A stock ranking selection processing method based on a quantitative model

Abstract

The application provides a stock ranking selection processing method based on a quantitative model, and belongs to the technical field of data processing, and specifically comprises the following steps: determining a matching coefficient of a stock and a matching stock according to a matching condition of the stock and different matching word groups of search words, and distribution data of the different matching word groups of search words on different dates; generating a plurality of matching stock groups based on the matching coefficients of the different matching stocks; determining a selection quantity of the matching stocks according to constituent data of the different matching stock groups and a matching condition of search words of a user on different dates; and determining a ranking processing result of the matching stocks of the user by using a quantitative model based on the selection quantity and financial data of the matching stocks, so that the ranking processing result of the matching stocks is ensured to be matched with the user.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, and in particular relates to a stock ranking and selection method based on a quantitative model. Background Technology

[0002] For stock investors, the sheer number of potential stock investment targets can lead to confusion during the stock selection process, making it difficult to achieve efficient investment decisions.

[0003] To improve the efficiency of investment decision-making for stock investors, the invention patent application CN202211256513.X, "A Company Selection Method Based on Stock Chart Similarity Detection," calculates the similarity of stock charts using multiple scales, thereby improving the accuracy of stock chart similarity detection and providing a more accurate reference for selecting high-performing companies. However, the following technical problems exist:

[0004] In stock selection, existing solutions neglect the stability of user profiles. They use the same number of recommendations for users with unstable profiles as for users with stable profiles. For users with unstable profiles, their interest in different stocks is more dispersed. Therefore, if the same recommendation method and number of recommendations are used, it cannot be guaranteed that the stock ranking results will match the user's needs.

[0005] To address the aforementioned technical issues, this application specifically provides a stock ranking and selection method based on a quantitative model. Summary of the Invention

[0006] To achieve the objectives of this invention, the following technical solution is adopted:

[0007] This application provides a stock ranking and selection method based on a quantitative model, specifically including:

[0008] S1 divides the search terms into different search term groups based on the similarity between the user's search terms, and determines the user's matching word group based on the distribution data of the search terms in different search term groups on different dates;

[0009] S2 determines the stability of the user profile based on the changes in the search terms of the matching words of the user across different dates. If the stability meets the requirements, proceed to the next step.

[0010] S3 determines the matching coefficient of the stock and the matching stock based on the matching results of the stock with different matching word groups and the distribution data of the matching word groups on different dates;

[0011] S4 generates multiple matching stock groups based on different matching coefficients of matching stocks. Based on the composition data of different matching stock groups and the matching situation of the user's search terms on different dates, the number of matching stocks to be selected is determined. Based on the number of selections and the financial data of the matching stocks, a quantitative model is used to determine the ranking result of the user's matching stocks.

[0012] The beneficial effects of this invention are as follows:

[0013] Based on the changes in the search terms of the matching phrases for the user across different dates, it is determined whether the stability of the user profile meets the requirements. This avoids the technical problem of unreliable identification of user profile stability caused by relying solely on changes in search terms. By fully considering the changes in search terms of the matching phrases across different dates, an accurate assessment of the stability of the user profile is achieved. It also lays the foundation for generating differentiated selection quantities of matching stocks based on differences in the stability of the user profile. By using a preset quantity to determine the selection quantity of matching stocks for users with less stable user profiles, the matching degree of the ranking recommendation results of the matching stocks can meet the requirements.

[0014] Based on the composition data of different matching stock groups and the matching situation with the user's search terms on different dates, the number of matching stocks to be selected is determined. This takes into account the differences in the number of stocks and the types of companies in the different matching stock groups, which lead to differences in the difficulty of sorting and the representativeness of the selected stocks. At the same time, it also takes into account the matching situation with the user's search needs on different dates. This achieves the determination of the number of matching stocks to be selected from multiple dimensions, which reduces the difficulty of sorting while also meeting the user's search needs and user profile requirements.

[0015] A further technical solution involves dividing the search terms into different search term groups, specifically including:

[0016] By matching different search terms with different stocks, we can identify stocks that match the same search terms simultaneously and treat them as similar matching stocks.

[0017] The search terms are divided into different search term groups based on the number of similar matching stocks.

