TWI716172B - Critical information risk warning apparatus and method - Google Patents

Abstract

A critical information risk warning apparatus and method. The apparatus stores a plurality of target keywords, and each of the target keywords corresponds to a risk parameter. The apparatus receives a plurality of critical information, each of which corresponds to a listed company and an announcement time. The apparatus determines the risk parameter corresponding to each critical information according to the plurality of critical information and the target keywords. The apparatus counts each risk parameter of each critical information in a time interval that corresponding to each listed company, and generates a risk estimate of each listed company, respectively. The apparatus generates a risk alert when the apparatus determines that at least one of the risk estimates exceeds a threshold.

Description

Important information risk early warning device and method

The invention relates to a major information risk early warning device and method. Specifically, the present invention relates to a risk early warning device and method for judging risks based on important information.

In the current securities market under the open principle, the competent authority has strict information disclosure requirements for listed companies. The listed companies must disclose their own major information in accordance with the regulations. Investors can use the major information announced by each listed company, such as: Increase capital, issue new shares, announce the basis date of dividend distribution, resign independent directors, change accounting firms and other important information to understand relevant information, important schedules, and even operating conditions of each listed company.

Furthermore, among these large amounts of material information, there may be some signs (that is, unfavorable material information) that can reflect the potential risks of the listed company's operating conditions, which can be used as a reference for investors in making decisions. For example, one of the reasons why the company changes its accounting firm may be that the original well-known accounting firm is unwilling to endorse it, so it is transferred to a less well-known accounting firm; the company’s independent director resigns, and the company’s financial supervisor One of the reasons for the resignation may be that the relevant persons are unwilling to bear relevant legal responsibilities for risky decisions, and these potentially unfavorable major messages may cause the stock price of the listed company to fluctuate.

However, due to the large number of listed companies and the larger amount of significant information related to them, the field lacks a method that can assist in assessing the risks of each important information, and can collect, sort and analyze a large number of important information to calculate each in real time. The risk assessment of listed companies and the early warning mechanism.

In view of this, how to provide a risk early warning mechanism based on publicly significant information to remind investors of possible risks and to assist investors in responding to these high-risk listed companies in advance to reduce possible risks. Loss is an urgent goal for the industry.

1 One purpose of the present invention is to provide a major information risk early warning device. The important information risk early warning device includes a memory, a transceiver and a processor, and the processor is electrically connected to the memory and the transceiver. The storage stores a plurality of target keywords, and each target keyword corresponds to a risk parameter. The transceiver receives a plurality of important messages, each of which corresponds to a listed company and an announcement time. The processor determines the risk parameter corresponding to each of the major messages according to the major messages and the target keywords. The processor counts each of the risk parameters of the major messages corresponding to each of the listed companies in a time interval, and generates a respective risk estimate of each of the listed companies. The processor determines that when at least one of the risk estimates exceeds a threshold, a risk warning is generated.

Another object of the present invention is to provide an early warning method for major information risks, which is suitable for an electronic device. The electronic device stores a plurality of target keywords, and each target keyword corresponds to a risk parameter. The method for early warning of major information risk includes the following steps: receiving a plurality of important messages, each of which corresponds to a listing cabinet. Companies and an announcement time; determine the risk parameters corresponding to each of the major messages according to the major messages and the target keywords; calculate the corresponding major messages of each listed company in a time interval Each of the risk parameters is used to generate a risk estimate of each listed company; and it is judged that when at least one of the risk estimates exceeds a threshold, a risk warning is generated.

The major message risk early warning technology (including at least the device and method) provided by the present invention receives multiple major messages, determines the risk parameters of each major message based on the major messages and target keywords, and determines whether each listed counter company is in one Whether the risk estimate of the major information published in the time interval exceeds a predetermined value, when at least one of the risk estimates exceeds the predetermined value, a corresponding risk warning is generated. The present invention further provides a method for generating risk parameters corresponding to the target keywords. By analyzing past historical data (including stock price data and important information), the risk parameter weight distribution of the target keywords is determined. Through the foregoing operations, the important information risk early warning technology provided by the present invention analyzes a large number of important information, provides real-time risk assessment of each listed company and proposes an early warning mechanism, reminding possible risks to help investors deal with these in advance Over-risk listed companies will respond to reduce possible losses and solve problems that cannot be solved by conventional technology.

