CN116226363A

CN116226363A - Reference information generation method and device

Info

Publication number: CN116226363A
Application number: CN202310504791.0A
Authority: CN
Inventors: 潘建东; 孙冰; 马张晖; 刘国杨
Original assignee: China Securities Co Ltd
Current assignee: China Securities Co Ltd
Priority date: 2023-05-08
Filing date: 2023-05-08
Publication date: 2023-06-06

Abstract

The embodiment of the invention provides a reference information generation method and device, which relate to the technical field of data processing, and the method comprises the following steps: obtaining event information of an event which has occurred; obtaining event days when the event determined based on the event information interferes with each security product in a preset industry; obtaining the profitability of each securities product and the securities trade market in an event period and the profitability of each securities product in an estimated period, and calculating the abnormal profitability of each securities product in the event period according to the obtained profitability; determining an interference period of the event interfering the industry in the event period according to the abnormal daily yield of each securities product in the event period; and determining reference information representing the interference degree of the event to the industry according to the event period and the interference period. By applying the reference information generation scheme provided by the embodiment of the invention, the reference information can be generated.

Description

Reference information generation method and device

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for generating reference information.

Background

The securities trade market is susceptible to interference from events occurring in real life, resulting in a stock exchange market surge. For this reason, when staff in a stock exchange provides services to clients, it is generally required to analyze the interference of events occurring in life to the market of the stock product in the industry, and provide a recommendation scheme of the stock product to the user based on the interference of the events to the market of the stock product in the industry.

Because staff analysis events often affect the market conditions of the industry where the certificate products are located, the staff analysis events are often influenced by age, experience and other artificial subjective factors, and therefore the accuracy of the recommended scheme is low. Therefore, it is desirable to provide a reference information generating scheme for generating reference information representing the interference degree of the event to the market of the industry where the certificate product is located, so as to be referred to by staff, thereby improving the accuracy of the recommended scheme.

Disclosure of Invention

The embodiment of the invention aims to provide a reference information generation method and device for providing reference information for a user. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a reference information generating method, where the method includes:

obtaining event information of an event which has occurred;

obtaining event days when the event determined based on the event information interferes with each security product in a preset industry;

obtaining the profitability of each securities product and the securities trade market in an event period and the profitability of each securities product in an estimated period, and calculating the abnormal profitability of each securities product in the event period according to the obtained profitability, wherein the event period and the estimated period are determined according to the event day, and the last day of the estimated period is the day before the first day of the event period;

Determining an interference period of the event interfering the industry in the event period according to the abnormal daily yield of each securities product in the event period;

and determining reference information representing the interference degree of the event to the industry according to the event period and the interference period.

In one embodiment of the present invention, the obtaining the event day when the event determined based on the event information interferes with each security product in the preset industry includes:

obtaining event information text related to the event information;

carrying out emotion type detection on each event information text to obtain emotion types corresponding to each event information text, wherein the emotion types comprise positive emotion types and negative emotion types;

determining, for each information release day on which an event information text is released, an emotion net value representing an overall emotion of information released within the information release day, based on the number of event information texts of different emotion types released at the information release day;

and determining the information release date corresponding to the emotion net value with the value larger than the target value as the event date when the event interferes with each security product in the preset industry.

In one embodiment of the present invention, when determining that a plurality of information release days corresponding to the absolute value of emotion whose value is greater than the target value are determined, the determining that the information release days corresponding to the absolute value of emotion whose value is greater than the target value is used as event days when the event interferes with each security product in the preset industry, the method includes:

determining each target information release day corresponding to the emotion net value with the value larger than the target value;

determining the earliest target information release day as the event day when the event interferes with each security product in the preset industry, and determining the target information release day as the event day if the interval between the target information release day and the previous target information release day exceeds the preset day aiming at each remaining target information release day.

In one embodiment of the present invention, the target value is determined based on a standard deviation of the net emotion value corresponding to each information distribution day.

In one embodiment of the present invention, the calculating an abnormal rate of return of each security product per day during the event period according to the obtained rate of return includes:

determining the income relation between each securities product and the securities trade market according to the income ratio of each securities product and the securities trade market in the estimated period;

And calculating the abnormal yield of each securities product in the event period according to the determined yield relationship, the yield of each securities product in the event period and the yield of the securities trade market in the event period.

In one embodiment of the present invention, after said calculating an abnormal rate of return for each security product per day during said event period in accordance with the obtained rate of return, the method further comprises:

and (3) carrying out standardized processing on the abnormal yield of each securities product in the event period every day aiming at each securities product to obtain the processed abnormal yield.

In one embodiment of the present invention, the determining the interference period of the event interfering with the industry in the event period according to the abnormal daily yield rate of each securities product in the event period includes:

determining an average abnormal rate of return of the industry per day during the event period according to the abnormal rate of return of each security product per day during the event period;

calculating a first interference parameter representing the interference degree of the event, which interferes with the industry every day in the event period, according to the determined average abnormal yield and the abnormal yield of each securities product every day in the event period;

Determining a date corresponding to a first interference parameter with a parameter value greater than a first parameter threshold as a date in an interference period when the event interferes with the industry;

the determining, according to the event period and the interference period, reference information characterizing the interference degree of the event to the industry includes:

and calculating the proportion of the days of the interference period to the event period as reference information for representing the interference degree of the event to the industry.

In one embodiment of the present invention, in a case where a plurality of event days are determined, the calculating the proportion of the days of the interference period to the event period as the reference information for representing the interference degree of the event to the industry includes:

calculating the proportion of the number of days of the interference period in the event period corresponding to each event day to the number of days of the event period corresponding to each event day;

and determining reference information representing the interference degree of the event on the industry according to the number of days corresponding to each event day.