[0018] A further technical solution is that the matching status of the search terms with different stocks is determined based on the matching status of the search terms with the business scope, company name, and products of the companies corresponding to the stocks.

[0019] A further technical solution involves dividing the search terms into different search term groups based on the number of similar matching stocks, specifically including:

[0020] Search terms that have more similar matching stocks than the preset number of stocks are grouped into the same search term group.

[0021] A further technical solution is that the method for determining the user's matching phrases is as follows:

[0022] The number of times the search terms in different search term groups were processed on different dates was determined by analyzing the distribution data of search terms in different search term groups on different dates.

[0023] Based on the number of times the user retrieves data on different dates and the number of times the search terms in the search term group are retrieved on different dates, the matching retrieval date for the search term group is determined;

[0024] Whether a search term group is a matching term group is determined based on the number of matching search dates for that search term group.

[0025] Optionally, the type of enterprise is determined based on the business scope of the enterprise corresponding to the stock.

[0026] Optionally, the method for determining the matching coefficient of the stock composition is as follows:

[0027] Based on the deviation of the enterprise types corresponding to the stocks in the matched stock group, the deviation coefficient of the corresponding enterprise types between the stocks in the matched stock group is determined, and the average deviation coefficient is determined by the average of the deviation coefficients of the corresponding enterprise types between different stocks.

[0028] The matching coefficient of the stock composition of the matching stock group is determined based on the ratio of the average deviation coefficient to the number of stocks in the matching stock group.

[0029] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0030] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0031] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings;

[0032] Figure 1 This is a flowchart of a stock ranking and selection method based on a quantitative model;

[0033] Figure 2 This is a flowchart illustrating the method for determining the user's matching phrases;

[0034] Figure 3 It is a flowchart for determining whether a user's profile is stable enough to meet the requirements;

[0035] Figure 4 This is a flowchart illustrating the method for determining the matching coefficients of stocks;

[0036] Figure 5 This is a flowchart illustrating the method for determining the number of stocks to be selected for matching. Detailed Implementation

[0037] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0038] In this application, by using the user's search terms and the matching information of the companies corresponding to the stocks, stocks that meet the requirements of matching the user's needs are identified, and the stock ranking results are automatically generated based on the stocks that meet the requirements of matching the user's needs.

[0039] Based on the ratio of the number of times search terms in a search term group are retrieved on different dates to the number of times users retrieve terms on different dates, the matching status of different dates with the search term group is determined. Dates with a matching status greater than 0.6 are designated as matching retrieval dates for the search term group. When the proportion of matching retrieval dates for a search term group is greater than 0.3, the search term group is determined as a matching term group.

[0040] Based on the search terms used by users to match words between different dates, the matching word groups corresponding to different dates are determined, and the dates are divided into different matching word group date groups based on the matching word groups corresponding to different dates;

[0041] When the proportion of dates in the matched date group is greater than 0.6, it is determined that the stability of the user profile meets the requirements.

[0042] The percentage of times a stock matches different search terms is determined by comparing the stock with each search term in a matching phrase, and this percentage is used as the matching coefficient of the stock with each search term in a matching phrase. The percentage of dates for which search terms in a matching phrase exist is determined, and this percentage is used as the weight coefficient of the matching phrase. The matching coefficient of a stock is determined by summing the products of the matching coefficient and the weight coefficient of the stock with different search terms in a matching phrase. Stocks with a matching coefficient of 0.6 or higher are considered matching stocks.

[0043] The stock matching deviation coefficient of the matching stock groups for different dates is determined by the proportion of search terms with no matching stocks on a given date. The effective matching stock groups in the matching stock groups are determined by the average of the stock matching deviation coefficients of the matching stock groups for different dates. The number of matching stocks to be selected is determined by the number of stocks in the effective matching stock group with the fewest matching stocks.

[0044] Example 1

[0045] like Figure 1 As shown, this application provides a first aspect: a stock ranking and selection method based on a quantitative model, specifically including:

[0046] S1 divides the search terms into different search term groups based on the similarity between the user's search terms, and determines the user's matching word group based on the distribution data of the search terms in different search term groups on different dates;

[0047] Furthermore, the search terms are divided into different search term groups, specifically including:

[0048] By matching different search terms with different stocks, we can identify stocks that match the same search terms simultaneously and treat them as similar matching stocks.