The detailed technology and implementation manners of the present invention are described below in conjunction with the drawings, so that those with ordinary knowledge in the technical field of the present invention can understand the technical features of the claimed invention.

The following will explain the major information risk early warning device and method provided by the present invention through implementations. However, these embodiments are not intended to limit the present invention to be implemented in any environment, application or method as described in these embodiments. Therefore, the description of the embodiments is only for the purpose of explaining the present invention, rather than limiting the scope of the present invention. It should be understood that, in the following embodiments and drawings, elements that are not directly related to the present invention have been omitted and are not shown, and the size of each element and the size ratio between the elements are only examples, and are not used to limit the present invention. The scope.

The first embodiment of the present invention is a major information risk early warning device 1, and its schematic structure is depicted in FIG. 1. The important information risk early warning device 1 includes a storage 11, a transceiver 13 and a processor 15, and the processor 15 is electrically connected to the storage 11 and the transceiver 13. The storage 11 can be a memory, a Universal Serial Bus (USB) disk, a hard disk, an optical disk, a flash drive, or a person with ordinary knowledge in the technical field of the present invention knows and has the same Any other storage medium or circuit for the function.

The transceiver 13 is an interface that can receive and transmit messages or other interfaces that can receive and transmit messages known to those skilled in the art to which the present invention pertains. The processor 15 may be various processors, central processing units, microprocessors, digital signal processors, or other computing devices known to those skilled in the art to which the present invention belongs. In some embodiments, the important information risk early warning device 1 can be set separately, or the important information risk early warning device 1 can be integrated into a transaction processing server, and the content of the important information risk early warning device 1 is not limited by the present invention.

In this embodiment, the storage 11 of the important information risk early warning device 1 pre-stores a plurality of target keywords (the definition and generation method of the target keywords will be described later). Then, a plurality of important messages are received, and the risk parameter corresponding to each important message is determined according to the important messages and the target keywords. Subsequently, by calculating each of the risk parameters of the major information corresponding to the listed companies in a time interval, the respective risk estimates of the listed companies are generated to determine when the risk estimates are When at least one of them exceeds the threshold, a risk warning is generated. It should be noted that the operation of the important information risk early warning device 1 still includes other relevant details, but the focus of the present invention is the calculation and analysis of important information and target keywords, so the following paragraphs will only describe the implementation details related to the present invention in detail .

In this embodiment, the storage 11 stores a plurality of target keywords (not shown), and each target keyword corresponds to a risk parameter. It should be noted that in the present invention, keywords that may appear in major messages that have an adverse impact on the company’s stock price/operations are defined as target keywords, and target keywords can be generated in a variety of ways (for example, by investors themselves) Decisions, professional judgments, results generated through machine learning analysis, etc.). For example, professionals judge that when target keywords such as "independent director change", "financial director change", "change of accounting firm" and other target keywords appear on major messages, it will have an adverse effect on the company's company stock price/operation. In some implementations, the target keywords may also be based on the investor's own investment strategy or investment habits, and each investor may determine the target keywords that he considers important through the operation interface. In addition, a plurality of target keywords can be directly received from the outside or directly generated by the important information risk early warning device 1. The present invention does not limit the source and generation method of the target keywords.