In one embodiment of the present invention, in a case where a plurality of event days are determined, the determining, according to an abnormal daily yield rate of each security product in the event period, a disturbance period during the event period in which the event disturbs the industry includes:

For each event day, calculating the abnormal rate of return of each securities product in the estimated period corresponding to the event day according to the rate of return of each securities product and the securities trade market in the estimated period corresponding to the event day, determining the accumulated rate of return of each securities product in the event period corresponding to the event day according to the abnormal rate of return of each securities product in the event period corresponding to the event day, and determining a second interference parameter representing the interference degree of the event in the event period corresponding to the event day, wherein the second interference parameter represents the interference degree of the industry in the event period corresponding to the event day according to the calculated abnormal rate of return and the determined accumulated rate of return;

determining an event day corresponding to a second interference parameter with a parameter value larger than a second parameter threshold, and determining an event period corresponding to the determined event day as an interference period;

and calculating the quantity proportion of the interference period to the event period corresponding to each event day, and taking the quantity proportion as reference information for representing the interference degree of the event to the industry.

In one embodiment of the invention, the method further comprises:

Determining a text vector representing each event information text;

for each security product, obtaining association parameters representing the correlation between each event information text and the security product according to the determined text vector and the product vector representing the security product;

calculating the proportion of the interference period to the period of the event period;

and determining reference information representing the interference degree of the event to the industry according to the deadline proportion and the associated parameters.

In one embodiment of the invention, the product vector of the security product is determined according to the map information of the security product in a pre-constructed product knowledge map of the industry.

In one embodiment of the present invention, the determining a text vector representing each event information text includes:

determining entity keywords which belong to map entities contained in the product knowledge map and appear in at least one event information text, and determining the occurrence times of the entity keywords in each event information text;

and determining a text vector representing the text of each event information according to the word vector and the occurrence times of the entity keywords.

In a second aspect, an embodiment of the present invention further provides a reference information generating apparatus, where the apparatus includes:

the information acquisition module is used for acquiring event information of an event which occurs;

the event day acquisition module is used for acquiring event days when the event determined based on the event information interferes with various security products in a preset industry;

the system comprises a yield calculation module, a calculation module and a calculation module, wherein the yield calculation module is used for obtaining the yield of each securities product and the securities market in an event period and the yield of each securities market in an estimated period, and calculating the abnormal yield of each securities product in the event period according to the obtained yield, wherein the event period and the estimated period are determined according to the event day, and the last day of the estimated period is the day before the first day of the event period;

the interference period determining module is used for determining an interference period of the event interfering the industry in the event period according to the abnormal daily yield of each securities product in the event period;

and the information generation module is used for determining reference information representing the interference degree of the event on the industry according to the event period and the interference period.

In one embodiment of the present invention, the event day acquisition module includes:

A text obtaining sub-module for obtaining event information text related to the event information;

the emotion detection sub-module is used for detecting emotion types of the event information texts to obtain emotion types corresponding to the event information texts, wherein the emotion types comprise positive emotion types and negative emotion types;

the net value determining submodule is used for determining the emotion net value representing the overall emotion of the information released in each information release day according to the quantity of event information texts of different emotion types released in the information release day aiming at each information release day of releasing event information texts;

the first determining sub-module is used for determining information release days corresponding to the emotion net value with the value larger than the target value, and the information release days are used as event days when the event interferes with various security products in the preset industry.

In one embodiment of the present invention, in the case of determining that a plurality of information release days corresponding to the net emotion values with values greater than the target value, the date determination submodule is specifically configured to:

In one embodiment of the present invention, the yield calculation module is specifically configured to:

In one embodiment of the invention, the apparatus further comprises:

the standardized processing module is used for carrying out standardized processing on the abnormal rate of return of each securities product in the event period according to each securities product after calculating the abnormal rate of return of each securities product in the event period according to the obtained rate of return, so as to obtain the processed abnormal rate of return.

In one embodiment of the present invention, the interference period determining module includes:

the profit rate determination submodule is used for determining the average profit rate of the industry in the event period according to the daily abnormal profit rate of each securities product in the event period;

A parameter calculation sub-module for calculating a first interference parameter representing the interference degree of the event interfering the industry every day in the event period according to the determined average abnormal yield and the abnormal yield of each securities product every day in the event period;

a second determining submodule, configured to determine a date corresponding to a first interference parameter whose parameter value is greater than a first parameter threshold, as a date in an interference period when the event interferes with the industry;

the information generation module comprises:

and the proportion calculation sub-module is used for calculating the proportion of the days of the interference period to the event period and taking the proportion as reference information for representing the interference degree of the event to the industry.

In one embodiment of the present invention, in the case that a plurality of event days are determined, the proportion calculating submodule is specifically configured to:

In one embodiment of the present invention, in the case that a plurality of event days are determined, the interference period determining module is specifically configured to:

the information generation module is specifically configured to:

In one embodiment of the invention, the apparatus further comprises:

the vector determining module is used for determining a text vector for representing each event information text;

The parameter determining module is used for obtaining association parameters representing the relativity between the information texts of all events and the securities according to the determined text vectors and the product vectors representing the securities for each securities;

the information generation module is specifically configured to:

In one embodiment of the present invention, the vector determination module is specifically configured to:

In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;

A memory for storing a computer program;

a processor for implementing the method steps of any of the above first aspects when executing a program stored on a memory.

In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium having stored therein a computer program which, when executed by a processor, implements the method steps of any of the first aspects described above.

The embodiment of the invention has the beneficial effects that:

from the above, when the scheme provided by the embodiment of the invention is applied to generate the reference information, the abnormal rate of return of each securities product in the event period and the rate of return of the securities market in the estimation period can be calculated according to the rate of return of each securities product in the event period, so that the interference period of the event interfering industry is determined according to the calculated abnormal rate of return, if the interference period is longer, the higher the interference degree of the event to the industry is indicated, and if the interference periods are shorter, the lower the interference degree of the event to the industry is indicated, and the reference information for representing the interference degree of the event to the industry can be accurately determined according to the event period and the interference period. Therefore, by applying the reference information generation scheme provided by the embodiment of the invention, accurate reference information can be provided for a user.

Of course, it is not necessary for any one product or method of practicing the invention to achieve all of the advantages set forth above at the same time.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other embodiments may be obtained according to these drawings to those skilled in the art.