[0049] The search terms are divided into different search term groups based on the number of similar matching stocks.

[0050] Specifically, the matching of the search terms with different stocks is determined based on the matching of the search terms with the business scope, company name, and products of the companies corresponding to the stocks.

[0051] It should be noted that the search terms are divided into different search term groups based on the number of similar matching stocks, specifically including:

[0052] Search terms that have more similar matching stocks than the preset number of stocks are grouped into the same search term group.

[0053] Optionally, such as Figure 2 As shown, the method for determining the user's matching phrases is as follows:

[0054] The number of times the search terms in different search term groups were processed on different dates was determined by analyzing the distribution data of search terms in different search term groups on different dates.

[0055] Based on the number of times the user retrieves data on different dates and the number of times the search terms in the search term group are retrieved on different dates, the matching retrieval date for the search term group is determined;

[0056] Whether a search term group is a matching term group is determined based on the number of matching search dates for that search term group.

[0057] Furthermore, the matching retrieval date for the search term group is the date on which the percentage of the number of retrievals of the search terms in the search term group is greater than the percentage of the number of retrievals of the user.

[0058] It is understood that when the number of matching search dates for the search term group is greater than the preset number of matching dates, the search term group is determined to be a matching term group.

[0059] Specifically, the method for determining the user's matching phrases is as follows:

[0060] The number of times the search terms in different search term groups were processed on different dates was determined by analyzing the distribution data of search terms in different search term groups on different dates.

[0061] Based on the average number of retrievals of the search terms in the search term group on different dates, it is determined whether the search term group is a matching term group.

[0062] S2 determines the stability of the user profile based on the changes in the search terms of the matching words of the user across different dates. If the stability meets the requirements, proceed to the next step.

[0063] Furthermore, the changes in the search terms for matching phrases between different dates are determined based on the changes in the number of times the search terms for matching phrases are processed between different dates.

[0064] It should be noted that, as Figure 3 As shown, determining whether the stability of the user profile meets the requirements specifically includes:

[0065] Based on the user's search terms for matching word groups across different dates, determine the matching word groups corresponding to different dates;

[0066] Based on the matching words corresponding to different dates, the dates are divided into different matching word date groups;

[0067] Based on the number of dates in different matching term date groups, determine whether the stability of the user profile meets the requirements.

[0068] Furthermore, the dates in the matching phrase date group are the same dates as the matching phrases.

[0069] It is understandable that when there is no matching date group with a date quantity greater than the preset percentage threshold, the stability of the user profile of the user is determined to be unsatisfactory.

[0070] Specifically, when the stability of the user profile does not meet the requirements, the variation coefficient of the user profile is determined by multiplying the proportion of the number of dates in different matching word date groups, and the number of matching stocks is determined based on the preset selection number corresponding to the variation coefficient.

[0071] It should be noted that the number of stocks selected for matching users whose user profiles do not meet the stability requirements is greater than the number of stocks selected for matching users whose user profiles meet the stability requirements.

[0072] In another possible embodiment, determining that the stability of the user profile meets the requirements specifically includes:

[0073] Based on the user's search terms for matching word groups across different dates, determine the matching word groups corresponding to different dates;

[0074] The profile stability coefficient for different matching phrases is determined based on the percentage of dates corresponding to different matching phrases.

[0075] By using the profile stability coefficient of different matching word groups, it is determined whether the stability of the user profile meets the requirements.

[0076] Furthermore, if there are no matching word groups whose profile stability coefficient does not meet the requirements, then it is determined that the stability of the user profile does not meet the requirements.

[0077] In another possible embodiment, determining that the stability of the user profile meets the requirements specifically includes:

[0078] S31 uses the search terms of the user's matching word groups across different dates to determine the matching word groups corresponding to different dates, and determines the profile stability coefficient of different matching word groups based on the proportion of dates corresponding to different matching word groups.

[0079] S32 determines the retrieval distribution dispersion coefficient between dates in the matching term group based on the number of retrievals of the search terms in the matching term group corresponding to the dates in different matching term group date groups;

[0080] S33 determines the weight coefficient of different matching term date groups based on the proportion of dates in different matching term date groups, and determines the user profile stability coefficient of the user by combining the retrieval distribution dispersion coefficient of different matching term date groups, and uses the user profile stability coefficient to determine whether the stability of the user profile meets the requirements.

