Financial product field updating method, device, equipment and storage medium

Abstract

The present invention relates to the field of character processing and discloses a method, apparatus, device, and storage medium for updating financial product fields. The method comprises: obtaining N updated product fields to generate an updated product field set, where N is a positive integer; performing numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm to generate a character feature data set and a conversion mapping hash table; performing priority identification processing on the character feature data set based on a preset KNN neighbor algorithm to obtain priority numbers corresponding to the N character feature data in the character feature data set; transferring the priority numbers corresponding to the N character feature data to the updated product field corresponding to each character feature data according to the conversion mapping hash table to obtain priority numbers corresponding to the N updated product fields; and performing replacement and update processing on N financial product usage fields corresponding to the N updated product fields based on the priority numbers corresponding to the N updated product fields to obtain updated N financial product usage fields.

Description

Technical Field

[0001] The present invention relates to the field of character processing, and in particular to a method, device, equipment and storage medium for updating a financial product field. Background Art

[0002] Under existing technology, a priority is generally defined for all data sources of all fields of financial products. When new data arrives, the relationship between the source of the existing data field and the source priority of the new data is first determined. If the priority of the new source is higher than that of the existing data, the data is updated; otherwise, it remains unchanged.

[0003] A table of priorities related to financial product fields and a table storing the update sources for all existing data fields are used. Each time old data is updated, both the field priority table and the source table of the fields to be updated in the existing data must be queried. The priority of the source of the field to be updated is then compared with the priority of the existing field source. If the source of the field to be updated has a higher priority, it is updated; otherwise, it remains unchanged. Financial product data comes from many sources, and different data sources have different focuses on financial products. This may result in Data Source 1 primarily updating Field A, while Data Source 2 primarily updates Field B. This requires breaking the entire financial product data into individual fields, recording the priorities of different data sources for each field, and then updating them accordingly. However, this update method, which updates data based on entire data fields, is too coarse-grained. The priority order of individual fields must be fixed, making it impossible to dynamically modify or automatically select priorities based on different product features. Therefore, a technology is needed to address the existing inability to dynamically and automatically modify fields based on priority. Summary of the Invention

[0004] The main purpose of the present invention is to solve the technical problem that fields cannot be modified dynamically and automatically according to priorities.

[0005] A first aspect of the present invention provides a method for updating a financial product field, the method comprising:

[0006] Get N updated product fields and generate an updated product field set, where N is a positive integer;

[0007] According to a preset conversion algorithm, the N updated product fields in the updated product field set are sequentially converted to numerical values ​​to generate a character feature data set and a conversion mapping hash table;

[0008] Based on a preset KNN neighbor algorithm, priority recognition processing is performed on the character feature data set to obtain priority numbers corresponding to N character feature data in the character feature data set;

[0009] According to the conversion mapping hash table, the priority numbers corresponding to the N character feature data are transferred to the update product field corresponding to each character feature data to obtain the priority numbers corresponding to the N update product fields;

[0010] Based on the priority numbers corresponding to the N updated product fields, the N financial product usage fields corresponding to the N updated product fields are replaced and updated to obtain the updated N financial product usage fields.

[0011] Optionally, in a first implementation of the first aspect of the present invention, after acquiring N updated product fields and generating an updated product field set, and before sequentially performing numerical conversion processing on the N updated product fields in the updated product field set according to a preset conversion algorithm to generate a character feature dataset and a conversion mapping hash table, the method further includes:

[0012] Read the data sources of N updated product fields and obtain N field source data;

[0013] According to the preset source value mapping table, the source data of N fields are mapped to obtain the binary strings corresponding to the source data of N fields;

[0014] The binary strings corresponding to the N field source data are set as labels of the updated product fields corresponding to the N field source data, thereby obtaining N updated product fields with binary labels.

[0015] Optionally, in a second implementation of the first aspect of the present invention, before setting the binary strings corresponding to the N field source data as labels for the updated product fields corresponding to the N field source data to obtain N updated product fields with source labels, and after setting the binary strings corresponding to the N field source data as labels for the updated product fields corresponding to the N field source data to obtain N updated product fields with binary labels, the method further includes:

[0016] The binary labels of the N updated product fields are converted to long integers in sequence to obtain N updated product fields with long integer labels.

