An information publishing and query system for financial services

Abstract

The present invention discloses an information publishing and query system for financial services, which belongs to the technical field of financial services, and comprises a demand module, a collection module, a data processing module and a retrieval module; the demand module is used for a user to input a corresponding demand range, generates a corresponding demand target based on the demand range input by the user, and displays the generated demand target; the collection module is used for collecting financial information according to each demand target, and obtains the financial information corresponding to each demand target; the data processing module is used for processing the financial information of each demand target and establishing a financial database; the financial information in each storage node is processed for meaning representation, and each recommended display data corresponding to each demand target is obtained; the retrieval module is used for a user to query financial information, identifies the query keywords input by the user, obtains the query data according to the query keywords, and displays the query data to the user.

Description

Technical Field

[0001] The invention belongs to the technical field of financial services, and in particular is an information publishing and query system for financial services. Background Art

[0002] Financial information refers to the signals, instructions, data, situations, and messages generated in the process of organizing and managing currency circulation, various financial securities transactions, credit activities, and fund settlement. Its content includes financial information of relevant departments, exchange rate information, securities market information, customer credit information, etc.

[0003] However, it is currently difficult for many individual users to fully understand relevant financial information, and they can often only learn about relevant information through a few channels. Due to the lack of channels, it is difficult to accurately assist users in financial transactions.

[0004] Based on this, in order to help a large number of individual users to fully understand the required financial information, the present invention provides an information publishing and query system for financial services. Summary of the invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an information publishing and query system for financial services.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An information publishing and query system for financial services, comprising a demand module, a collection module, a data processing module, and a retrieval module;

[0008] The demand module is used for the user to input the corresponding demand range, generate the corresponding demand target based on the demand range input by the user, and display the generated demand target;

[0009] The collection module is used to collect financial information according to each demand target and obtain the financial information corresponding to each demand target.

[0010] Furthermore, the working method of the acquisition module includes:

[0011] Retrieve all information channels with financial information published by the target of demand, screen each of the information channels according to data repeatability and information channel credibility, and determine each data collection channel corresponding to each of the target of demand;

[0012] Data collection is performed based on each data collection channel corresponding to each of the demand targets to obtain corresponding initial data; the obtained initial data is deduplicated to obtain financial information corresponding to the demand target.

[0013] The data processing module is used to process the financial information of each of the demand targets, obtain the financial information corresponding to each demand target, establish a financial database, set storage nodes corresponding to each demand target in the financial database, and store the obtained financial information in the corresponding financial database in the financial database;

[0014] Perform meaning representation processing on the financial information in each storage node to obtain various recommended display data corresponding to each demand target; classify and display each recommended display data according to each demand target.

[0015] Furthermore, optimize the storage management of the financial database, identify various optimization methods available to the user, evaluate each optimization method, determine the corresponding target optimization method, and perform storage optimization based on the target optimization method.

[0016] Furthermore, the method for evaluating each optimization method includes:

[0017] Perform efficiency simulation on each optimization method to form corresponding storage efficiency and scheduling efficiency;

[0018] According to the formula Calculate the corresponding optimization value;

[0019] In the formula: YZ is the optimization value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 < 1, 0 < b2 < 1; XC is the storage efficiency; XD is the scheduling efficiency; CB is the optimization cost;

[0020] Select the optimization method with the highest optimization value as the target optimization method.

[0021] Furthermore, the method for performing meaning representation processing on the financial information in each storage node includes:

[0022] Identify each piece of financial data in the financial information and determine the financial meaning of each piece of financial data; establish mutual associations between the financial data according to each financial meaning to form several association chains;

[0023] Identify the number of connections corresponding to each financial meaning in the association chain; set corresponding adjustment coefficients according to each financial meaning;

[0024] Calculate the corresponding representative value according to the formula LMTi = λi × Ni; in the formula: Ni represents the number of connections corresponding to the corresponding financial meaning; i represents the corresponding financial meaning, i = 1, 2,..., n, n is a positive integer; λi is the adjustment coefficient corresponding to the financial meaning; LMZi represents the representative value of the corresponding financial meaning;

[0025] Select the financial data corresponding to the financial meaning with the highest representative value as the main display data, and mark the financial data of non-main display data in the association chain as auxiliary associated data; integrate the obtained main display data and auxiliary associated data into the recommended display data for the demand target.