[0081] S3 determines the matching coefficient of the stock and the matching stock based on the matching results of the stock with different matching word groups and the distribution data of the matching word groups on different dates;

[0082] It should be noted that the matching results of the stock with different matching terms are determined based on the business scope, company name, and product matching results of the company corresponding to the stock with the search terms.

[0083] Specifically, such as Figure 4 As shown, the method for determining the matching coefficient of the stock is as follows:

[0084] The percentage of matches between the stock and different matching phrases is determined based on the matching results of the stock with different matching phrases, and this percentage is used as the matching coefficient between the stock and different matching phrases.

[0085] Based on the distribution data of search terms with different matching word groups on different dates, the proportion of dates in which the search terms with the matching word groups exist is determined, and this proportion is used as the weight coefficient of the matching word groups.

[0086] The matching coefficient of the stock is determined based on the matching coefficients of the stock with different matching word groups and the weight coefficients of different matching word groups.

[0087] Furthermore, the matching coefficient of the stock ranges from 0 to 1, and when the matching coefficient of the stock is greater than a preset matching coefficient threshold, the stock is determined to be a matching stock.

[0088] S4 generates multiple matching stock groups based on different matching coefficients of matching stocks. Based on the composition data of different matching stock groups and the matching situation of the user's search terms on different dates, the number of matching stocks to be selected is determined. Based on the number of selections and the financial data of the matching stocks, a quantitative model is used to determine the ranking result of the user's matching stocks.

[0089] It should be noted that multiple matching stock groups are generated, specifically including:

[0090] Based on the matching coefficients of different matching stocks, from largest to smallest, the selection results of matching stock groups with different numbers of matching stocks are determined;

[0091] The results of selecting matching stock groups with different numbers of matching stocks will be combined to generate multiple matching stock groups.

[0092] Specifically, if there are stocks A / B / C with matching coefficients of 0.6, 0.7, and 0.8 respectively, then the matching stock for a stock group with a quantity of 1 is C, the matching stock for a stock group with a quantity of 2 is B / C, and the matching stock for a stock group with a quantity of 3 is A / B / C.

[0093] Specifically, such as Figure 5 As shown, the method for determining the number of stocks to be matched is as follows:

[0094] Based on the matching status of the matched stocks in the matched stock group and the user's search terms on different dates, determine the search terms that have matched stocks and the search terms that do not have matched stocks on different dates, and determine the stock matching deviation coefficient of the matched stock group on different dates based on the proportion of search terms that do not have matched stocks on the date.

[0095] The effective matching stock groups in the matching stock groups are determined based on the average stock matching deviation coefficient of the matching stock groups on different dates;

[0096] Based on the composition data of the effective matching stock groups, the number of stocks in the effective matching stock groups is determined, and the optimal stock group in the effective matching stock groups is determined based on the number of stocks and the deviation of the enterprise types corresponding to the stocks in different matching stock groups. The number of matching stocks selected is determined based on the number of matching stocks in the optimal stock group.

[0097] Furthermore, the effective matching stock group in the matching stock group is the effective matching stock group whose average stock matching deviation coefficient on different dates meets the requirements.

[0098] It should be noted that the method for determining the optimal stock group is as follows:

[0099] Based on the deviation of the enterprise type corresponding to the stocks in the effective matching stock group, identify stocks whose business scope overlaps more than a preset overlap threshold and classify them into the same overlapping stock group.

[0100] The optimal stock group in the effective matching stock group is determined based on the number of overlapping stock groups in the effective matching stock group and the stock data in the different overlapping stock groups.

[0101] A further technical solution is that the optimal stock group is the effective matching stock group with the fewest overlapping stock groups reaching the target group number threshold.

[0102] Optionally, the type of enterprise is determined based on the business scope of the enterprise corresponding to the stock.

[0103] It is understandable that the ranking of the user's matched stocks using a quantitative model includes the following specific steps:

[0104] Using the selected quantity as a constraint, the matching stock selection result of the user is determined based on the matching coefficients of different matching stocks from large to small.