[0017] Optionally, in a third implementation of the first aspect of the present invention, performing numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm to generate a character feature dataset and a conversion mapping hash table includes:

[0018] Obtaining a training field feature set, wherein each training field feature in the training field feature set includes: a training field and a conversion value;

[0019] Inputting the training field feature set into a preset neural network model for recognition training processing to obtain a recognition transformation matrix set;

[0020] According to the recognition conversion matrix set, the N updated product fields in the updated product field set are sequentially recognized and converted to obtain a character feature data set and a conversion mapping hash table.

[0021] Optionally, in a fourth implementation of the first aspect of the present invention, the priority identification processing is performed on the character feature dataset based on a preset KNN neighbor algorithm, and the priority numbers corresponding to the N character feature data in the character feature dataset are obtained, including:

[0022] Setting a priority number label for each training field feature in the training field feature set to obtain a level training field feature set;

[0023] Inputting the series training field feature set into a preset neural network model for series training processing to obtain a feature series matrix;

[0024] Based on a preset KNN neighbor algorithm, N character feature data in the character feature data set are sequentially subjected to distance determination processing with feature elements in the feature series matrix to obtain the feature element with the closest distance to the character feature data;

[0025] According to the correspondence between the feature elements and the level elements in the feature level matrix, the level element corresponding to the character feature data is determined as the priority number.

[0026] Optionally, in a fifth implementation of the first aspect of the present invention, after sequentially performing numerical conversion processing on the N updated product fields in the updated product field set according to a preset conversion algorithm to generate a character feature data set and a conversion mapping hash table, and before transferring the priority numbers corresponding to the N character feature data to the updated product fields corresponding to each character feature data according to the conversion mapping hash table to obtain the priority numbers corresponding to the N updated product fields, the method further includes:

[0027] Normalizing the character feature dataset to obtain a new character feature dataset.

[0028] Optionally, in a sixth implementation of the first aspect of the present invention, the N financial product usage fields corresponding to the N updated product fields are replaced and updated based on the priority numbers corresponding to the N updated product fields, and the updated N financial product usage fields obtained include:

[0029] Determine whether the priority number of the updated product field is greater than the priority number of the financial product usage field corresponding to the updated product field;

[0030] If it is greater, the financial product usage field corresponding to the updated product field is replaced by the updated product field;

[0031] If not, the financial product usage field corresponding to the updated product field will not be changed.

[0032] A second aspect of the present invention provides a device for updating a financial product field, the device comprising:

[0033] An acquisition module is used to acquire N updated product fields and generate an updated product field set, where N is a positive integer;

[0034] A conversion module, configured to perform numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm, and generate a character feature data set and a conversion mapping hash table;

[0035] an identification module, configured to perform priority identification processing on the character feature data set based on a preset KNN neighbor algorithm to obtain priority numbers corresponding to N character feature data in the character feature data set;

[0036] A mapping transfer module, configured to transfer the priority numbers corresponding to the N character feature data to the update product field corresponding to each character feature data according to the conversion mapping hash table, to obtain the priority numbers corresponding to the N update product fields;

[0037] The replacement update module is used to replace and update the N financial product usage fields corresponding to the N update product fields based on the priority numbers corresponding to the N update product fields, and obtain the updated N financial product usage fields.

[0038] A third aspect of the present invention provides a device for updating a financial product field, comprising: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line; the at least one processor calls the instructions in the memory so that the device for updating a financial product field executes the above-mentioned method for updating a financial product field.

[0039] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, enable the computer to execute the above-mentioned method for updating financial product fields.

[0040] In an embodiment of the present invention, by training a field priority model for financial products, then using the trained model to predict and update new and existing financial product field priority data, and using the updated financial product field priority data to update existing private equity product data, the effect of dynamically and automatically modifying the fields of financial products according to priority is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of an embodiment of a method for updating a financial product field in an embodiment of the present invention;

[0042] Figure 2 This is a schematic diagram of an embodiment of a device for updating a financial product field according to an embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram of another embodiment of a device for updating a financial product field according to an embodiment of the present invention;

[0044] Figure 4 Schematic diagram of an embodiment of a device for updating a financial product field in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] Embodiments of the present invention provide a method, apparatus, device, and storage medium for updating a financial product field.