Fig. 1 is a flowchart of a first reference information generating method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of date division according to an embodiment of the present invention;

fig. 3 is a flowchart of a second reference information generating method according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of an emotion net value sequence provided by an embodiment of the present invention;

fig. 5 is a flowchart of a third reference information generating method according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of an event analysis system according to an embodiment of the present invention;

fig. 7 is a schematic structural diagram of a reference information generating device according to an embodiment of the present invention;

Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by the person skilled in the art based on the present invention are included in the scope of protection of the present invention.

Referring to fig. 1, fig. 1 is a flowchart of a first reference information generating method according to an embodiment of the present invention, where the method includes the following steps S101 to S105.

Step S101: event information of an event that has occurred is obtained.

The above event information may be understood as information of text, video, keywords, etc. for describing an event.

When event information of an event is obtained, event information of an event input by a user may be obtained, or information specified by the user in event information of various events that have been recorded may be obtained.

Step S102: and obtaining event days when the event determined based on the event information interferes with each security product in the preset industry.

The preset industry may be determined manually, such as a utility industry, a power equipment industry, a nonferrous metal industry, an electronic industry, and the like.

The security products may include stocks, options, futures, etc.

Specifically, in the first implementation manner, the event day may be determined through steps S102A-S102D in the embodiment shown in fig. 3 below based on the obtained event information when the event day is determined, which is not described in detail herein.

In a second implementation, a date input by the user based on the event information may be obtained as the event date.

Step S103: obtaining the profitability of each securities product and the securities trade market in the event period and the profitability in the estimation period, and calculating the abnormal profitability of each securities product in the event period according to the obtained profitability.

The estimated period is determined according to the event day, and the last day of the estimated period is the day before the first day of the event period.

Specifically, after determining the event date, the event period in which the event date is located and the estimated period between the event periods may be determined according to the event date.

For example, as shown in the schematic diagram of date division shown in fig. 2, t0 represents the event day, t1 represents the first day of the event period, t1 is a first preset number of days before t0 and spaced from t0, t1+

Represents the last day of the event period t1+.>

After t1, < >>

The second preset number of days is represented, the second preset number of days is larger than the first preset number of days, t2 represents the first day of the estimated period, and t2 and t1 are separated by a third preset number of days. Thus, the estimated period is in the date range of t2 to t1-1, and the event period is t1 to t1 +.>

Is a date range of (c).

After the estimation period and the event period are determined, the daily gain of each securities product in the estimation period, the daily gain of each securities product in the event period, the daily gain of the securities market in the estimation period and the daily gain of the securities market in the event period can be obtained, so that the abnormal daily gain of each securities product in the event period is calculated according to the obtained various gain.

Specific implementations of calculating abnormal yields of security products based on the various yields described above may be found in the subsequent embodiments and will not be described in detail herein.

Step S104: and determining the interference period of the event interference industry in the event period according to the abnormal daily yield of each securities product in the event period.

In particular, the specific implementation of determining the above-mentioned interference period may be referred to the subsequent embodiments, which will not be described in detail here.

Step S105: and determining reference information representing the interference degree of the event to the industry according to the event period and the interference period.

Specifically, the interference period is a part or all of the date in the event period, and the proportion of the interference period to the number of days in the event period can be determined according to the number of days in the event period and the number of days in the interference period and used as reference information for representing the interference degree of the event to the industry.

The reference information generating method of the first implementation mode is described below as applied to determining the time of day of an event.

In an embodiment of the present invention, referring to fig. 3, a flowchart of a second reference information generating method is provided, and in this embodiment, the above step S102 may be implemented by the following steps S102A-S102D.

Step S102A: event information text associated with the event information is obtained.

The event information text may include news, papers, comments, etc. for the event.

Specifically, various published information texts can be collected in advance, so that after the event information is obtained, event information texts related to the event information can be determined from the collected information texts.

For example, the event information text may be determined by either of the following two implementations.

In a first implementation, the event information may include keywords of the event, so that when determining the event information text, the event information text including the keywords of the event may be determined.

In a second implementation manner, various information texts collected in advance can be analyzed and classified, and after the event information is obtained, a classification type of the event is determined according to the event information, so that the information text in the classification type is determined to be the event information text.

Step S102B: and carrying out emotion type detection on each event information text to obtain emotion types corresponding to each event information text.

Wherein the emotion types include a positive emotion type and a negative emotion type.

Specifically, when emotion type detection is performed on each event information text, emotion type detection may be performed on the event information text using emotion detection algorithms, models, and the like.

For example, an emotion detection model, such as an ALBERT model, may be trained in advance, so that when emotion type detection is performed with an event information text, the event information text may be input into the model, thereby obtaining an emotion type output by the model.

In the emotion detection model training process, the sample information text data set can be divided into a training set, a verification set and a test set, the sample information text in the training set is used for training the model, the sample information text in the verification set is used for verifying whether the model after training is accurate, and the sample information text in the test set is used for testing the model after training is completed.

Step S102C: for each information release date of release event information text, determining emotion net value representing the overall emotion of the information released in the information release date according to the quantity of event information texts of different emotion types released in the information release date.

For simplicity of description, the number of event information texts belonging to a positive emotion type that are published within the same information publication day is referred to herein as a positive number, and the number of event information texts belonging to a negative emotion type is referred to herein as a negative number.

In determining the emotion net value, in one implementation, the positive number and the negative number may be subtracted to obtain a subtraction result, which is used as the emotion net value.

In another implementation, the ratio of the positive number to the sum of the positive number and the negative number may be calculated, or the ratio of the negative number to the sum of the positive number and the negative number may be calculated, and the calculated ratio is subtracted by 0.5 to obtain the subtraction result as the absolute emotion value.

Step S102D: and determining the information release date corresponding to the emotion net value with the value larger than the target value as the event date when the event interferes with each security product in the preset industry.

The target value may be set manually or determined based on the net emotion value corresponding to each date of release.

For example, the target value is a product of a standard deviation of the emotion net value corresponding to each information release day and a preset multiple.

The preset multiple may be determined by the user according to his own experience, for example, 2 times.