[0105] Based on the financial data of the matched stock selection results, the update time of the financial data of the matched stock selection results is determined, and the analysis results of the financial data are used to determine the financial data in different dimensions.

[0106] Based on the update time of financial data for different stock selection results and financial data in different dimensions, a quantitative model is used to determine the effective coefficient of financial data for different stock selection results.

[0107] The ranking results of different matching stock selections are determined by the validity coefficient of the financial data.

[0108] Furthermore, the different dimensions of financial data include debt-to-equity ratio, current ratio, quick ratio, accounts receivable turnover, inventory turnover, return on equity, sales profit, and cost-to-profit ratio.

[0109] It should be noted that the quantization model is constructed using one or more of the following: BP neural network, LSTM neural network, and RNN neural network.

[0110] Example 2

[0111] Optionally, the method for determining the number of stocks to be matched is as follows:

[0112] S41 determines the search terms that have matching stocks and do not have matching stocks on different dates based on the matching status of the matching stocks in the matching stock group and the search terms of the user on different dates. Then, it determines the stock matching coefficient of the matching stock group on different dates by combining the number of search processing times of the search terms that have matching stocks and the search terms that do not have matching stocks. Based on the stock matching coefficient of the matching stock group on different dates, it determines the group matching coefficient of the matching stock group.

[0113] S42 determines the number of stocks in the matching stock group based on the composition data of the matching stock group, and determines the stock composition matching coefficient of the matching stock group by combining the deviation of the enterprise type corresponding to the stocks in the matching stock group.

[0114] S43 determines the comprehensive matching coefficient in the matching stock group using the group matching coefficient and the stock composition matching coefficient, determines the optimal stock group in the matching stock group based on the comprehensive matching coefficient, and determines the number of matching stocks to be selected based on the number of matching stocks in the optimal stock group.

[0115] Optionally, the type of enterprise is determined based on the business scope of the enterprise corresponding to the stock.

[0116] Optionally, the overall matching coefficient is determined based on the ratio of the group matching coefficient to the stock composition matching coefficient.

[0117] Optionally, the method for determining the matching coefficient of the stock composition is as follows:

[0118] Based on the deviation of the enterprise types corresponding to the stocks in the matched stock group, the deviation coefficient of the corresponding enterprise types between the stocks in the matched stock group is determined, and the average deviation coefficient is determined by the average of the deviation coefficients of the corresponding enterprise types between different stocks.

[0119] The matching coefficient of the stock composition of the matching stock group is determined by multiplying the average deviation coefficient by the number of stocks in the matching stock group.

[0120] Optionally, the optimal stock group in the matched stock group is the matched stock group with the largest comprehensive matching coefficient.

[0121] Optionally, step S41 above includes the following:

[0122] S411 uses the matching status of the matched stocks in the matched stock group with the user's search terms on different dates to determine the search terms that have matched stocks and the search terms that do not have matched stocks on different dates. The number of search terms that do not have matched stocks is used to determine the search term matching deviation dates on the dates. When the number of search term matching deviation dates of the matched stock group does not meet the requirements, it is determined that the matched stock group does not belong to the optimal stock group. When the number of search term matching deviation dates of the matched stock group meets the requirements, proceed to the next step.

[0123] It should be noted that for stock groups with a large number of dates that deviate from the search terms, it indicates that their matching degree with the user's search terms is low. Therefore, without combining other data, it can be directly determined that they do not belong to the optimal stock group.

[0124] It is understandable that the search term match deviation date is the date with a large number of search terms that do not match stocks, and the specific date can be determined using a fixed threshold.

[0125] Specifically, to determine whether the number of dates with matching deviations for the search terms in a stock group meets the requirements, a fixed threshold for the number of dates with matching deviations for the search terms can be used.

[0126] By filtering stock groups with significant date discrepancies in their search term matching, most of the matching stock groups can be excluded, thereby improving the efficiency of filtering the optimal stock groups.

[0127] S412 determines the total number of search terms that do not match stocks based on the number of search terms that do not match stocks on different dates. If the total number of search terms that do not match stocks does not meet the requirements, it is determined that the matching stock group does not belong to the optimal stock group. If the total number of search terms that do not match stocks meets the requirements, proceed to the next step.