[0046] The terms "first," "second," "third," "fourth," and the like (if any) in the description and claims of the present invention and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0047] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 An embodiment of a method for updating a financial product field in an embodiment of the present invention includes:

[0048] 101. Obtain N updated product fields and generate an updated product field set, where N is a positive integer;

[0049] In this embodiment, it is first necessary to extract characteristic data of the data source from each data source, such as update frequency, update field, update source, credibility, etc. Then, characteristic data such as product name, product company, product manager, etc. are extracted from private equity products. The recording method needs to save two fields: field name + update source, such as: {product name: source 1, product company: source 2, product manager: source 3}, where product name, product company, and product manager refer to specific names, and the position is fixed. It should be noted that the updated product field set can also be a specific field such as {product name, product company, product manager}, and then there is a corresponding source record {source 1, source 2, source 3}.

[0050] Furthermore, after step 101 and before step 102, the following steps are executed:

[0051] 1011. Read the data sources of N updated product fields to obtain N field source data;

[0052] 1012. Map the source data of the N fields according to the preset source value mapping table to obtain binary strings corresponding to the source data of the N fields;

[0053] 1013. Set the binary strings corresponding to the N field source data as labels of the updated product fields corresponding to the N field source data to obtain N updated product fields with binary labels.

[0054] In steps 1011-1013, the record of {Product Name: Source 1, Product Company: Source 2, Product Manager: Source 3} has the disadvantage of being too long. When there are many products and product fields, it will cause a large waste of database space. In addition, when processing data, the processing speed will be relatively slow because the string needs to be parsed. The above record is represented in binary. Its length is 64 bits. It is divided into groups of 4 bits, which can be divided into 16 groups in total. Each group represents the update source information of a field, and a total of 16 private equity product fields can be represented. The update source of each field is represented by 4 bits of binary, and a total of 24 = 16 data sources can be represented. In this way, a long integer can be used to represent the update records of 16 private equity product fields + 16 data sources, greatly reducing the space occupied by the database. For example, {source 1, source 2, source 3, source 4, source 5, source 6}, each field uses a 4-bit binary to represent its source, source 1 is represented by binary 0000, source 2 is represented by binary 0001, and so on. Ultimately, six 4-bit binary 0000 0001 0010 0011 0100 0101 are used to represent complex information.

[0055] Furthermore, after step 1013 and before step 102, the following steps may be performed:

[0056] 1014. Convert the binary labels of the N updated product fields into long integers in sequence to obtain N updated product fields with long integer labels.

[0057] A long integer is 64 bits long. The data above only has 24 bits. Padded with 0s to the remaining 40 bits is: 0000 0001 0010 0011 0100 0101 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000. Converting the binary number to a long integer is 81985084525117440. This means that only one long integer is needed to represent the complex field update source information of {Source 1, Source 2, Source 3, Source 4, Source 5, Source 6}. Another benefit of using long integer data for storage is that it facilitates data calculations. For example, to determine the source of the first field in the example above, we can perform bitwise operations on the long integer, such as ANDing: 1111 0000 000000000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000. Then, right-shifting by 64-4 = 60 bits yields the update source of the first field. Because bitwise operations are extremely fast, they can effectively improve processing speed when processing large amounts of data. This split can also be adjusted based on actual business needs. If a 3-bit binary representation of the update source of a field is used, 21 update fields can be represented.

[0058] 102. Performing numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm to generate a character feature data set and a conversion mapping hash table;

[0059] In this embodiment, for the specific field {AQQQDSF, BQFTYD, CQXGJ}, different character combinations will produce different meanings, but for the entire character, a combination of characters and letters is used. For thousands of individual characters, a conversion processing method such as Z0, Z1, Z2, ..., L001, L002, L003 can be generated, which has a character identifier in front and a code in the back. {AQQQDSF, BQFTYD, CQXGJ} can be converted to {Z01Z17Z17Z17Z04Z19Z06, Z02Z17Z06Z20Z25Z04, Z03Z17Z24Z07Z10}. Note that AQQQDSF, BQFTYD, CQXGJ are actually Chinese characters or Chinese and English characters. They are converted into one-to-one correspondence Z01Z17Z17Z17Z04Z19Z06, Z02Z17Z06Z20Z25Z04, Z03Z17Z24Z07Z10 to facilitate computer calculations in the later stage.