Because if the event interferes with the coupon product, the whole emotion of the information released on the event day will tend to be one of positive emotion and negative emotion, namely, the emotion net value corresponding to the event day is more extreme than the emotion net value corresponding to each information release day, and the standard deviation of the emotion net value corresponding to each information release day reflects the distribution condition of the emotion net value corresponding to each information release day, so that the target value is determined according to the emotion net value corresponding to each information release day, if the value of the emotion net value corresponding to the information release day is greater than the target value, the emotion net value is described as the extreme value in each emotion net value, namely, the emotion net value is more extreme than the emotion net value corresponding to each information release day, therefore, the accuracy of determining the event day can be improved according to the emotion net value corresponding to each information release day, and the accuracy of the generated reference information can be improved.

Specifically, after the target value is obtained, for each information release day, whether the value of the emotion net value corresponding to the information release day is greater than the target value can be judged, and if so, the information release day is determined to be the event day.

As shown in fig. 4, fig. 4 is a schematic diagram of a net emotion value sequence, wherein each value in the horizontal direction represents a time distribution, and each value in the vertical direction represents a net emotion value distribution. As can be seen from fig. 4, the absolute emotion value fluctuation is large between 22 nd year 2 and 9 nd 3 rd year 2022, and the intermediate value between 22 nd year 2 and 9 nd year 2022, that is, 1 nd 3 rd year 2022, can be taken as the event day, and the absolute emotion value corresponding to 2 nd year 2022 and 9 nd month 2 is also large, so that it can be determined that 2 nd month 2022 is another event day.

Therefore, according to the method and the device for generating the reference information, when the event interferes with the certificate product, the quantity difference of the event information corresponding to different emotion types released by the event day is large, so that the emotion net value corresponding to each information release day is determined according to the quantity of the event information texts of different emotion types released by each information release day, the event day when the event interferes with each security product can be accurately determined according to the value of the emotion net value, and the accuracy of the generated reference information can be improved.

In one embodiment of the present invention, if the step S102D determines that the plurality of information release days corresponding to the emotion net value with the value greater than the target value, the event day may be further determined from the determined information release days.

For example, when determining the event day, first determining each target information release day corresponding to the emotion net value with the value greater than the target value, determining the earliest target information release day as the event day when the event interferes with each security product in the preset industry, and determining the target information release day as the event day if the interval between the target information release day and the previous target information release day exceeds the preset day for each remaining target information release day.

The preset number of days may be manually determined, for example, 7 days, 8 days, etc.

For example, if four target information distribution days are determined, namely, 1 month and 1 day, 1 month and 8 days, 1 month and 12 days, and 1 month and 20 days, respectively, then when the event day is determined, the earliest target information distribution day may be determined as the event day, that is, 1 month and 1 day may be determined as the event day. In addition, the remaining three target information release days are separated from the previous target information release day by 6 days, 3 days and 7 days, and if the preset day is 7 days, the fourth target information release day can be determined as the event day, namely, the number 1 month and 20 are determined as the event day.

Therefore, when the scheme provided by the embodiment of the invention is applied to generate the reference information, the event can have discontinuous multiple interference on the certificate product, so that after each target information release date is determined, if the interval between two adjacent target information release dates is smaller, the event can be considered to generate the same interference on the certificate product between the two target information release dates, and if the interval between the two adjacent target information release dates is larger, the event can be considered to generate the two interferences on the certificate product, and therefore, each event date can be accurately determined according to the interval between each target information release date and the previous target information release date, thereby improving the accuracy of the generated reference information.

A specific implementation of calculating the abnormal yield of the security product in the above step S103 will be described below.

For example, the abnormal daily rate of return for each security product over the event period may be calculated by either of two implementations.

In a first implementation, a revenue relationship between a securities product and a securities market may be determined based on the securities product and the rate of return of the securities market in the estimated period; and calculating the abnormal yield of the securities products in the event period according to the determined yield relationship, the yield of the securities products in the event period and the yield of the securities trading market in the event period.

Specifically, the linear regression can be performed on the yield of the securities product and the securities market in the estimated period to obtain the linear relationship between the yield of the securities product and the yield of the securities market, which is used as the yield relationship between the securities product and the securities market. After the income relation is obtained, the expected income ratio of the securities product in the event period can be calculated according to the income relation and the daily income ratio of the securities trade market in the event period, so that the actual income ratio of the obtained securities product in the event period is subtracted from the expected income ratio, the abnormal income ratio of the securities product in the event period can be accurately obtained, the reference information can be generated according to the more accurate abnormal income ratio of the securities product, and the accuracy of the generated reference information can be improved.

In one embodiment of the present invention, the linear relationship between the security product yield and the security trade market yield may be expressed according to the following expression:

wherein, the liquid crystal display device comprises a liquid crystal display device,

representing the profitability of the securities product i +.>

Representing the profitability of the securities trade market m, +.>

Intercept representing the above-mentioned linear relationship, +.>

Slope representing the above linear relationship, +.>

、/>

And calculating according to the yield of the securities products and the securities trade market in the estimated period.

The abnormal daily rate of return of a security product during an event period can be calculated by the following expression:

representing the abnormal rate of return of a security product i on a date t during an event, < >>

Representing the profitability of a security product i at a date t during an event,/>

Representing the profitability of the securities trade market m at the date t during the event.

In a second implementation, the daily rate of return of the securities product during the event period may be predicted based on the rate of return of the securities product and the securities market during the estimated period, and a first difference between the actual daily rate of return of the securities product during the event period and the predicted rate of return may be calculated as an abnormal rate of return of the securities product during the event period.

In addition, the daily income ratio of the securities trade market in the event period can be predicted, a second difference value between the actual daily income ratio and the predicted income ratio of the securities trade market in the event period is calculated, and the first difference value of the same day is adjusted according to the calculated second difference value to obtain an adjusted first difference value which is used as the abnormal income ratio of the securities product in the event period.

There is an autocorrelation effect on the rate of return of a security product on an adjacent date, i.e. the rate of return of a security product on a day is affected by the rate of return of the security product on the day before. Since the security products have an autocorrelation influence of the yield, the accuracy of the abnormal yield of the security products calculated based on the yield of the security products is low.

In order to eliminate the autocorrelation influence of the security products, in one embodiment of the present invention, after calculating the abnormal rate of return of each security product in an event period, the abnormal rate of return of each security product in an event period may be normalized to obtain a processed abnormal rate of return.

