Stock screening method and system

Abstract

The invention relates to the technical field of stock recommendation, in particular to a stock screening method and system. The method comprises the steps of collecting multiple pieces of user information, dividing user levels based on the multiple pieces of user information, then collecting stock information, calculating the recommendation level of each stock based on the stock information, screening out multiple alternative stocks in combination with the user levels of users and the recommendation levels of the stocks, and finally calculating the expected income of each alternative stock. According to the stock recommendation method and device, different user levels are divided for different users based on the user information, multi-source information integration is achieved, the user information has a more important proportion in the stock recommendation process, and the user recommendation efficiency is improved. Therefore, it is ensured that the recommended stock can meet the personal conditions of the user and has reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of stock recommendation, and particularly to a stock screening method and system. Background Art

[0002] A stock is a certificate issued by a joint stock limited company to prove the shares held by shareholders, indicating that the holder of the stock has ownership of a part of the capital of the joint stock company. Since stocks contain economic interests and can be listed and circulated, stocks are also a type of negotiable securities.

[0003] Chinese Patent with publication number CN118134640A discloses a method for intelligent stock recommendation. By integrating the DIKW knowledge graph into stock recommendation, before constructing the stock DIKW model, a user DIKWP model is first constructed based on user basic information. The stock screening conditions obtained from the user DIKWP model can be used to screen stocks, reducing the calculation amount and improving the efficiency and accuracy of recommendation. However, in the prior art, the relevant conditions of users are not integrated, resulting in the recommended stocks being difficult to meet the actual conditions and needs of users. Summary of the Invention

[0004] The purpose of the present invention is to propose a stock screening method and system for the problems existing in the background art.

[0005] The technical solution of the present invention:

[0006] On the one hand, the present application provides a stock screening method, including:

[0007] Collecting multiple user information and dividing user levels based on the multiple user information;

[0008] Collecting stock information and calculating the recommendation level of each stock based on the stock information;

[0009] Combining the user level of the user with the recommendation level of the stock to screen out multiple alternative stocks;

[0010] Calculating the expected return of each alternative stock, judging whether the expected return of the alternative stock meets the expected target of the customer, and screening out the alternative stocks that meet the expected target of the customer as recommended stocks.

[0011] Preferably, collecting multiple user information and dividing user levels based on the multiple user information includes:

[0012] Creating a user information table;

[0013] Collecting the user information of multiple users and putting all the collected user information into the user information table.

[0014] Preferably, multiple user information is collected, and user levels are divided based on the multiple user information. It further includes:

[0015] Select the user information of one user from the user information table;

[0016] Calculate the user score corresponding to this user based on the user information through Formula 1;

[0017]

[0018] where P is the user score, M i is the i-th user information, N i is the weight corresponding to the i-th user information, and c is the total number of user information;

[0019] Set user levels and the user score ranges corresponding to each user level;

[0020] Set the user level of this user according to the user score range where the user score corresponding to this user is located;

[0021] Return the user information of one user selected from the user information table until all users in the user information table have been selected, and obtain the user levels of each user.

[0022] Preferably, stock information is collected, and the recommended level of each stock is calculated based on the stock information, including:

[0023] Create a stock information table;

[0024] Collect multiple stock information and put all the collected stock information into the stock information table.

[0025] Preferably, stock information is collected, and the recommended level of each stock is calculated based on the stock information. It further includes:

[0026] Select one stock information from the stock information table;

[0027] Collect user intentions and adjust the weights of each stock information based on the user intentions;

[0028] Calculate the stock score of this stock through Formula 2 in combination with the adjusted weights;

[0029]

[0030] where Q is the stock score, M i is the i-th stock information, N i is the weight corresponding to the i-th stock information, and k is the total number of stock information;

[0031] Set the recommended levels and the stock score ranges corresponding to each recommended level;

[0032] Set the stock rating of the stock according to the stock rating range corresponding to the stock.