[0060] In step 102, the following steps may also be performed:

[0061] 1021. Obtain a training field feature set, wherein each training field feature in the training field feature set includes: a training field and a conversion value;

[0062] 1022. Input the training field feature set into a preset neural network model for recognition training to obtain a recognition transformation matrix set;

[0063] 1023. According to the recognition conversion matrix set, the N updated product fields in the updated product field set are sequentially recognized and converted to obtain a character feature data set and a conversion mapping hash table.

[0064] In steps 1021-1023, the text-CNN neural network model is used to extract fields corresponding to different products from existing private equity product fields, and the above-mentioned feature data is put into the machine learning training engine for training to obtain a matrix set of the conversion model, that is, the model.

[0065] The trained model converts the specific field {AQQQDSF, BQFTYD, CQXGJ} into the corresponding feature data {0214, 3350, 234532}. The conversion performed by the text-CNN neural network model doesn't actually involve converting individual bytes, but rather multiple character combinations. After the combination is complete, a conversion mapping hash table is generated for the information containing 0214-AQQQDSF, 3350-BQFTYD, 234532-CQXGJ. This can be implemented in a variety of ways, including one where the first element of the first table for {AQQQDSF, BQFTYD, CQXGJ}, {0214, 234532, 3350} corresponds to the first element of the second table, the second element of the first table corresponds to the third element of the second table, and the third element of the first table corresponds to the second element of the second table.

[0066] After step 102 and before step 103, the following steps may be performed:

[0067] 1035. Normalize the character feature dataset to obtain a new character feature dataset.

[0068] In this embodiment, the feature data of {0214,,234532,3350} are collected to form a feature data set, and then the feature data set is normalized to facilitate subsequent model prediction.

[0069] 103. Performing priority recognition processing on the character feature data set based on a preset KNN neighbor algorithm to obtain priority numbers corresponding to N character feature data in the character feature data set;

[0070] In this embodiment, the KNN neighbor algorithm performs priority recognition processing on the converted feature data of {Z01Z17Z17Z17Z04Z19Z06, Z02Z17Z06Z20Z25Z04, Z03Z17Z24Z07Z10}. The priority judgment can be the feature character-priority table in the original priority determination table, and the distance between Z01Z17Z17Z17Z04Z19Z06, Z02Z17Z06Z20Z25Z04, Z03Z17Z24Z07Z10 and the feature characters in the priority determination table is determined to obtain the priority of the closest feature character in the priority determination table.

[0071] In step 103, the following steps may also be performed:

[0072] 1031. Setting a priority number label for each training field feature in the training field feature set to obtain a level training field feature set;

[0073] 1032. Input the series training field feature set into a preset neural network model to perform series training processing to obtain a feature series matrix;

[0074] 1033. Based on a preset KNN neighbor algorithm, perform distance determination processing on the N character feature data in the character feature data set and the feature elements in the feature series matrix in sequence to obtain the feature element with the closest distance to the character feature data;

[0075] 1034. According to the correspondence between the feature elements and the series elements in the feature series matrix, determine the series element corresponding to the character feature data as the priority number.

[0076] In steps 1031-1034, a text-CNN neural network model is used to assign training priority labels to the private equity product fields trained in step 102. This feature data is then fed into a machine learning training engine for training, resulting in a feature series matrix, which is the priority number corresponding to the output of each feature field. The feature series matrix has two columns or rows: one column or row is a feature element, and the other column or row is a series element. Feature elements correspond one-to-one to series elements. Using the KNN neighbor algorithm, the characters {0214,,234532,3350} are compared with the eigenvalue columns of the feature series matrix to obtain the closest feature element. The series element corresponding to the feature element is read and used as the priority number.

[0077] 104. According to the conversion mapping hash table, the priority numbers corresponding to the N character feature data are transferred to the update product field corresponding to each character feature data to obtain the priority numbers corresponding to the N update product fields;

[0078] In this embodiment, the priority numbers of {0214,,234532,3350} are 123, 125, and 127 respectively. Based on the correspondence between {AQQQDSF, BQFTYD, CQXGJ} and {0214,234532,3350}, 123, 125, and 127 are passed to {AQQQDSF, BQFTYD, CQXGJ}, and the priority numbers of {AQQQDSF, BQFTYD, CQXGJ} are 123, 125, and 127 respectively.