Specifically, the standard deviation of the abnormal yield of the securities product in the event period can be calculated, the abnormal yield of the securities product in the event period in each day is divided with the calculated standard deviation, and the divided standard abnormal yield is obtained and used as the processed abnormal yield.

In one embodiment of the present invention, the abnormal rate of return of a coupon product per day during an event period may be normalized by the following expression:

standard abnormal yield representing the date t of securities product i during the day of the event,/>

Representing the existence of a security product iAbnormal yield of date t in the piece day, +.>

Representing the standard deviation of the abnormal yield of the security product i.

From the above, when the scheme provided by the embodiment of the invention is applied to generate the reference information, after calculating the daily abnormal yield of the security product in the event period, the abnormal yield of the security product in the event period is standardized, so that the influence of the autocorrelation of the security product on the abnormal yield is eliminated, the accuracy of the abnormal yield of the security product can be improved, and the accuracy of the generated reference information can be improved.

A specific implementation manner of determining the interference period in the step S104 is described below.

The interference period is a part or all of an event period, and the event period is determined according to an event day. It may be appreciated from the foregoing embodiments that there may be one or more event days, and thus one or more event periods, so that when determining the interference period, different implementations may be used to determine the interference period for different situations of the event period.

For example, in the case where an event day is determined, an average abnormal rate of return of the industry per day during the event period may be determined based on the abnormal rates of return of the individual securities products per day during the event period; according to the determined average abnormal yield and the abnormal yield of each securities product in the event period every day, calculating a first interference parameter which corresponds to each date in the event period and represents the interference degree of the event interference industry; and determining the date corresponding to the first interference parameter with the parameter value larger than the first parameter threshold as the date in the interference period of the event interference industry.

The first parameter threshold may be a manually determined threshold.

Specifically, an average value of the abnormal yields of the securities products on the same day in the event period can be calculated as the average abnormal yields of the industry on the same day. After calculating the average daily abnormal yield of the industry in the event period, calculating a first interference parameter representing the interference degree of the event to the industry in the event period according to the calculated average abnormal yield and the daily abnormal yield of each securities product in the event period.

In calculating the first interference parameter, in one implementation manner, according to each average abnormal yield and each abnormal yield of securities products in each day in the event period, the first interference parameter may be calculated by the following expression:

first disturbance parameter corresponding to date t in the event period,/->

Mean abnormal yield representing the date t of the industry during the event,/->

The standard abnormal yield of the date t of the securities product i in the event period is represented, namely the abnormal yield of the date t of the securities product i in the event period after standardized processing is carried out, and N represents the quantity of each securities product in the industry.

In another implementation, the first interference parameter may be calculated by using a parameter checking algorithm according to each average abnormal rate of return and each abnormal rate of return of securities products in each day during the event period.

After the first interference parameters corresponding to each day in the event period are calculated, as the first interference parameters represent the interference degree of the event interference industry, the larger the parameter value of the first interference parameters is, the larger the interference degree of the event interference industry is, the smaller the parameter value of the first interference parameters is, the smaller the interference degree of the event interference industry is, therefore, according to the parameter values of the first interference parameters, the first interference parameters with the parameter values larger than the first parameter threshold value are determined, the date corresponding to the determined first interference parameters can be accurately determined as the date in the interference period of the event interference industry, and after all the first interference parameters are traversed, the date range formed by the determined date can be accurately determined as the interference period.

In addition, after the interference period in the event period is determined, when the reference information for representing the interference degree of the event to the industry is determined according to the event period and the interference period, the proportion of the interference period to the number of days of the event period can be calculated and used as the reference information for representing the interference degree of the event to the industry.

The interference degree of the event in the interference period is stronger, if the interference period is longer, the interference degree of the event on the whole industry is stronger, so that the number of days of the interference period in the event period is calculated, the number of days can accurately reflect the interference degree of the event on the whole industry, and therefore the number of days of the interference period in the event period is taken as reference information, and the accuracy of the reference information can be improved.

For another example, in the case of determining a plurality of event days, the interference period may be determined by any one of the following two implementations, and after the interference period is determined, reference information characterizing the interference degree of the event to the industry is determined according to the interference period and the event period.

In a first implementation manner, one event day corresponds to one event period, and a plurality of event days correspond to a plurality of event periods, so that for each event day, an interference period in the event period corresponding to the event day can be determined, after the interference period in the event period corresponding to each event day is determined, the number of days proportion of the interference period in the event period corresponding to each event day in each event period can be calculated, and reference information representing the interference degree of the event to the industry is determined according to the calculated number of days proportion.

Specifically, when determining the interference period in the event period corresponding to each event day, the interference period may be determined by adopting the manner of determining the interference period under the condition of one event day, which is not described herein.

After the interference period in the event period corresponding to each event day is determined, the number of days proportion of the interference period in the event period corresponding to each event day to each event period can be calculated, the number of days proportion corresponding to each event day reflects the interference degree of the event to the industry in the event period corresponding to the event day, so that the interference degree of the event to the industry can be reflected on the whole according to the parameter information determined by the number of days proportion corresponding to each event day, and therefore, the reference information for representing the interference degree of the event to the industry can be accurately determined according to the number of days proportion corresponding to each event day.

When the reference information is determined according to the number of days corresponding to each event day, the average value of the number of days corresponding to each event day can be calculated, and the calculated average number of days is used as the reference information for representing the interference degree of the event on the industry.

In addition, when calculating the average value of the number of days corresponding to each event day, the weight of the number of days corresponding to each event day may be determined based on the net emotion value corresponding to each event day, and the weighted average value of the number of days may be calculated based on the determined weight.

In a second implementation manner, for each event day, according to the profitability of each securities product and the securities trade market in the estimated period corresponding to the event day, calculating the abnormal profitability of each securities product in the estimated period corresponding to the event day; determining the cumulative income ratio of each security product in the event period corresponding to the event day according to the abnormal income ratio of each security product in the event period corresponding to the event day; and determining a second interference parameter representing the interference degree of the event in the event period corresponding to the event day, wherein the interference degree of the industry is interfered according to the calculated abnormal yield and the determined accumulated yield.