[0128] Understandably, if a large number of search terms do not match the stocks in a stock matching group, it indicates that the stock matching group and the user's search terms have a low degree of matching, and therefore these stocks can be directly excluded.

[0129] Specifically, the total number of search terms that do not match any stocks can be compared with a preset fixed threshold to determine whether the total number of search terms that do not match any stocks meets the requirements.

[0130] S413 determines the stock matching coefficient of the stock matching group for different dates based on the search terms that match stocks and the search terms that do not match stocks on different dates, and in combination with the number of search processing times for the search terms that match stocks and the search terms that do not match stocks. When there are dates where the stock matching coefficient does not meet the requirements, proceed to step S414. When there are no dates where the stock matching coefficient does not meet the requirements, proceed to step S415.

[0131] As an illustration, when there are dates where the stock matching coefficient does not meet the requirements, it indicates that there are dates where the user's matching degree with the stock group is low. For dates with low matching degree, the number of dates with low matching degree can be further combined to determine whether the matching degree with the user profile is difficult to meet the requirements.

[0132] The stock matching coefficient can be determined using mathematical models such as neural networks and analytic hierarchy process (AHP). The number of retrieval processes for each matching stock term and each non-matching stock term is taken as input.

[0133] Specifically, when the stock matching coefficient is less than a preset fixed threshold, the date can be directly determined to be a date where the stock matching coefficient does not meet the requirements.

[0134] S414 When the percentage of dates for which the stock matching coefficient does not meet the requirements is greater than the preset value of the percentage of dates, it is determined that the matched stock group does not belong to the optimal stock group. When the percentage of dates for which the stock matching coefficient does not meet the requirements is not greater than the preset value of the percentage of dates, proceed to the next step.

[0135] It is understandable that when there are a large number of dates where the stock matching coefficient does not meet the requirements, that is, when the proportion of dates where the stock matching coefficient does not meet the requirements is greater than the preset value of the proportion of dates, it can be directly determined that the matching degree between the matched stock group and the user profile is difficult to meet the requirements.

[0136] By further filtering matching stock groups that are difficult to meet the matching requirements, the efficiency of screening the optimal stock group is further improved.

[0137] S415 determines the group matching coefficient of the matching stock group based on the stock matching coefficient of the matching stock group on different dates. When the group matching coefficient of the matching stock group meets the requirements, proceed to step S42. When the group matching coefficient of the matching stock group does not meet the requirements, it is determined that the matching stock group does not belong to the optimal stock group.

[0138] Optionally, step S42 above includes the following:

[0139] S421 Based on the composition data of the matched stock group, determine the deviation of the enterprise type corresponding to the stock in the matched stock group, and determine the deviation coefficient of the enterprise type corresponding to the stock in the matched stock group based on the deviation. When the deviation coefficient of the enterprise type corresponding to different stocks is less than the preset deviation coefficient threshold, it is determined that the matched stock group does not belong to the optimal stock group. When the deviation coefficient of the enterprise type corresponding to different stocks is not less than the preset deviation coefficient threshold, proceed to step S422.

[0140] It is understandable that when the deviation coefficient of the company type of the stocks in the matching stock group is less than the preset deviation coefficient threshold, it means that the company types of the stocks in the matching stock group are relatively similar. Therefore, if they are recommended, the representative value of the matching stock group to the matching stocks will be too small, so it can be directly determined that it does not belong to the optimal stock group.

[0141] Specifically, the deviation coefficient for enterprise type is determined based on the number of deviations in the business scope of the enterprise corresponding to the stock. The more deviations there are, the larger the deviation coefficient will be.

[0142] S422 When the deviation coefficients of different stocks corresponding to different enterprise types are all within the preset deviation coefficient range, proceed to step S423; when the deviation coefficients of different stocks corresponding to different enterprise types are not all within the preset deviation coefficient range, proceed to step S424.

[0143] It should be noted that when the deviation coefficients of different stocks corresponding to different company types are all within the preset deviation coefficient range, it means that although the different stocks correspond to different company types, they are still related to a certain extent. At the same time, if there are a large number of stocks, it can be directly determined that they do not belong to the optimal stock group.

[0144] Therefore, when it is determined that the stock is within the preset deviation coefficient range, it is necessary to directly proceed to step S423 to further determine the number of stocks in the effective matching stock group.