[0033] Return to select a stock and obtain the stock information of the stock until all stocks are selected, and obtain the stock rating of each stock.

[0034] Preferably, screen out multiple alternative stocks by combining the user rating of the user and the recommended rating of the stock, including:

[0035] Obtain the user rating.

[0036] Screen out the recommended rating corresponding to the user rating; record the selected recommended rating as the alternative rating.

[0037] Sort all the stocks in the alternative rating from high to low according to the size of their stock ratings, and select the top N stocks from the alternative rating; record the selected stocks as alternative stocks.

[0038] Preferably, calculate the expected return of each alternative stock, judge whether the expected return of the alternative stock meets the expected target of the customer, and screen out the alternative stocks that meet the expected target of the customer as recommended stocks, including:

[0039] Obtain all alternative stocks and the rise and fall trends of each alternative stock within the set period; specifically, the set period can be a period set by the user himself, such as one month, one quarter or one year.

[0040] Calculate the maximum expected loss of each alternative stock by combining the customer's intended funds and the rise and fall trends of the alternative stocks.

[0041] Judge in turn whether the maximum expected loss of each alternative stock is greater than or equal to the maximum acceptable loss of the customer. If the maximum expected loss of the alternative stock is greater than the maximum acceptable loss of the customer, then exclude the alternative stock.

[0042] Record the remaining alternative stocks as recommended stocks.

[0043] Preferably, the user information includes user funds, user occupation and user age, and the stock information includes the industry where the stock is located, the financial status of the stock company and the trading information of the stock.

[0044] On the other hand, the present application also provides a stock screening system, including an information collection component and a control component, through which user information and stock information are collected. Through the stock screening method described in any of the above-mentioned control components, the user information and stock information collected by the information collection component are transmitted to the control component, and the control component combines the user information and stock information to calculate recommended stocks suitable for the user.

[0045] Preferably, the information collection component includes a customer module and a stock module, and user information is collected through the customer module, and stock information is collected through the stock module.

[0046] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:

[0047] By collecting information of multiple users and dividing users into levels based on the information of multiple users, then collecting stock information and calculating the recommendation level of each stock based on the stock information, multiple alternative stocks are screened out based on the user level of the user and the recommendation level of the stock, and finally the expected return of each alternative stock is calculated to determine whether the expected return of the alternative stock meets the customer's expected goals, and the alternative stocks that meet the customer's expected goals are screened out as recommended stocks. This application divides different user levels for different users based on user information, realizes multi-source information integration, and makes user information have a more important proportion in the stock recommendation process, thereby ensuring that the recommended stocks can meet the user's personal conditions and are reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a flow chart of a stock screening method proposed by the present invention;

[0049] Figure 2 This is a structural diagram of a stock screening system proposed by the present invention;

[0050] Reference numerals: 100, information collection component; 101, customer module; 102, stock module;

[0051] 200. Control components. DETAILED DESCRIPTION

[0052] Example 1, as Figure 1 As shown, the present invention proposes a stock screening method, comprising:

[0053] S100, collecting information of multiple users and classifying users based on the information of multiple users;

[0054] S200, collecting stock information and calculating the recommendation level of each stock based on the stock information;

[0055] S300, screen out multiple candidate stocks by combining the user level of the user with the recommendation level of the stocks;

[0056] S400, calculate the expected return of each candidate stock, determine whether the expected return of the candidate stock meets the expected target of the customer, and screen out the candidate stocks that meet the expected target of the customer as recommended stocks.

[0057] In the present invention, by collecting multiple user information and dividing user levels based on the multiple user information, then collecting stock information and calculating the recommendation level of each stock based on the stock information, combining the user level of the user with the recommendation level of the stocks to screen out multiple candidate stocks, and finally calculating the expected return of each candidate stock, determining whether the expected return of the candidate stock meets the expected target of the customer, and screening out the candidate stocks that meet the expected target of the customer as recommended stocks. This application divides different user levels for different users based on user information, realizes the integration of multi-source information, makes user information have a more important proportion in the stock recommendation process, so as to ensure that the recommended stocks can meet the personal conditions of the user and have reliability.