[0026] Further,

[0027] In the formula: λi is the adjustment coefficient corresponding to the financial meaning; Lzi is the historical similar view volume corresponding to the financial meaning.

[0028] The retrieval module is used for users to query financial information, identify the query keywords input by users, obtain the corresponding retrieval information from the financial database according to the obtained query keywords; determine the association chain corresponding to each retrieval information, and obtain the initial query data based on the association chain; the initial query data is composed of each unit data;

[0029] Retrieve the interpretation data corresponding to each unit data, evaluate the obtained interpretation data, and obtain the corresponding interpretation value; eliminate the interpretation data with the interpretation value lower than the threshold X2; and deduplicate the interpretation data based on the interpretation value.

[0030] Sort in descending order according to the interpretation to form the interpretation statistical data, and make the corresponding association between the obtained interpretation statistical data and the corresponding unit data; obtain the query data and display the query data to the user.

[0031] Further, the method for evaluating the interpretation data includes:

[0032] Set the corresponding quality value according to the interpretation data;

[0033] Identify the release source of each interpretation data; obtain the historical interpretation data release record of the release source, identify the quality value and result coefficient corresponding to each historical interpretation data in the historical interpretation data release record, and the value range of the result coefficient is [-1, 1]; mark each historical interpretation data as v, v = 1, 2,..., q, q is a positive integer; mark the obtained historical quality value as DELv; mark the obtained result coefficient as μv;

[0034] According to the formula Calculate the corresponding source value; in the formula: LYZ is the source value;

[0035] Calculate the corresponding interpretation value according to the formula PKR = b3×DF + b4×LYZ; in the formula: PKR is the interpretation value; b3 and b4 are both proportionality coefficients, and the value range is 0 < b3 < 1, 0 < b4 < 1; DF is the quality value.

[0036] Further, the method for setting the quality value includes:

[0037] Set a quality assessment template, which includes various quality assessment items and assessment standards; evaluate the interpretation data based on the preset quality assessment template to obtain the single item value corresponding to each quality assessment item; mark the obtained single item value as DEj; j represents the corresponding quality assessment item, j = 1, 2, ..., m, m is a positive integer; according to the formula Calculate the corresponding mass value; where: DF is the mass value.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] Through the cooperation among the demand module, acquisition module, data processing module, and retrieval module, comprehensive financial information is provided to users, and corresponding interpretation data is supplemented to facilitate users' interpretation of financial data. Through the cooperation among the demand module, acquisition module, and data processing module, the optimization processing of a large amount of financial data is realized to solve the problem of information overload for individual users; at the same time, in order to avoid the problem of insufficient storage space and information overload caused by a large amount of data, the storage management of the financial database is optimized to improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 This is a principle block diagram of the present invention;

[0042] Figure 2 This is an example diagram of the association chain of the present invention. DETAILED DESCRIPTION

[0043] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] like Figure 1 to Figure 2 As shown, an information publishing and query system for financial services includes a demand module, a collection module, a data processing module, and a retrieval module;

[0045] The demand module is used for users to input the corresponding demand range according to their own needs, such as the needs of a certain enterprise, stock, fund, industry, field and other related ranges; generate corresponding demand targets according to the user's demand range, and display the obtained demand targets; it is convenient for users to clarify whether each demand target is correct and to adjust the demand targets.

[0046] The collection module is used to collect financial information according to each demand target, determine the corresponding data collection channel according to each demand target, and the data collection channel is used to collect the real-time published financial information of the corresponding demand target; based on each data collection channel, the financial information of each demand target is collected in real time to obtain the financial information corresponding to each demand target.

[0047] Among them, there are many ways to determine the data collection channel, such as the common selection by humans based on their experience in the prior art, or other selection methods in the prior art; based on the background considerations of the comprehensiveness and accuracy of financial data, the present invention proposes the following method.

[0048] Retrieve all information channels with financial information published by the target of demand, evaluate the credibility of each information channel, the credibility value range is [0, 100], preset the standard information channel data corresponding to each credibility value such as 0, 10, 20, ..., 100, indicating the relevant credibility of the information channel of the credibility value, and then locate the corresponding standard credibility value according to the data matching, and then make corrections according to the corresponding difference, and set the credibility value in combination with the corresponding interpolation method; or use the existing information authenticity evaluation system, technology, etc. to evaluate; for example, according to the existing common sense in the field, first determine the credibility value level interval to which it belongs, and then confirm the corresponding specific credibility value according to the corresponding difference, with a larger operating error range;

[0049] Information channels with credibility values ​​lower than the threshold value X1 are eliminated, and the remaining information channels are screened according to the repeatability of financial information. That is, when the financial information related to the demand target in a certain information channel is the same as the corresponding financial information in other information channels, the information channels with low credibility values ​​are eliminated; the remaining information channels are marked as data collection channels.