In this embodiment, the risk parameter is a risk estimate corresponding to each major message. Specifically, when a target keyword corresponds to a higher risk parameter, it means that the risk of the major message of the target keyword is higher. Conversely, when the target keyword corresponds to a lower risk parameter, it means that the The lower the risk of this important message of the target keyword. For ease of understanding, please refer to a specific example of multiple target keywords and their corresponding risk parameters shown in Figure 2A, but they are not intended to limit the present invention. In this specific example, the target keywords "resignation of independent director", "resignation of supervisor", "change of accounting firm", "change of financial supervisor" and "change of spokesperson" correspond to risk parameters of 12 points, 10 points, 10 points, 6 points and 2 points. It can be seen from the table that the risk of the target keyword "resignation of independent directors" in important messages is obviously higher than that of the target keyword "spokesperson change". In addition, the method of generating the risk parameters corresponding to the target keywords will be described later.

Then, the major message risk early warning device 1 receives a plurality of major messages through the transceiver 13, wherein each major message corresponds to a listed company and an announcement time. Specifically, material information is public information on the stock market, which is announced by each listed counter company. Material information can be obtained on a variety of public platforms (for example, the public information observatory provided by the Taiwan Stock Exchange). The present invention does not limit material information. The method of obtaining and the content should be understood by those with ordinary knowledge in the technical field of the present invention, so it is not repeated.

For ease of understanding, please refer to a specific example of a plurality of important messages shown in Figure 2B, but it is not intended to limit the present invention. In this specific example, it exemplifies that companies from C1, C2, C3, C4 and C5 announced on July 16, 2018 that their main thrusts were "Announcement of the company’s spokesperson change" and "Announcement of the resignation of independent directors of the company." ”, “Announcement of the Company’s Change of Accountant”, “Announcement of the Board of Directors Resolution of Dividend Distribution”, and “Announcement of the Dividend Distribution Base Date”. It should be noted that generally important messages still include the stock code, company name, speaking time, detailed content and other related data of the important message. For the convenience of illustration, the present invention only retains the content related to the calculation and analysis of the present invention, so The following paragraphs will only describe the implementation details related to the present invention in detail.

Subsequently, the processor 15 determines the risk parameter corresponding to each important message according to the received important message and the stored target keywords. Specifically, the processor 15 compares the subject of each important message with the target keywords (for example: relevance comparison), and when the subject of the important message matches a target keyword, the processor 15 15 can determine the risk parameter of the important message as the risk parameter corresponding to the target keyword.

For example, please refer to Figure 2A and Figure 2B at the same time. The processor 15 compares the important message from C1 "Announcement of the company’s spokesperson change" with the "speaker change" in the target keyword. Therefore, the processor 15 The risk parameter that determines the significant message from C1 is 2. It should be clarified that the terminology of the same event may be inconsistent due to different important messages (for example: change, change, resignation, resignation, etc.). Therefore, the correlation comparison performed by the processor 15 can be trained through a machine learning. During the training phase, it can be trained through a large number of labeled vocabularies of the same type, and trained and integrated according to the generated comparison result feedback. Deal with inconsistencies in terms. In this way, the accuracy of the comparison can be improved. Those with ordinary knowledge in the technical field to which the present invention pertains should be able to understand the algorithm of the comparison and how to train it through machine learning, so it will not be repeated.

In some embodiments, in order to increase the accuracy of the comparison, the processor 15 may also perform a word segmentation process (for example: JIEBA word breaker) on the subject/content of each important message to generate a corresponding word The result of a word segmentation of the message. Then, the processor 15 then performs a correlation comparison between each of the word segmentation results and the target keywords. In some implementations, the correlation comparison can also be achieved through algorithms such as BM25 (Best Match 25), TF/IDF (Term frequency—inverse document frequency), and machine learning, etc., which can also include denoising , Segmentation, word segmentation and other operations. It should be clarified why the details of word segmentation processing and the content of relevance comparison are not the focus of the present invention. Those with ordinary knowledge in the technical field of the present invention should understand the content, so it will not be repeated.