[0058] In an optional embodiment, the 100 includes:

[0059] S110, create a user information table;

[0060] S120, collect the user information of multiple users and put all the collected user information into the user information table.

[0061] It should be noted that for new user who have just entered the stock market, there are huge differences between the user information corresponding to each new user. It is precisely because of the differences between different new users that each new user has differences in the stocks they want to choose. Therefore, by collecting the user information of multiple users and establishing user levels for different users according to different user information, it is convenient to recommend suitable stocks according to user recommendations.

[0062] In an optional embodiment, the S100 further includes:

[0063] S130, select the user information of a user from the user information table;

[0064] S140, calculate the user score corresponding to the user based on the user information through formula 1;

[0065]

[0066] Where P is the user score, M i is the i-th user information, N i is the weight corresponding to the i-th user information, and c is the total number of user information;

[0067] Specifically, the weight corresponding to each user information can be adjusted in real time according to different actual emphases. For example, if the invested funds are emphasized, the weight corresponding to the invested funds can be increased. However, in one calculation, the weight criteria adopted for all users should be kept consistent;

[0068] S150, set the user levels and the user score ranges corresponding to each user level;

[0069] S160, set the user level of the user according to the user score range in which the user score corresponding to the user is located;

[0070] S170, return to step S130 until all users in the user information table have been selected, and obtain the user levels of each user.

[0071] It should be noted that the amount of money that each user can invest in the stock market and the risk tolerance ability are different. Therefore, for different users, suitable stocks need to be recommended in combination with their user information. Then, in order to better recommend stocks to users, the user level of the user is calculated by combining the user information through formula 1, so as to ensure that the personal conditions of the user can be fully considered.

[0072] If there are currently 3 new users who have just entered the stock market, denoted as user A, user B, and user C respectively, and the user information of each user is shown in Table 1 - User Information Table.

[0073] Table 1

[0074] User User's Occupation User's Annual Income / 10,000 yuan Planned Investment / 10,000 yuan … User A Doctor 30 30 … User B Teacher 20 10 … User C Private Enterprise Manager 100 200 … … … … … …

[0075] It can be seen that the job stability of user A and user B is higher than that of user C. Therefore, although user C has a higher annual income, the job stability of user C is lower than that of user A and user B. So if only the criterion of "job" is used as the only item in formula 1, the risk resistance ability of user A and user B is higher than that of user C.

[0076] However, if job stability, annual income, and planned invested funds are comprehensively considered, and the corresponding weights are set to 0.3, 0.4, and 0.3 respectively, then after calculation by formula 1, the user scores corresponding to user A, user B, and user C are 39, 26, and 106 respectively (a stable job is recorded as 50, and an unstable job is recorded as 20).

[0077] In the case where the divided user levels and the range of each user level are shown in Table 2 - User Level Table, the user level of user A can be obtained as level 1, the user level of user B is level 1, and the user level of user C is level 2.

[0078] Table 2

[0079]

[0080]

[0081] In an optional embodiment, the step S200 includes:

[0082] S210, creating a stock information table;

[0083] S220: Collect multiple stock information and put all the collected stock information into a stock information table.

[0084] It should be noted that since stocks belong to different industries and each stock corresponds to a different company, different stocks have different appeal and suitability for different users. By collecting information on multiple stocks, it is easy to screen out the stocks that best meet the user's conditions.

[0085] In an optional embodiment, the S200 further includes:

[0086] S230, selecting a stock information from the stock information table;

[0087] S240, collecting user intentions, and adjusting the weights of various stock information based on the user intentions;

[0088] S250, calculating the stock score of the stock by using Formula 2 in combination with the adjusted weight;

[0089]

[0090] Among them, Q is the stock score, M i is the i-th stock information, N i is the weight corresponding to the i-th stock information, k is the total number of stock information;

[0091] S260, setting the recommended level and the stock score range corresponding to each recommended level;

[0092] S270, setting the stock grade of the stock according to the stock score range of the stock score corresponding to the stock;

[0093] S280, return to step S230, until all stocks are selected, and the stock level of each stock is obtained.