[0145] S423 Determine the number of stocks in the effective matching stock group. If the number of stocks in the effective matching stock group is greater than a preset stock number threshold, then determine that the matching stock group does not belong to the optimal stock group. If the number of stocks in the effective matching stock group is not greater than the preset stock number threshold, proceed to step S424.

[0146] Understandably, the preset stock quantity threshold can be determined through a preset method, or it can be determined based on the number of stocks matched by the user. Specifically, it can be determined by multiplying the number of matched stocks by a preset ratio factor.

[0147] S424 determines the stock composition matching coefficient of the effective matching stock group based on the deviation of the number of stocks in the effective matching stock group and the company type corresponding to the stocks in the matching stock group. When the stock composition matching coefficient of the effective matching stock group is within the preset composition matching coefficient range, it is determined that the matching stock group does not belong to the optimal stock group. When the stock composition matching coefficient of the effective matching stock group is not within the preset composition matching coefficient range, proceed to step S43.

[0148] Example 3

[0149] It should be further explained that the following conditions apply to proceeding to step S32:

[0150] When the number of dates corresponding to different matching word groups is within the preset date number range, or when there are no matching word groups that meet the requirements for the stability coefficient of the user profile, it can be directly determined that the stability of the user profile does not meet the requirements.

[0151] When there are matching word groups that meet the requirements for image stability coefficient and the number of dates corresponding to different matching word groups is not uniformly within the preset date number range, proceed to step S32.

[0152] By first determining the profile stability coefficient of the matching group, we can first screen users whose distribution is relatively discrete in different matching word groups. For users whose distribution is relatively discrete, the stability of their user profile will inevitably not meet the requirements.

[0153] For example, before proceeding to step S33, it is necessary to further determine whether there is a matching word group date group whose retrieval distribution dispersion coefficient meets the requirements. If there is no matching word group date group whose retrieval distribution dispersion coefficient meets the requirements, it can be directly determined that the stability of the user profile does not meet the requirements. If and only if there is a matching word group date group whose retrieval distribution dispersion coefficient meets the requirements, proceed to step S33.

[0154] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0155] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0156] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A stock ranking and selection method based on a quantitative model, characterized in that, Specifically, it includes: The search terms are divided into different search term groups based on the similarity between the users' search terms, and the matching word groups for the users are determined based on the distribution data of the search terms in the different search term groups on different dates; Based on the changes in the search terms matching the user's terms across different dates, once it is determined that the stability of the user profile meets the requirements, proceed to the next step; Based on the matching results of stocks with different matching word groups and the distribution data of different matching word groups on different dates, the matching coefficient of the stock and the matching stocks are determined. The search terms are divided into different search term groups, specifically including: By matching different search terms with different stocks, we can identify stocks that match the same search terms simultaneously and treat them as similar matching stocks. By utilizing the number of similar matching stocks, the search terms are divided into different search term groups, and search terms with a number of similar matching stocks greater than a preset number of stocks are divided into the same search term group; The method for determining the user's matching phrases is as follows: The number of times the search terms in different search term groups were processed on different dates was determined by analyzing the distribution data of search terms in different search term groups on different dates. Based on the number of times the user retrieves data on different dates and the number of times the search terms in the search term group are retrieved on different dates, the matching retrieval date for the search term group is determined; Whether a search term group is a matching term group is determined based on the number of matching search dates for that search term group; When the number of matching search dates for the search term group is greater than the preset number of matching dates, the search term group is determined to be a matching term group; Determining that the stability of the user profile meets the requirements specifically includes: Based on the user's search terms for matching word groups across different dates, determine the matching word groups corresponding to different dates; Based on the matching words corresponding to different dates, the dates are divided into different matching word date groups; Based on the number of dates in different matching term date groups, determine whether the stability of the user profile meets the requirements.

2. The stock ranking and selection method based on a quantitative model as described in claim 1, characterized in that, The matching of the search terms with different stocks is determined based on the matching of the search terms with the business scope, company name, and products of the companies corresponding to the stocks.

3. The stock ranking and selection method based on a quantitative model as described in claim 1, characterized in that, The matching retrieval date for the search term group is the date on which the percentage of the number of retrievals of the search terms in the search term group is greater than the preset percentage of the number of retrievals.

Images