In this embodiment, after the processor 15 determines the risk parameter corresponding to each of the major messages, the processor 15 then counts the major messages corresponding to each of the listed companies in a time interval (ie, announces Time falls within this time interval of each of the risk parameters of the major information) to generate a risk estimate for each of the listed companies. Finally, the processor 15 determines that when at least one of the risk estimates exceeds a threshold, a risk warning is generated. Therefore, subsequent investors can adjust their investment strategies based on these risk warnings, or the risk control personnel/professional financial personnel of the investment company can make further professional judgments. In some embodiments, the time interval is the first N months from a current time point, and N is a positive integer.

Specifically, the processor 15 may sum up the risk parameters of the major information corresponding to the listed companies in the certain time interval (for example: within the past 6 months) to generate each The respective risk estimates of the listed companies. For example, please refer to Figure 3, which illustrates the statistical results of the processor 15 on the major information released by companies such as C1, C2, C3, C4, and C5 in the past 6 months (meaning that each company is The total risk parameters of major information released within 6 months), and the statistical results show that the respective risk estimates of C1, C2, C3, C4, and C5 are "12", "8", and "24". , "6" and "10". Therefore, when the threshold is set to 20, the processor 15 determines that the risk estimate of the C3 company has exceeded the threshold, and the processor 15 generates a risk warning for the C3 company.

The following paragraphs will detail the manner in which the processor 15 generates the risk parameter corresponding to the target keyword. Specifically, the processor 15 may analyze/statistically analyze historical data of major stock price crashes in the past to infer which major messages appearing more frequently may be relatively important, and give such major messages a higher risk weight. In some implementations, the risk parameter corresponding to each target keyword is calculated through the following operation: According to a preset rule and a historical stock price, a plurality of target listed companies (for example, : Process one target listed OTC company at a time); extract a plurality of historical important information corresponding to these target listed OTC companies in a first time zone (for example: N months) to generate a historical important information set; An occurrence number of each target keyword in the historical important information set; and based on the occurrence number, the risk parameter corresponding to each target keyword is determined. In some embodiments, the predetermined rule is the target listed companies that meet the stock price limit for M consecutive days, and M is a positive integer.

For example, if an investor wants to find a listed company whose share price has fallen limit for several consecutive days in historical data, and analyze the major information released by such companies in the past. Investors can use the "5 consecutive down limit for 5 days" as the default rule, and the processor 15 will search for listed companies with the "5 consecutive down limit for 5 days" in historical stock prices based on the default rule. Then, the processor 15 retrieves all the major information released by these listed companies in the 6 months from the first day's downward limit. Subsequently, the processor 15 counts the number of occurrences of major messages of the target keywords based on the target keywords described in the preceding paragraph (for example, "independent director resignation", "financial director change" and "spokesperson change"). Finally, the processor 15 can adjust the risk parameter corresponding to the target keyword with the highest number of occurrences (for example: "Resignation of Independent Director") to the highest and the second highest number of occurrences according to the number of occurrences of the target keywords. The risk parameter corresponding to the word (for example: "Finance Director Change") is adjusted to the second highest, the risk parameter corresponding to the target keyword with the lowest occurrence (for example: "Speaker Change") is adjusted to the lowest, and so on.

From the above description, it can be seen that the major information risk early warning device 1 provided by the present invention receives multiple major messages, determines the risk parameters of each major message according to the major messages and target keywords, and determines that each listed counter company is at a time Whether the risk estimate of the significant information published in the interval exceeds a predetermined value, when at least one of the risk estimates exceeds the predetermined value, a corresponding risk warning is generated. In addition, the important information risk early warning device 1 further provides a method for generating risk parameters corresponding to the target keywords, and determines the weight distribution of the risk parameters of the target keywords by analyzing past historical data (including stock price data and important information). Through the foregoing operations, the important information risk early warning device 1 analyzes a large number of important information, provides real-time risk assessment of each listed company and proposes an early warning mechanism, reminding possible risks to assist investors in anticipating these excessive risks. The listed companies have responded to reduce possible losses and solve problems that cannot be solved by conventional technology.