[0079] 105. Based on the priority numbers corresponding to the N updated product fields, perform a replacement update process on the N financial product usage fields corresponding to the N updated product fields to obtain the updated N financial product usage fields.

[0080] In this example, the product name, product company, and product manager fields in {AQQQDSF, BQFTYD, CQXGJ} have field priorities of 281, 65, and 115. When the priority of the updated product field is 50 units higher than the priority of the financial product usage field, the update is performed. The product company field of BQFTYD is replaced from the original financial product usage field to the updated product field of BQFTYD.

[0081] 1051. Determine whether the priority number of the updated product field is greater than the priority number of the financial product usage field corresponding to the updated product field;

[0082] 1052. If it is greater, replace the financial product usage field corresponding to the updated product field with the updated product field;

[0083] 1053. If not, the financial product usage field corresponding to the updated product field remains unchanged.

[0084] In steps 1051-1053, a circular comparison is performed, where the priority of one updated product field is compared at a time, and the circular comparison is completed until all N updated product fields are compared with the corresponding N financial product usage fields. The comparison rule is strict priority number size, and the product field with a larger priority number is determined to be the used field.

[0085] In an embodiment of the present invention, by training a field priority model for financial products, then using the trained model to predict and update new and existing financial product field priority data, and using the updated financial product field priority data to update existing private equity product data, the effect of dynamically and automatically modifying the fields of financial products according to priority is achieved.

[0086] The above describes the method for updating the financial product field in the embodiment of the present invention. The following describes the device for updating the financial product field in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a device for updating a financial product field includes:

[0087] The acquisition module 201 is used to acquire N updated product fields and generate an updated product field set, where N is a positive integer;

[0088] The conversion module 202 is configured to perform numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm to generate a character feature data set and a conversion mapping hash table;

[0089] The recognition module 203 is configured to perform priority recognition processing on the character feature data set based on a preset KNN neighbor algorithm to obtain priority numbers corresponding to N character feature data in the character feature data set;

[0090] A mapping transfer module 204 is configured to transfer the priority numbers corresponding to the N character feature data to the update product field corresponding to each character feature data according to the conversion mapping hash table, to obtain the priority numbers corresponding to the N update product fields;

[0091] The replacement and update module 205 is configured to perform replacement and update processing on the N financial product usage fields corresponding to the N updated product fields based on the priority numbers corresponding to the N updated product fields, to obtain the updated N financial product usage fields.

[0092] In an embodiment of the present invention, by training a field priority model for financial products, then using the trained model to predict and update new and existing financial product field priority data, and using the updated financial product field priority data to update existing private equity product data, the effect of dynamically and automatically modifying the fields of financial products according to priority is achieved.

[0093] See also Figure 3 Another embodiment of the device for updating the financial product field in the embodiment of the present invention includes:

[0094] The acquisition module 201 is used to acquire N updated product fields and generate an updated product field set, where N is a positive integer;

[0095] The conversion module 202 is configured to perform numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm to generate a character feature data set and a conversion mapping hash table;

[0096] The recognition module 203 is configured to perform priority recognition processing on the character feature data set based on a preset KNN neighbor algorithm to obtain priority numbers corresponding to N character feature data in the character feature data set;

[0097] A mapping transfer module 204 is configured to transfer the priority numbers corresponding to the N character feature data to the update product field corresponding to each character feature data according to the conversion mapping hash table, to obtain the priority numbers corresponding to the N update product fields;

[0098] The replacement and update module 205 is configured to perform replacement and update processing on the N financial product usage fields corresponding to the N updated product fields based on the priority numbers corresponding to the N updated product fields, to obtain the updated N financial product usage fields.

[0099] The apparatus for updating the financial product field further includes a source conversion module 206, which is specifically configured to:

[0100] Read the data sources of N updated product fields and obtain N field source data;

[0101] According to the preset source value mapping table, the source data of N fields are mapped to obtain the binary strings corresponding to the source data of N fields;

[0102] The binary strings corresponding to the N field source data are set as labels of the updated product fields corresponding to the N field source data, thereby obtaining N updated product fields with binary labels.

[0103] The apparatus for updating the financial product field further includes a long integer conversion module 207, which is specifically configured to:

[0104] The binary labels of the N updated product fields are converted to long integers in sequence to obtain N updated product fields with long integer labels.