Specifically, when calculating the abnormal rate of return of each securities product per day during the estimation period, the rate of return of each securities product and the securities trade market per day during the estimation period can be obtained. For each securities product, carrying out linear regression on the obtained securities product and the obtained security market, obtaining a linear relation between the security product and the security market, calculating the expected daily rate of the securities product in the estimated period according to the linear relation and the obtained daily rate of the security market in the estimated period, and subtracting the obtained actual daily rate of the securities product from the expected daily rate of the securities product in the estimated period, thus obtaining the abnormal daily rate of the securities product in the estimated period.

When the cumulative yield of the securities in the event period is determined, the abnormal yield of the securities in each day in the event period can be accumulated to obtain the accumulated cumulative yield.

For example, the abnormal rate of return of a security product per day over an event period may be accumulated by the following expression:

representing the cumulative yield of the security product i +.>

First day of the presentation of event period, +.>

Day of event period>

The standard abnormal yield of the date t of the securities product i in the event period is represented, namely the abnormal yield of the date t of the securities product i in the event period after standardized processing.

In addition, when the abnormal yields of the ticket products per day during the event period are accumulated, the abnormal yields of the ticket products per day during the event period may also be weighted, so that the abnormal yields of the ticket products per day during the event period are weighted and summed.

In one embodiment of the invention, the abnormal rate of return of the coupon product per day during the event period may be weighted by the following expression:

the weight of the abnormal yield of the securities product i on the date t in the event period is smaller if the date t is smaller, namely, the date t is closer to the first day of the event period, and the weight is larger if the date t is larger, namely, the date t is closer to the last day of the event period.

After calculating the cumulative rate of return of each security product in the event period corresponding to the event day, the cumulative abnormal rate of return of each security product can be calculated according to the calculated cumulative rates of return.

In one embodiment of the invention, the cumulative abnormal yield for each security product may be calculated according to the following expression:

representing cumulative abnormal yield of securities product i, < ->

An average value of the cumulative yield of each security product is represented.

After calculating the abnormal rate of return of each securities in each day in the estimated period corresponding to the event day and the accumulated abnormal rate of return of each securities in the event period corresponding to the event day, the abnormal rate of return and the accumulated abnormal rate of return can be used as an observation point, the observation point and the calculated abnormal rates of return are ordered, if the event does not interfere with the industry or has lower interference degree, the accumulated rate of return is located at the middle position of the arranged sequence, and if the event interferes with the industry to a greater degree, the accumulated rate of return is located at the front or back position of the arranged sequence.

On the pair of pairsAfter the abnormal yield and the accumulated abnormal yield are ranked, the generalized abnormal yield of the securities product i can be obtained

：

According to the generalized abnormal yield of each securities product, a second interference parameter which characterizes the interference degree of the event in the event period corresponding to the event day and interfering the industry can be determined through the following expression:

indicates estimated days,/->

Characterizing the degree of deviation of the abnormal yield of date t from the generalized abnormal yield intermediate position in the generalized abnormal yield of security product i +.>

Indicating +.about.of all securities products in industry at date t>

T represents the average value of securities productionThe number of elements contained in the generalized abnormal yield of the product,/->

Necessary normalization parameter item representing the above second interference parameter,/or->

Indicating +.about.of all securities products in industry on event day>

Average value of>

Representing the second interference parameter.

After determining the second interference parameters corresponding to each event day, the second interference parameters with parameter values larger than the second parameter threshold value can be determined according to the parameter values of each second interference parameter, so that the event day corresponding to the determined second interference parameters can be accurately determined, and the event period corresponding to the determined event day can be accurately determined to be the interference period.

After the interference period in each event period is determined, the number proportion of the interference period to the event period corresponding to each event day can be calculated and used as reference information for representing the interference degree of the event to the industry.

For example, if 5 event days are determined, the number of event periods is also 5, at this time, if 3 event periods in the 5 event periods are determined to be interference periods, the number proportion of the interference periods to each event period may be calculated to be 3/5=0.6, and the calculated number proportion is the reference information representing the interference degree of the event to the industry.

The interference degree of the event in the interference period is stronger, if the interference period is more, the interference degree of the event on the whole industry is stronger, so that the number proportion of the interference period to the event period is calculated, the number proportion can accurately reflect the interference degree of the event on the whole industry, and therefore the number proportion of the interference period to the event period is taken as reference information, and the accuracy of the reference information can be improved.

The embodiment of the invention utilizes various profitability to determine the reference information, and can also combine various profitability and other information to determine the reference information.

In an embodiment of the present invention, referring to fig. 5, a flowchart of a third reference information generating method is provided, in this embodiment, the method further includes the following steps S106 to S107, and the step S105 may be implemented by the following steps S105A to S105B.

Step S106: a text vector is determined that characterizes each event information text.

Wherein the text vector is used for representing each event information text. For example, if there are 1000 event information texts, the text vector is used to characterize the 1000 event information texts.

Specifically, each event information text may be converted into a text vector representing each event information text by a text conversion algorithm, model, or the like.

The specific implementation of determining the text vector described above may be found in the following embodiments, which are not described in detail herein.

Step S107: for each security product, obtaining association parameters representing the correlation between each event information text and the security product according to the determined text vector and the product vector representing the security product.

Wherein, the product vector for representing the securities product can be preset.

In one embodiment of the present invention, the product vector of the security product is determined according to the map information of the security product in the product knowledge map of the pre-established industry.

When the product vector of the security product is determined, the security product can be vectorized according to the entity node information of the security product in the product knowledge graph of the industry and the information of the entity nodes on the upstream and downstream of the security product, so that the product vector of the security product is obtained, and the product vector of the security product can be accurately determined based on the product knowledge graph of the industry.

For example, in vectorizing a coupon product, a deepflk algorithm may be employed to effect vectorizing of the coupon product.

After the text vector and the product vector are determined, a correlation parameter representing the correlation between each event information text and the security product can be obtained based on the text vector and the product vector.

The above-described association parameters may be obtained, for example, by either of the following two implementations.