[0050] Exemplarily, the method for collecting financial information of each demand target based on each data collection channel is as follows:

[0051] Financial information that meets the demand target is collected from various data collection channels in real time, marked as initial data, and the obtained initial data is deduplicated, the same or similar data is eliminated, the initial data corresponding to the data collection channels with high credibility values ​​are retained, and the output is the financial information of the demand target.

[0052] The data processing module is used to process the financial information of each demand target collected by the collection module, obtain the financial information corresponding to each demand target, establish a financial database, set storage nodes corresponding to each demand target in the financial database, and store the obtained financial information in the corresponding financial database in the financial database;

[0053] Perform meaning representation processing on the financial information in each storage node to obtain each recommended display data corresponding to each demand target; classify and display each recommended display data according to each demand target.

[0054] In one embodiment, due to differences in the equipment and other conditions of individual users, a large amount of data will cause situations such as insufficient user storage space and information overload. Therefore, it is necessary to optimize the storage management of the financial database; the specific optimization method is as follows:

[0055] Identify various optimization methods available to the user, which need to be analyzed according to the user's actual conditions, such as external storage, various cloud disks, storage data retention duration, deletion, and other methods, not limited to one optimization method; evaluate each optimization method to determine the corresponding target optimization method, and perform storage optimization based on the target optimization method.

[0056] Among them, the methods for evaluating each optimization method include:

[0057] Perform efficiency simulation on each optimization method to form corresponding storage efficiency and scheduling efficiency, that is, perform simulation operations of storing and retrieving data of a preset size, obtain the corresponding storage duration and retrieval duration, calculate with the corresponding data size, and obtain the corresponding storage efficiency and scheduling efficiency;

[0058] According to the formula Calculate the corresponding optimization value; in the formula: YZ is the optimization value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 < 1, 0 < b2 < 1; XC is the storage efficiency; XD is the scheduling efficiency; CB is the estimated optimization cost of applying this optimization method.

[0059] Select the optimization method with the highest optimization value as the target optimization method, but it can be manually adjusted and determined according to the user's needs.

[0060] The methods for performing meaning representation processing on the financial information in each storage node include:

[0061] Identify each piece of financial data in the financial information, determine the financial meaning of each piece of financial data, that is, the meaning it wants to express, or the goal of publishing the financial data, and use the existing data analysis system to determine; exemplarily, establish a corresponding meaning analysis model based on a CNN network or a DNN network, and establish a corresponding training set for training by artificial means, the training set includes input data and output data, the input data is each piece of financial data set in simulation, and the output data is the corresponding financial meaning; analyze through the meaning analysis model after successful training to determine the financial meaning of each piece of financial data;

[0062] According to the financial meaning of each financial data, the financial data are related to each other to form several association chains; that is, first determine whether their financial meanings are related, and if so, they are considered to be related. For example, the financial statement data for the whole year released by a certain enterprise may be related to the financial statement data of each quarter, the financial data of each specific item, etc.; determine whether they are related according to the relevance of their meanings; the corresponding financial meaning association table can be preset based on the experience of professionals, and then corresponding identification and matching can be performed; after determining the financial meanings of each correlation, they are connected according to their corresponding correlation relationship to form an association chain; such as Figure 2 As shown, in the figure, meaning 1 etc. represent the corresponding financial meanings, and the corresponding lines represent the corresponding associations, thereby forming an association chain.

[0063] Identify the number of links corresponding to each financial meaning in the association chain, the number of links connected to the financial meaning;

[0064] According to each financial meaning, the corresponding adjustment coefficient is set according to the number of users' browsing of the corresponding historical financial data of the same type, and the corresponding adjustment coefficient is set according to the corresponding browsing proportion; the corresponding browsing volume is marked as the historical similar browsing volume, and each financial meaning is marked as i, i = 1, 2, ..., n, n is a positive integer; the corresponding historical similar browsing volume is marked as Lzi; according to the formula Calculate the corresponding adjustment coefficient; where: λi is the adjustment coefficient corresponding to the financial meaning;

[0065] The corresponding representative value is calculated according to the formula LMTi=λi×Ni; wherein: Ni represents the number of lines corresponding to the corresponding financial meaning; LMZi represents the representative value of the corresponding financial meaning.