[0105] The conversion module 202 is specifically configured to:

[0106] Obtaining a training field feature set, wherein each training field feature in the training field feature set includes: a training field and a conversion value;

[0107] Inputting the training field feature set into a preset neural network model for recognition training processing to obtain a recognition transformation matrix set;

[0108] According to the recognition conversion matrix set, the N updated product fields in the updated product field set are sequentially recognized and converted to obtain a character feature data set and a conversion mapping hash table.

[0109] The identification module 203 is specifically configured to:

[0110] Setting a priority number label for each training field feature in the training field feature set to obtain a level training field feature set;

[0111] Inputting the series training field feature set into a preset neural network model for series training processing to obtain a feature series matrix;

[0112] Based on a preset KNN neighbor algorithm, N character feature data in the character feature data set are sequentially subjected to distance determination processing with feature elements in the feature series matrix to obtain the feature element with the closest distance to the character feature data;

[0113] According to the correspondence between the feature elements and the level elements in the feature level matrix, the level element corresponding to the character feature data is determined as the priority number.

[0114] The apparatus for updating the financial product field further includes a normalization module 208, which is specifically configured to:

[0115] Normalizing the character feature dataset to obtain a new character feature dataset.

[0116] The replacement and update module 205 is specifically configured to:

[0117] Determine whether the priority number of the updated product field is greater than the priority number of the financial product usage field corresponding to the updated product field;

[0118] If it is greater, the financial product usage field corresponding to the updated product field is replaced by the updated product field;

[0119] If not, the financial product usage field corresponding to the updated product field will not be changed.

[0120] In an embodiment of the present invention, by training a field priority model for financial products, then using the trained model to predict and update new and existing financial product field priority data, and using the updated financial product field priority data to update existing private equity product data, the effect of dynamically and automatically modifying the fields of financial products according to priority is achieved.

[0121] above Figure 2 and Figure 3 The updating device for the financial product field in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The updating device for the financial product field in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0122] Figure 4FIG2 is a schematic diagram of the structure of a device for updating financial product fields provided in an embodiment of the present invention. The device 400 for updating financial product fields may vary significantly depending on configuration or performance. The device may include one or more central processing units (CPUs) 410 (e.g., one or more processors), memory 420, and one or more storage media 430 (e.g., one or more mass storage devices) storing application programs 433 or data 432. The memory 420 and storage media 430 may be either transient or persistent storage. The program stored in the storage medium 430 may include one or more modules (not shown), each of which may include a series of instructions for operating on the device 400 for updating financial product fields. Furthermore, the processor 410 may be configured to communicate with the storage medium 430, allowing the device 400 to execute the series of instructions stored in the storage medium 430.

[0123] The updating device 400 based on the financial product field may further include one or more power supplies 440, one or more wired or wireless network interfaces 450, one or more input and output interfaces 460, and / or one or more operating systems 431, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 4 The illustrated structure of the updating device for the financial product field does not limit the updating device based on the financial product field and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0124] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, which, when executed on a computer, cause the computer to execute the steps of the method for updating the financial product field.

[0125] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0126] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program code.

[0127] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for updating a financial product field, characterized in that: Methods include: Get N updated product fields and generate an updated product field set, where N is a positive integer; Read the data sources of N updated product fields and obtain N field source data; According to the preset source value mapping table, the source data of N fields are mapped to obtain the binary strings corresponding to the source data of N fields; Set the binary strings corresponding to the N field source data as labels of the updated product fields corresponding to the N field source data to obtain N updated product fields with binary labels; Convert the binary labels of the N updated product fields to long integers in sequence to obtain N updated product fields with long integer labels; According to a preset conversion algorithm, numerical conversion processing is sequentially performed on N updated product fields in the updated product field set to generate a character feature data set and a conversion mapping hash table, including obtaining a training field feature set, wherein each training field feature in the training field feature set includes: a training field and a conversion value; inputting the training field feature set into a preset neural network model for recognition training processing to obtain a recognition conversion matrix set; and according to the recognition conversion matrix set, sequentially performing recognition conversion processing on the N updated product fields in the updated product field set to obtain a character feature data set and a conversion mapping hash table; Based on a preset KNN neighbor algorithm, priority identification processing is performed on the character feature data set to obtain priority numbers corresponding to N character feature data in the character feature data set, including setting a priority number label for each training field feature in the training field feature set to obtain a series training field feature set; the series training field feature set is input into a preset neural network model for series training processing to obtain a feature series matrix; based on the preset KNN neighbor algorithm, distance determination processing is performed on the N character feature data in the character feature data set and the feature elements in the feature series matrix in sequence to obtain the feature element with the closest distance to the character feature data; and according to the correspondence between the feature elements and the series elements in the feature series matrix, the series element corresponding to the character feature data is determined as the priority number; According to the conversion mapping hash table, the priority numbers corresponding to the N character feature data are transferred to the update product field corresponding to each character feature data to obtain the priority numbers corresponding to the N update product fields; Based on the priority numbers corresponding to the N updated product fields, the N financial product usage fields corresponding to the N updated product fields are replaced and updated to obtain the updated N financial product usage fields.