The second embodiment of the present invention is a major information risk early warning method, and the flowchart is depicted in FIG. 4. The important information risk early warning method is suitable for an electronic device, such as the important information risk early warning device 1 described in the first embodiment. The electronic device stores a plurality of target keywords, and each target keyword corresponds to a risk parameter, such as the plurality of target keywords in the first embodiment. The major information risk early warning method generates a risk early warning through steps S401 to S407.

In step S401, the electronic device receives a plurality of important messages, and each of the important messages corresponds to a listed company and an announcement time.

In step S403, the electronic device determines the risk parameter corresponding to each of the important messages according to the important messages and the target keywords. In some embodiments, the step of determining the risk parameter corresponding to each of the major messages includes: performing a word segmentation processing on each of the major messages to generate a word segmentation result corresponding to each of the major messages; and The word segmentation result is compared with the target keywords to determine the risk parameter corresponding to each of the important messages.

Then, in step S405, the electronic device calculates each of the risk parameters of the major messages corresponding to each of the listed companies in a time interval to generate a risk estimate of each of the listed companies. In some embodiments, the step of generating the risk estimate includes: adding up each of the risk parameters of the material information corresponding to each of the listed companies in the time interval to generate each of the listed companies The estimated value of this risk. In some embodiments, the time interval is the first N months from a current time point, and N is a positive integer.

Subsequently, in step S407, the electronic device determines that when at least one of the risk estimates exceeds a threshold, a risk warning is generated.

In some embodiments, the risk parameter corresponding to each target keyword is calculated through step S501 to step S507. In step S501, the electronic device determines a plurality of target listed companies that meet the preset rule according to a preset rule and a historical stock price. Then, in step S503, the electronic device captures a plurality of historical important messages corresponding to the target listed companies in a first time interval to generate a historical important message set. Subsequently, in step S505, the electronic device counts the number of occurrences of each target keyword in the historical important message set. Finally, in step S507, the electronic device determines the risk parameter corresponding to each target keyword based on the number of occurrences. In some embodiments, the preset rule is the target listed companies whose stock prices meet the M consecutive down limit, and M is a positive integer.

In addition to the above steps, the second embodiment can also execute all the operations and steps of the important information risk early warning device 1 described in the first embodiment, have the same functions, and achieve the same technical effects. Those with ordinary knowledge in the technical field to which the present invention pertains can directly understand how the second embodiment performs these operations and steps based on the above-mentioned first embodiment, has the same functions, and achieves the same technical effects, so it will not be repeated.

To sum up, the major information risk early warning technology (including at least the device and method) provided by the present invention receives multiple major messages, determines the risk parameters of each major message based on the major messages and target keywords, and judges each Whether the risk estimate of the major information announced by the listed company in a time interval exceeds the predetermined value, when at least one of the risk estimates exceeds the predetermined value, a corresponding risk warning is generated. In addition, the present invention further provides a method for generating risk parameters corresponding to the target keywords. By analyzing past historical data (including stock price data and important information), the risk parameter weight distribution of the target keywords is determined. Through the foregoing operations, the important information risk early warning technology provided by the present invention analyzes a large number of important information, provides real-time risk assessment of each listed company and proposes an early warning mechanism, reminding possible risks to help investors deal with these in advance Over-risk listed companies will respond to reduce possible losses and solve problems that cannot be solved by conventional technology.

The above-mentioned embodiments are only used to exemplify part of the implementation aspects of the present invention and explain the technical features of the present invention, rather than to limit the protection scope and scope of the present invention. Any change or equal arrangement that can be easily completed by a person with ordinary knowledge in the technical field of the present invention belongs to the scope of the present invention, and the scope of protection of the rights of the present invention is subject to the scope of the patent application.

