one embodiment, the interest item selection unit 106 calculates different types of stock trend indexes for the stocks, and the stock indexes indicating the upward trend of the stock price indexes with a predetermined number or more, . Specifically, the interested item selection unit 106 can select an item of interest based on whether each stock price tendency prediction indicator predicts a rise or a fall of the stock price. In other words, the interested item selection unit 106 can select an item of interest using only the fluctuation indicated by a plurality of stock price trend indicators. For example, the interest item selection unit 106 calculates four stock price trend indicators for each stock item, and if two or more stock price trend predictors predict a stock price increase, It can be selected as an item. In addition, the interest item selection unit 106 may re-group the items of interest based on whether some of the stock trend predictors predict a rise.

In this way, the interested item selection unit 106 can save resources and time consumed in predicting the stock price of each stock item by primarily selecting the items of interest among the plurality of stock items.

The stock price fluctuation predicting unit 108 is a module for predicting stock price fluctuations in detail for each of the primary stocks of interest. According to an embodiment, when new news is generated in relation to the selected interest item, the fluctuation of the stock price of the item of interest can be predicted using the keyword extracted from the new news and the pre-built keyword dictionary. New news refers to news that has occurred since the keyword dictionary was built or updated. That is, the keywords included in the new news may be keywords that have not yet been reflected in the keyword dictionary.

Specifically, when new news related to the item of interest occurs, the keyword extracting unit 102 can extract keywords from the new news as described above. The stock price fluctuation predicting unit 108 can predict the fluctuation of the stock price of the interested stock using the keyword extracted from the new news and the keyword dictionary constructed in advance. Predicting stock price volatility may mean not only whether the stock price is increasing or decreasing but also forecasting the increase or decrease. According to an embodiment, the stock price fluctuation predicting unit 108 can predict the fluctuation of the stock price in accordance with the result of the fluctuation of the stock price corresponding to the combination of the keywords extracted from the new news in the keyword dictionary. For example, when keyword 1, keyword 3, and keyword 5 are extracted from new news, the stock fluctuation predicting unit 108 can predict that the share price of the interested stock increases by one.

On the other hand, the stock fluctuation predicting unit 108 can predict the fluctuation of the stock price using a SVM (support vector machine) from the combination of the keyword dictionary and the extracted second keyword.

According to the embodiments of the present invention, it is possible to improve the accuracy of stock price prediction by primarily selecting the stocks of interest and analyzing the selected stocks in detail.

FIG. 2 is a flowchart for explaining a stock price prediction method 200 according to an embodiment of the present invention. The method shown in FIG. 2 can be performed, for example, by the above-described stock prediction apparatus 100. FIG. In the illustrated flow chart, the method is described as being divided into a plurality of steps, but at least some of the steps may be performed in reverse order, combined with other steps, performed together, omitted, divided into detailed steps, One or more steps may be added and performed.

The keyword extracting unit 102 may extract one or more keywords from the news generated in connection with the stock market (S202). According to one embodiment, the keyword extracting unit 102 may extract the keyword from morpheme analysis from the news.

The keyword dictionary construction unit 104 may construct a keyword dictionary indicating a correlation between the extracted keyword and the stock price of the corresponding stock item changed by the occurrence of the news (S204). The keyword dictionary may include an up / down result of a stock price corresponding to each combination of one or more of the keywords. Also, the keyword dictionary construction unit 104 can update the keyword dictionary based on the keywords extracted from the new news and the stock price of the interest item changed by the generation of the new news.

The interested item selection unit 106 may select an item of interest using one or more stock price trend indicators that predict the fluctuation of each stock price from a plurality of stock items (S206). According to one embodiment, the interest item selection unit 106 calculates different types of stock trend indexes for the stocks, and the stock indexes indicating the upward trend of the stock price indexes with a predetermined number or more, .

The stock price fluctuation predicting unit 108 predicts the fluctuation of the stock price of the interested stock using the keyword extracted from the new news and the pre-established keyword dictionary when new news is generated in relation to the selected interested stock (S208). According to an embodiment, the stock price fluctuation predicting unit 108 can predict the fluctuation of the stock price in accordance with the result of the fluctuation of the stock price corresponding to the combination of the keywords extracted from the new news in the keyword dictionary. Also, the stock fluctuation predicting unit 108 can predict the fluctuation of the stock price using a SVM (support vector machine) from the combination of the keyword dictionary and the extracted second keyword.

3 is a block diagram illustrating and illustrating a computing environment 10 that includes a computing device suitable for use in the exemplary embodiments. In the illustrated embodiment, each of the components may have different functions and capabilities than those described below, and may include additional components in addition to those described below.

The illustrated computing environment 10 includes a computing device 12. In one embodiment, the computing device 12 may be a stock prediction device 100.

The computing device 12 includes at least one processor 14, a computer readable storage medium 16, The processor 14 may cause the computing device 12 to operate in accordance with the exemplary embodiment discussed above. For example, processor 14 may execute one or more programs stored on computer readable storage medium 16. The one or more programs may include one or more computer-executable instructions, which when executed by the processor 14 cause the computing device 12 to perform operations in accordance with the illustrative embodiment .