[0094] It should be noted that by combining stock information to calculate the stock score corresponding to each stock, and dividing all stocks into different levels according to the stock score, recommendations are made based on the stock level and user level to ensure the rationality of the recommended stocks.

[0095] In an optional embodiment, the S300 includes:

[0096] S310, obtaining the user level;

[0097] S320, screening the recommended levels corresponding to the user level; and recording the selected recommended level as the alternative level;

[0098] S330, sorting all the stocks in the alternative level in descending order according to their stock scores, and selecting the top N stocks from the alternative level; and recording the selected stocks as alternative stocks.

[0099] It should be noted that since the stock level and the user level are calculated by considering stock information and user information respectively, the stock level can reflect the relevant information of the stock, and the user level can reflect the relevant information of the user. Therefore, the stock level and the user level can be directly matched to ensure the rationality of the match.

[0100] In an optional embodiment, the S400 includes:

[0101] S410, obtaining all the alternative stocks and the rise and fall trends of each alternative stock within the set period;

[0102] Specifically, the set period can be a period set by the user himself, such as one month, one quarter or one year;

[0103] S420, calculating the maximum expected loss of each alternative stock by combining the customer's intended funds and the rise and fall trends of the alternative stocks;

[0104] Optionally, calculating the VaR of the alternative stocks based on the historical simulation method through Formula 3;

[0105] VaR = E(ω) - ω * Formula 3;

[0106] where E(ω) is the average daily income, and ω * is the lowest income level under the premise of a confidence level of a;

[0107] S430, sequentially determining whether the maximum expected loss of each alternative stock is greater than or equal to the customer's maximum acceptable loss. If the maximum expected loss of the alternative stock is greater than the customer's maximum acceptable loss, then exclude the alternative stock;

[0108] S440, recording the remaining alternative stocks as recommended stocks.

[0109] It should be noted that the maximum expected loss of the alternative stocks is calculated through VaR. VaR is generally referred to as "Value at Risk" or "Risk Value at Risk", which refers to the maximum possible loss of a financial asset (or portfolio of securities) within a specific period in the future at a certain confidence level.

[0110] In an alternative embodiment, the user information includes user funds, user occupation, and user age, and the stock information includes the industry where the stock is located, the financial status of the stock company, and the trading information of the stock.

[0111] It should be noted that after collecting the user information and stock information, corresponding scores need to be assigned to each piece of user information and stock information to facilitate the calculation of user scores and stock scores. For different types of stocks, for example, stocks in the AI chip industry should have a higher score than stocks in the civil engineering industry. <9>

[0112] When assigning scores, different scores can be assigned according to different emphases, but the consistency of the scores needs to be maintained during the calculation process.

[0113] As Figure 2 shown, the present application also provides a stock screening system, including an information collection component and a control component. The user information and stock information are collected through the information collection component, and the stock screening method described in any one of the first embodiments is executed through the control component. The user information and stock information collected by the information collection component are transmitted into the control component, and the recommended stocks suitable for the user are calculated by the control component in combination with the user information and stock information.

[0114] It should be noted that after the information collected by the information collection component is transmitted to the control component, the control component needs to calculate the user level and the stock level by combining the user information and the stock information respectively, and then recommend stocks corresponding to the level based on the user level.

[0115] In an alternative embodiment, the information collection component includes a customer module and a stock module. The user information is collected through the customer module, and the stock information is collected through the stock module.

[0116] It should be noted that both the customer module and the stock module are communicatively connected to the control component to facilitate information transmission.

[0117] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made without departing from the spirit of the present invention within the scope of knowledge possessed by those skilled in the art to which the present invention pertains.