[0066] The financial data corresponding to the financial meaning with the highest representative value is selected as the main display data, and the corresponding financial data in the association chain that is not the main display data is marked as auxiliary associated data; the main display data and auxiliary associated data obtained are integrated into the recommended display data for the demand target, that is, the displayed data is the main display data, but a quick access link to the auxiliary associated data will be inserted in the main display data for quick access according to available needs.

[0067] Through the cooperation among demand module, collection module and data processing module, we can optimize the processing of large amounts of financial data and solve the problem of information overload for individual users. At the same time, in order to avoid problems such as insufficient storage space and information overload caused by large amounts of data, we optimize the storage management of financial databases and improve the user experience.

[0068] The retrieval module is used for users to query financial information, identify the query keywords input by the user, retrieve the corresponding financial information from the financial database according to the obtained query keywords, and mark it as retrieval information. The above process is implemented based on existing retrieval technology; determine the association chain corresponding to each retrieval information, obtain other financial data in the association chain, and integrate it into initial query data in combination with the corresponding retrieval data; mark each financial data in the initial query data as unit data;

[0069] Retrieve the interpretation data corresponding to each unit data, that is, the existing interpretation data for the financial data, evaluate each obtained interpretation data, and obtain the corresponding interpretation value; remove the interpretation data with an interpretation value lower than the threshold value X2; and remove the duplication of the interpretation data based on the interpretation value, and remove the interpretation data with a low interpretation value for the same interpretation data;

[0070] Sort the interpretations from high to low to form interpretation statistics, and associate the obtained interpretation statistics with the corresponding unit data;

[0071] The current initial query data is marked as query data, and the query data is displayed to the user.

[0072] Methods for evaluating the data from each interpretation include:

[0073] The corresponding quality value is set according to the interpretation data. The quality value is evaluated according to the specific content of the interpretation data. There are several quality evaluation items preset, such as the application of professional terms, logic, content richness and other angles to set the quality evaluation items. Specifically, the expert group discusses and sets the corresponding quality evaluation template for each field, including each quality evaluation item and evaluation standard; the interpretation data is evaluated based on the preset quality evaluation template to obtain the corresponding single item value of each quality evaluation item; the obtained single item value is marked as DEj; j represents the corresponding quality evaluation item, j = 1, 2, ..., m, m is a positive integer; according to the formula Calculate the corresponding mass value; where: DF is the mass value.

[0074] Identify the publishing source of each interpretation data, which refers to the corresponding publishing account; obtain the historical interpretation data publishing record of the publishing source, identify the quality value and result coefficient corresponding to each historical interpretation data in the historical interpretation data publishing record, the result coefficient is set according to the final result of the historical interpretation data, the interpretation is correct, the interpretation is wrong, and the results in between, the value range of the result coefficient is [-1, 1], -1 means the interpretation is completely wrong, 1 means all the interpretations are correct; others are assigned values ​​according to the differences, several reference standards can be preset in the middle, and then matched based on the reference standards to reduce the error, and multiple reference standards can be set continuously to further reduce the error until the accuracy requirement is met; each historical interpretation data is marked as v, v = 1, 2, ..., q, q is a positive integer; the obtained historical quality value is marked as DELv; the obtained result coefficient is marked as μv;

[0075] According to the formula Calculate the corresponding source value; where LYZ is the source value.

[0076] The corresponding interpretation value is calculated according to the formula PKR = b3 × DF + b4 × LYZ; where: PKR is the interpretation value; b3 and b4 are both proportional coefficients, with a value range of 0 <b3<1,0<b4<1。

[0077] Through the mutual cooperation among the demand module, collection module, data processing module and retrieval module, it is possible to provide users with comprehensive financial information and supplement corresponding interpretation data to facilitate users' interpretation of financial data.