The computer-readable storage medium 16 is configured to store computer-executable instructions or program code, program data, and / or other suitable forms of information. The program 20 stored in the computer-readable storage medium 16 includes a set of instructions executable by the processor 14. In one embodiment, the computer-readable storage medium 16 may be any type of storage medium such as a memory (volatile memory such as random access memory, non-volatile memory, or any suitable combination thereof), one or more magnetic disk storage devices, Memory devices, or any other form of storage medium that can be accessed by the computing device 12 and store the desired information, or any suitable combination thereof.

Communication bus 18 interconnects various other components of computing device 12, including processor 14, computer readable storage medium 16.

The computing device 12 may also include one or more input / output interfaces 22 and one or more network communication interfaces 26 that provide an interface for one or more input / output devices 24. The input / output interface 22 and the network communication interface 26 are connected to the communication bus 18. The input / output device 24 may be connected to other components of the computing device 12 via the input / output interface 22. The exemplary input and output device 24 may be any type of device, such as a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touch pad or touch screen), a voice or sound input device, An input device, and / or an output device such as a display device, a printer, a speaker, and / or a network card. The exemplary input and output device 24 may be included within the computing device 12 as a component of the computing device 12 and may be coupled to the computing device 102 as a separate device distinct from the computing device 12 It is possible.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, . Therefore, the scope of the present invention should not be limited to the above-described embodiments, but should be determined by equivalents to the appended claims, as well as the appended claims.

100: stock price prediction device
102: Keyword extracting unit
104: Keyword dictionary construction unit
106: Interested item selection
108: Stock price fluctuation predicting section

Claims (14)

A keyword extracting unit for extracting one or more keywords from the news generated in connection with the stock market;
A keyword dictionary construction unit for constructing a keyword dictionary indicating a relation between the extracted keyword and a stock price of a corresponding stock item changed after the occurrence of the news;
An interest item selecting an item of interest using one or more stock price trend indicators calculated from each stock price index from a plurality of stock items; And
And a stock price fluctuation predicting unit for predicting fluctuation of the stock price of the item of interest using the keyword extracted from the new news and the pre-established keyword dictionary when new news is generated in relation to the selected interest item, Device.
The method according to claim 1,
Wherein the keyword dictionary construction unit updates the keyword dictionary based on a keyword extracted from the new news and a stock price of the interest item changed by the generation of the new news.
The method according to claim 1,
Wherein the keyword extracting unit extracts the keyword from morpheme analysis from the news.
The method according to claim 1,
Wherein the keyword dictionary includes an up-shift result of a stock price corresponding to each combination of one or more keywords.
The method according to claim 1,
Wherein the interest item selection unit calculates different types of stock trend indicators for the stock items and selects a stock item showing an increase in the stock price as a stock item of interest with a predetermined number or more of the stock trend predictions.
The method according to claim 1,
Wherein the stock price fluctuation predicting unit predicts the fluctuation of the stock price in accordance with a result of fluctuation of stock price corresponding to a combination of keywords extracted from the new news in the keyword dictionary.
The method according to claim 1,
Wherein the stock price fluctuation predicting unit predicts the fluctuation of the stock price using a SVM (support vector machine) from the combination of the keyword dictionary and the extracted second keyword.
One or more processors, and
A method performed in a computing device having a memory storing one or more programs executed by the one or more processors,
Extracting one or more keywords from the news generated in connection with the stock market;
Constructing a keyword dictionary indicating a relation between the extracted keyword and a stock price of a corresponding stock item changed after the news is generated;
Selecting an item of interest using at least one stock price trend index calculated from each stock index from a plurality of stock indexes; And
And a forecasting unit for predicting a change in stock price of the item of interest using the keyword extracted from the new news and the pre-built keyword dictionary when new news occurs related to the selected interest item.
The method of claim 8,
Wherein the constructing step includes the step of updating the keyword dictionary based on a keyword extracted from the new news and a stock price of the item of interest fluctuated by the generation of the new news.
The method of claim 8,
Wherein the extracting step extracts the keyword through morphological analysis from the news.
The method of claim 8,
Wherein the keyword dictionary includes an up / down result of a stock price corresponding to each combination of one or more keywords.
The method of claim 8,
Wherein the selecting step calculates each of the different types of stock trend predictive indexes for the stock stock and selects the stock stocks in which the stock trend predictive index with a predetermined number or more indicates the increase in stock price as the stock of interest.
The method of claim 8,
Wherein the predicting step predicts the fluctuation of the stock price in accordance with the result of the fluctuation of the stock price corresponding to the combination of the keywords extracted from the new news in the keyword dictionary.
The method of claim 8,
Wherein the predicting step predicts the fluctuation of the stock price using a SVM (support vector machine) from a combination of the keyword dictionary and the extracted second keyword.

Images