Claims

1. A stock screening method, characterized in that, Including: Collect multiple user information and divide user levels based on the multiple user information; Collect stock information and calculate the recommended level for each stock based on the stock information; Combine the user level of the user with the recommended level of the stock to screen out multiple alternative stocks; Calculate the expected return of each alternative stock, determine whether the expected return of the alternative stock meets the customer's expected goal, and screen out the alternative stocks that meet the customer's expected goal as recommended stocks.

2. The stock screening method according to claim 1, wherein Collect multiple user information and divide user levels based on the multiple user information, including: Create a user information table; Collect the user information of multiple users and put all the collected user information into the user information table.

3. The stock screening method according to claim 2, wherein Collect multiple user information and divide user levels based on the multiple user information, and also include: Select the user information of one user from the user information table; Calculate the user score corresponding to the user based on the user information through Formula 1; Among them, P is the user rating, and M i is the i-th user information, and N i is the weight corresponding to the i-th user information, and c is the total number of user information; Set the user level and the user score range corresponding to each user level; Set the user level of the user according to the user score range where the user score corresponding to the user is located; Return the user information of one user selected from the user information table until all users in the user information table have been selected, and obtain the user level of each user.

4. A stock screening method according to claim 3, characterized in that, Collect stock information and calculate the recommended level for each stock based on the stock information, including: Create a stock information table; Collect multiple stock information and put all the collected stock information into the stock information table.

5. A stock screening method according to claim 4, characterized in that, Collect stock information and calculate the recommended level for each stock based on the stock information, and also include: Select one stock information from the stock information table; Collect user intentions and adjust the weights of each stock information based on the user intentions; Calculate the stock score of the stock through Formula 2 in combination with the adjusted weights; Where Q is the stock rating, M i is the i-th stock information, N i is the weight corresponding to the i-th stock information, and k is the total number of stock information; Set the recommended level and the stock score range corresponding to each recommended level; Set the stock level of the stock according to the stock score range where the stock score corresponding to the stock is located; Return to select one stock and obtain the stock information of the stock until all stocks have been selected, and obtain the stock level of each stock.

6. A stock screening method according to claim 5, wherein Combine the user level of the user with the recommended level of the stock to screen out multiple alternative stocks, including: Obtain the user level; Screen the recommended levels corresponding to the user level; record the selected recommended levels as alternative levels; Sort all the stocks in the alternative levels from high to low according to the size of their stock scores, and select the top N stocks from the alternative levels; record the selected stocks as alternative stocks.

7. A stock screening method according to claim 6, characterized in that, Calculate the expected return of each alternative stock, determine whether the expected return of the alternative stock meets the customer's expected goal, and screen out the alternative stocks that meet the customer's expected goal as recommended stocks, including: Obtain all the alternative stocks and the rise and fall trends of each alternative stock within the set period; calculate the maximum expected loss of each alternative stock in combination with the customer's intended funds and the rise and fall trends of the alternative stocks; Judge in turn whether the maximum expected loss of each alternative stock is greater than or equal to the customer's maximum acceptable loss. If the maximum expected loss of the alternative stock is greater than the customer's maximum acceptable loss, then eliminate the alternative stock; The reserved alternative stocks are recorded as recommended stocks.

8. A stock screening method according to claim 7, characterized in that, The user information includes user funds, user occupation, and user age, and the stock information includes the industry where the stock is located, the financial status of the stock company, and the trading information of the stock.

9. A stock screening system, characterized in that, It includes: An information collection component, which collects user information and stock information through the information collection component; A control component, which executes the stock screening method according to any one of claims 1 to 8 through the control component. The user information and stock information collected by the information collection component are transmitted into the control component, and the control component calculates the recommended stocks suitable for the user by combining the user information and the stock information.

10. A stock screening system according to claim 8, characterized in that, The information collection component includes a customer module and a stock module, and collects user information through the customer module and collects stock information through the stock module.

Images