In a first implementation, the similarity between the text vector and the product vector may be used as the above-mentioned association parameter.

In a second implementation, the distance between the text vector and the product vector may be calculated as the above-mentioned association parameter.

In one embodiment of the present application, after obtaining the associated parameters corresponding to each security product, the obtained associated parameters may be normalized.

For example, the associated parameters corresponding to each security product may be normalized by the following expression:

representing the associated parameters after normalization +.>

Representing the associated parameters prior to normalization.

After the above-described association parameters are obtained, the above-described step S105 can be implemented by the following steps S105A to S105B.

Step S105A: and calculating the proportion of the period of the interference period to the period of the event period.

The specific implementation of calculating the term ratio may refer to the implementation of calculating the number of days ratio and the number ratio in the above embodiment, and will not be described herein.

Step S105B: and determining reference information representing the interference degree of the event to the industry according to the deadline proportion and the associated parameters.

Specifically, according to the term proportion and the associated parameters, the average value of the term proportion and the associated parameters can be calculated and used as reference information for representing the interference degree of the event to the industry.

In addition, the term ratio and the associated parameter may be regarded as parameters for evaluating the degree of interference of the event to the industry from different angles, and thus, weights of the two parameters may be set in advance, so that when calculating the average value of the term ratio and the associated parameter, a weighted average value of the term ratio and the associated parameter may be calculated based on the set weights as the above-mentioned reference information.

From the above, when the scheme provided by the embodiment of the invention is applied to generate the reference information, not only the interference period and the event period are considered, but also the correlation between the text vector representing the text of each event information and the product vector representing the security product is considered, so that the information considered when the reference information is generated is richer, and the accuracy of the generated reference information can be improved.

A specific implementation of determining the above text vector is described below.

For example, the text vector described above may be determined by either of two implementations.

In a first implementation manner, map entities contained in a product knowledge map belonging to the industry and entity keywords appearing in at least one event information text can be determined, and the occurrence times of the entity keywords in each event information text can be determined; and determining a text vector representing the text of each event information according to the word vector and the occurrence times of the entity keywords.

For example, if the product knowledge graph includes a graph entity of "express," and "express" appears in at least one text in the event information texts, it may be determined that the word of "express" is an entity keyword, and the number of occurrences of the word of "express" in each event information text is counted.

After determining the keywords of each entity and the occurrence times thereof, the weight of the keywords of each entity can be determined according to the occurrence times of the keywords of each entity, wherein the weight is larger as the occurrence times are more, and the weight is smaller as the occurrence times are less. In addition, for each map entity in the product knowledge map, a word vector of each map entity may be preset, for example, the word vector of the map entity may be determined by using NLP (Natural Language Processing ) technology, so that after the entity keyword is determined, a word vector corresponding to the entity keyword may be determined, and thus, a text vector representing each event information text may be calculated according to the word vector of the entity keyword and the weight determined based on the occurrence number.

From the above, when the scheme provided by the embodiment of the invention is applied to generate the reference information, the text vector representing each event information text is determined by the entity keyword and the occurrence frequency thereof, and the entity keyword belongs to the map entity contained in the product knowledge map, so that the text vector can be considered to be derived from the product knowledge map and each event information text, and the product vector is also derived from the product knowledge map, therefore, the related parameters are obtained by using the text vector and the product vector with the same source part, and the accuracy of the related parameters can be improved, so that the accuracy of the generated reference information can be improved.

In a second implementation, a text conversion model may be trained in advance, and each event information text is input into the text conversion model, so as to obtain a vector output by the model, which is used as a text vector representing each event information text.

In an embodiment of the present invention, referring to fig. 6, a schematic structural diagram of an event analysis system is provided, where in this embodiment, the system includes an information preparation module, a knowledge preparation module, an emotion analysis module, a metering statistics module, a text analysis module, and an integrated analysis module.

The information preparation module is used for recording various news information texts and searching event information texts related to the event information from the recorded news information texts according to event information of events input by a user;

the knowledge preparation module is used for constructing and maintaining knowledge graphs of intra-industry coupon products of various industries;

the emotion analysis module is used for obtaining event information texts searched by the information preparation model, detecting emotion types of the event information texts to obtain emotion types corresponding to the event information texts, and determining emotion net values corresponding to the same day according to the number of the event information texts corresponding to different emotion types of the same day, so that after emotion net values of a plurality of dates are determined according to the emotion types of all the event information texts, event days of the event can be determined according to the emotion net values;

the metering statistics module is used for determining an event period and an estimated period corresponding to the event according to the event day determined by the emotion analysis module, and calculating a first interference parameter representing the interference degree of the event in the event period for interfering the industry every day and/or a second interference parameter representing the interference degree of the event in the event period corresponding to the event day according to the profitability of the securities products and the securities trade market in the event period and the estimated period respectively;

The text analysis module is used for establishing association between the event information text and the securities products and obtaining association parameters for representing the relativity between the event information text and the securities products;

the comprehensive analysis module is used for obtaining the interference parameters output by the metering statistics module and the associated parameters output by the text analysis module, determining the interference period in the event period based on the interference parameters, calculating the period proportion of the interference period to the event period, integrating the period proportion and the associated parameters, and finally determining the reference information representing the interference degree of the event to the industry.

Corresponding to the above reference information generating method, the embodiment of the invention also provides a reference information generating device.

In one embodiment of the present invention, referring to fig. 7, there is provided a schematic structural diagram of a reference information generating apparatus, in this embodiment, the apparatus includes:

an information obtaining module 701, configured to obtain event information of an event that has occurred;

an event day obtaining module 702, configured to obtain an event day when the event determined based on the event information interferes with each security product in a preset industry;

a profitability calculation module 703, configured to obtain a profitability of each securities product and a securities market in an event period and a profitability of each securities market in an estimated period, and calculate an abnormal profitability of each securities product in the event period according to the obtained profitability, wherein the event period and the estimated period are determined according to the event day, and a last day of the estimated period is a day before a first day of the event period;

A disturbance period determining module 704, configured to determine a disturbance period during the event period when the event interferes with the industry according to an abnormal daily yield rate of each security product during the event period;

and the information generating module 705 is configured to determine, according to the event period and the interference period, reference information that characterizes the interference degree of the event on the industry.