[0078] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0079] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An information publishing and query system for financial services, characterized in that: It includes a requirements module, a collection module, a data processing module, and a retrieval module; The requirements module is used for the user to input the corresponding requirements scope, generate the corresponding requirements target based on the requirements scope input by the user, and display the generated requirements target; The collection module is used to collect financial information according to each requirements target and obtain the financial information corresponding to each requirements target; The data processing module is used to process the financial information of each requirements target, obtain the financial information corresponding to each requirements target, establish a financial database, set storage nodes corresponding to each requirements target in the financial database, and store the obtained financial information in the corresponding financial database in the financial database; Perform meaning representation processing on the financial information in each storage node to obtain the recommended display data corresponding to each requirements target; classify and display each recommended display data according to each requirements target; The retrieval module is used for the user to query financial information, identify the query keywords input by the user, and obtain the corresponding retrieval information from the financial database according to the obtained query keywords; determine the association chain corresponding to each retrieval information, and obtain the initial query data based on the association chain; The initial query data is composed of each unit data; Retrieve the interpretation data corresponding to each unit data, evaluate the obtained interpretation data to obtain the corresponding interpretation value;剔除 the interpretation data with an interpretation value lower than the threshold X2; and perform deduplication of the interpretation data based on the interpretation value; Sort in descending order according to the interpretation to form the interpretation statistical data, and associate the obtained interpretation statistical data with the corresponding unit data; Obtain the query data and display the query data to the user; Optimize the storage management of the financial database, identify various optimization methods available to the user, evaluate each optimization method, determine the corresponding target optimization method, and perform storage optimization based on the target optimization method; The method for evaluating each optimization method includes: Perform efficiency simulation on each optimization method to form the corresponding storage efficiency and scheduling efficiency; According to the formula Calculate the corresponding optimization value; Where: YZ is the optimization value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 < 1, 0 < b2 < 1; XC is the storage efficiency; XD is the scheduling efficiency; CB is the optimization cost; Select the optimization method with the highest optimization value as the target optimization method; The method for performing meaning representation processing on the financial information in each storage node includes: Identify each financial data in the financial information and determine the financial meaning of each financial data; establish mutual associations between the financial data according to each financial meaning to form several association chains; Identify the number of connections corresponding to each financial meaning in the association chain; set the corresponding adjustment coefficient according to each financial meaning; According to the formula Calculate the corresponding representative value; where: Ni represents the number of lines corresponding to the corresponding financial meaning; i represents the corresponding financial meaning, i=1, 2, ..., n, n is a positive integer; λi is the adjustment coefficient of the corresponding financial meaning; LMZi represents the representative value of the corresponding financial meaning; Select the financial data corresponding to the financial meaning with the highest representative value as the main display data, and mark the financial data in the association chain that is not the main display data as auxiliary associated data; integrate the obtained main display data and auxiliary associated data into the recommended display data of the requirements target.

2. The information publishing and query system for financial services according to claim 1, characterized in that: The working method of the collection module includes: Retrieve all information channels for financial information with demand target releases, screen each of the information channels according to data repeatability and information channel credibility, and determine each data collection channel corresponding to each demand target; Based on each data collection channel corresponding to each demand target, collect data to obtain corresponding initial data; deduplicate the obtained initial data to obtain the financial information corresponding to the demand target.

3. The information publishing and query system for financial services according to claim 1, characterized in that: ； Where: λi is the adjustment coefficient corresponding to the financial meaning; Lzi is the historical similar view volume corresponding to the financial meaning.

4. The information publishing and query system for financial services according to claim 1, characterized in that: Set the corresponding quality value according to the interpretation data; Identify the release sources of each interpretation data; obtain the historical interpretation data release records of the release sources, and identify the quality values and result coefficients corresponding to each historical interpretation data in the historical interpretation data release records, and the value range of the result coefficient is [-1, 1]; Mark each historical interpretation data as v, v = 1, 2,..., q, q is a positive integer; mark the obtained historical quality value as DELv; mark the obtained result coefficient as μv; According to the formula Calculate the corresponding source value; where: LYZ is the source value; Calculate the corresponding interpretation value according to the formula PKR = b3×DF + b4×LYZ; where: PKR is the interpretation value; Both b3 and b4 are proportionality coefficients, and the value range is 0 < b3 < 1, 0 < b4 < 1; DF is the quality value.

5. The information publishing and query system for financial services according to claim 4, characterized in that: The method for setting the quality value includes: Set a quality assessment template, which includes various quality assessment items and assessment standards; evaluate the interpretation data based on the preset quality assessment template to obtain the single item value corresponding to each quality assessment item; mark the obtained single item value as DEj; j represents the corresponding quality assessment item, j=1, 2, ..., m, m is a positive integer; according to the formula Calculate the corresponding mass value; where: DF is the mass value.

Images