1: Important information risk warning device 11: Storage 13: Transceiver 15: processor C1, C2, C3, C4, C5: company code S401~S407: steps S501~S507: steps

Figure 1 is a schematic diagram depicting the structure of the important information risk early warning device 1 of the first embodiment;

Figure 2A depicts a specific example of multiple targets and their corresponding risk parameters;

Figure 2B depicts a specific example of multiple important messages;

Figure 3 depicts a specific example of multiple listed companies and their risk estimates;

Figure 4 depicts a partial flowchart of the method for early warning of major information risks in the second embodiment; and

Figure 5 depicts a partial flowchart of the method performed by some embodiments.

S401~S407: steps

Claims (10)

A major message risk early warning device, comprising: a memory storing a plurality of target keywords, and each target keyword corresponds to a risk parameter; a transceiver, receiving a plurality of major messages, wherein each of the major messages corresponds to each To a listed cabinet company and an announcement time; and a processor, which is electrically connected to the memory and the transceiver, and performs the following operations: According to the major messages and the target keywords, determine the respective major messages Corresponding to the risk parameter; to count each of the risk parameters of the major information corresponding to each listed company in a time interval to generate a respective risk estimate of each listed company; and to determine when When at least one of the risk estimates exceeds a threshold, a risk warning is generated; the risk parameter to which each target keyword corresponds is calculated through the following operation: According to a preset rule and a historical stock price, it is judged to meet the forecast A plurality of target listed companies with rules; extract a plurality of historically significant messages corresponding to the target listed companies in a first time interval to generate a historically significant message set; count each target keyword in the historically significant The number of occurrences of the message set; and Based on the number of occurrences, the risk parameter corresponding to each target keyword is determined. For the major message risk early warning device described in claim 1, wherein the processor determines the risk parameter corresponding to each major message through the following operations: each major message is subjected to a word segmentation process to generate a corresponding to each major message A word segmentation result of the message; and a correlation comparison between each word segmentation result and the target keywords to determine the risk parameter corresponding to each of the important messages. The material information risk early warning device described in claim 1, wherein the processor generates the risk estimate through the following operations: summing up each risk of the material information corresponding to each listed company in the time interval Parameters to generate the respective risk estimates of the listed companies. The important information risk early warning device of claim 1, wherein the time interval is the first N months from a current time point, and N is a positive integer. The material information risk early warning device described in claim 1, wherein the preset rule is the target listed companies that meet the stock price falling limit for M consecutive days, and M is a positive integer. A major message risk early warning method, which is suitable for an electronic device, the electronic device stores a plurality of target keywords, and each target keyword corresponds to a risk parameter, the major message risk early warning method includes the following steps: receiving a plurality of Significant information, each of which corresponds to a listed company and an announcement time; According to the major messages and the target keywords, determine the risk parameters corresponding to each major message; calculate the risk parameters of the major messages corresponding to each listed company in a time interval to Generate a risk estimate of each listed company; and determine when at least one of the risk estimates exceeds a threshold, generate a risk warning; wherein each of the target keywords corresponds to the risk parameter Calculate through the following steps: according to a preset rule and a historical stock price, determine a plurality of target listed companies that meet the preset rule; extract a plurality of historically significant corresponding to the target listed companies in a first time interval Information, generate a historical important message set; calculate the number of occurrences of each target keyword in the historical important message set; and based on the number of occurrences, determine the risk parameter to which each target keyword corresponds. For the material information risk early warning method described in claim 6, wherein the step of determining the risk parameter corresponding to each of the material information includes: performing a word segmentation processing on each of the material information to generate a corresponding to each of the material information Word segmentation results; and a correlation comparison between each of the word segmentation results and the target keywords to determine the risk parameter corresponding to each of the major messages. For the material information risk early warning method described in claim 6, wherein the step of generating the risk estimate includes: adding up each of the risk parameters of the material information corresponding to each of the listed companies in the time interval to Generate the respective risk estimates of the listed companies. The method for early warning of major information risk as described in claim 6, wherein the time interval is the first N months from a current time point, and N is a positive integer. The material information risk early warning method described in claim 6, wherein the preset rule is the target listed companies that meet the stock price limit for M consecutive days, and M is a positive integer.

Images