In this scheme, if the event interferes with the coupon product, the overall emotion of the information released on the event day will tend to be one of positive and negative emotion, that is, the emotion net value corresponding to the event day is more extreme than the emotion net value corresponding to each information release day, and the standard deviation of the emotion net value corresponding to each information release day reflects the distribution condition of the emotion net value corresponding to each information release day, so that the target value is determined according to the emotion net value corresponding to each information release day, and if the value of the emotion net value corresponding to the information release day is greater than the target value, it is indicated that the emotion net value is the extreme value in each emotion net value, that is, the emotion net value is more extreme than the emotion net value corresponding to each information release day, so that the accuracy of determining the event day can be improved according to the emotion net value determination corresponding to each information release day, and the accuracy of the generated reference information can be improved.

According to the scheme, according to the determined income relation, the income ratio of each securities product in the event period and the income ratio of the securities trade market in the event period, the abnormal income ratio of each securities product in the event period can be accurately calculated, so that the reference information is generated according to the relatively accurate abnormal income ratio of the securities product, and the accuracy of the generated reference information can be improved.

In one embodiment of the invention, the apparatus further comprises:

The information generation module comprises:

According to the scheme, according to the first interference parameters, the interference period can be accurately determined, the number of days of the interference period occupying the event period is calculated, if the interference period is longer, the stronger the interference degree of the event to the whole industry is indicated, so that the number of days of the interference period occupying the event period is calculated, the number of days of the event can be accurately reflected, and therefore the number of days of the interference period occupying the event period is taken as reference information, and the accuracy of the reference information can be improved.

In the scheme, the number of days proportion corresponding to each event day can accurately reflect the interference degree of the event to the whole industry in the event period corresponding to each event day, so that the reference information for representing the interference degree of the event to the industry can be accurately determined according to the number of days proportion corresponding to each event day.

the information generation module is specifically configured to:

In the scheme, the interference degree of the event in the interference period is stronger, if the interference period is more, the interference degree of the event on the whole industry is stronger, so that the number proportion of the interference period to the event period is calculated, the number proportion can accurately reflect the interference degree of the event on the whole industry, and therefore the number proportion of the interference period to the event period is used as reference information, and the accuracy of the reference information can be improved.

In one embodiment of the invention, the apparatus further comprises:

the information generation module is specifically configured to:

In the scheme, when the product vector of the security product is determined, the security product can be vectorized according to the entity node information of the security product in the product knowledge graph of the industry and the information of the entity nodes on the upstream and the downstream of the security product, so that the product vector of the security product is obtained, and the product vector of the security product can be accurately determined based on the product knowledge graph of the industry.

The embodiment of the present invention further provides an electronic device, as shown in fig. 8, including a processor 801, a communication interface 802, a memory 803, and a communication bus 804, where the processor 801, the communication interface 802, and the memory 803 complete communication with each other through the communication bus 804,

a memory 803 for storing a computer program;

the processor 801, when executing the program stored in the memory 803, implements the following steps:

obtaining event information of an event which has occurred;

Other schemes for generating the reference information by the processor 801 executing the program stored in the memory 803 are the same as those mentioned in the foregoing method embodiment, and will not be repeated here.

The communication bus mentioned above for the electronic devices may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, etc. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.

The communication interface is used for communication between the electronic device and other devices.

The Memory may include random access Memory (Random Access Memory, RAM) or may include Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.

The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-ProgrammableGate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

In yet another embodiment of the present invention, there is also provided a computer-readable storage medium having stored therein a computer program which, when executed by a processor, implements the steps of any of the above-described reference information generating methods.

In yet another embodiment of the present invention, there is also provided a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the reference information generating methods of the above embodiments.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), etc.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the apparatus, electronic device, computer readable storage medium, and computer program product embodiments, the description is relatively simple, as relevant to the method embodiments being referred to in the section of the description of the method embodiments.

The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A reference information generation method, the method comprising:

obtaining event information of an event which has occurred;

2. The method of claim 1, wherein the obtaining an event day when the event determined based on the event information interferes with each security product in a preset industry comprises:

obtaining event information text related to the event information;

3. The method according to claim 2, wherein in the case where it is determined that a plurality of information release days corresponding to the net emotion value having a value greater than the target value are determined, the determining of the information release days corresponding to the net emotion value having a value greater than the target value as the event day when the event interferes with each security product in the preset industry includes:

4. A method according to claim 2 or 3, wherein the target value is determined based on a standard deviation of net emotion values corresponding to respective information distribution days.

5. A method according to any one of claims 1-3, wherein said calculating an abnormal daily rate of return for each security product over the event period in dependence on the obtained rate of return comprises:

6. The method of claim 1, wherein after said calculating an abnormal daily rate of return for each security product over the event period in accordance with the obtained rates of return, the method further comprises:

7. A method according to any one of claims 1 to 3, wherein said determining a period of interference during said event during which said event interferes with said industry based on an abnormal daily rate of return of each security product during said event period comprises:

8. The method of claim 7, wherein in the event that a plurality of event days are determined, the calculating the proportion of days of the period of interference to the period of event as reference information characterizing the degree of interference of the event to the industry comprises:

9. A method according to any one of claims 1-3, wherein, in the event that a plurality of event days are determined, said determining a period of interference during said event that said event interferes with said industry based on an abnormal daily rate of return of each security product during said event period comprises:

10. A method according to claim 2 or 3, characterized in that the method further comprises:

determining a text vector representing each event information text;

11. The method of claim 10, wherein the product vector of the security product is determined from map information of the security product in a pre-constructed product knowledge map of the industry.

12. The method of claim 11, wherein said determining a text vector characterizing each event information text comprises:

13. A reference information generating apparatus, characterized in that the apparatus comprises:

14. The electronic equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;

a memory for storing a computer program;

a processor for carrying out the method steps of any one of claims 1-12 when executing a program stored on a memory.

15. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored therein a computer program which, when executed by a processor, implements the method steps of any of claims 1-12.