2. The method for updating a financial product field according to claim 1, characterized in that: After sequentially performing numerical conversion processing on the N updated product fields in the updated product field set according to a preset conversion algorithm to generate a character feature data set and a conversion mapping hash table, and before transferring the priority numbers corresponding to the N character feature data to the updated product fields corresponding to each character feature data according to the conversion mapping hash table to obtain the priority numbers corresponding to the N updated product fields, the method further includes: Normalizing the character feature dataset to obtain a new character feature dataset.

3. The method for updating a financial product field according to claim 1, characterized in that: The N financial product usage fields corresponding to the N updated product fields are replaced and updated based on the priority numbers corresponding to the N updated product fields, and the updated N financial product usage fields include: Determine whether the priority number of the updated product field is greater than the priority number of the financial product usage field corresponding to the updated product field; If it is greater, the financial product usage field corresponding to the updated product field is replaced by the updated product field; If not, the financial product usage field corresponding to the updated product field will not be changed.

4. A device for updating a financial product field, characterized in that: The device comprises: An acquisition module is used to acquire N updated product fields and generate an updated product field set, where N is a positive integer; A conversion module, configured to perform numerical conversion processing on the N updated product fields in the updated product field set in sequence according to a preset conversion algorithm, and generate a character feature data set and a conversion mapping hash table; an identification module, configured to perform priority identification processing on the character feature data set based on a preset KNN neighbor algorithm to obtain priority numbers corresponding to N character feature data in the character feature data set; A mapping transfer module, configured to transfer the priority numbers corresponding to the N character feature data to the update product field corresponding to each character feature data according to the conversion mapping hash table, to obtain the priority numbers corresponding to the N update product fields; A replacement and update module, configured to perform a replacement and update process on N financial product usage fields corresponding to the N update product fields based on the priority numbers corresponding to the N update product fields, to obtain the updated N financial product usage fields; The device further includes a source conversion module, which is specifically configured to: Read the data sources of N updated product fields and obtain N field source data; According to the preset source value mapping table, the source data of N fields are mapped to obtain the binary strings corresponding to the source data of N fields; Set the binary strings corresponding to the N field source data as labels of the updated product fields corresponding to the N field source data to obtain N updated product fields with binary labels; The device further includes a long integer conversion module, which is specifically configured to: Convert the binary labels of the N updated product fields to long integers in sequence to obtain N updated product fields with long integer labels; The conversion module is specifically configured to: Obtaining a training field feature set, wherein each training field feature in the training field feature set includes: a training field and a conversion value; Inputting the training field feature set into a preset neural network model for recognition training processing to obtain a recognition transformation matrix set; According to the recognition conversion matrix set, the N updated product fields in the updated product field set are sequentially recognized and converted to obtain a character feature data set and a conversion mapping hash table; The identification module is specifically used for: Setting a priority number label for each training field feature in the training field feature set to obtain a level training field feature set; Inputting the series training field feature set into a preset neural network model for series training processing to obtain a feature series matrix; Based on a preset KNN neighbor algorithm, N character feature data in the character feature data set are sequentially subjected to distance determination processing with feature elements in the feature series matrix to obtain the feature element with the closest distance to the character feature data; According to the correspondence between the feature elements and the level elements in the feature level matrix, the level element corresponding to the character feature data is determined as the priority number.

5. A device for updating a financial product field, characterized in that: The device for updating the financial product field includes: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line; The at least one processor calls the instructions in the memory to enable the financial product field updating device to execute the financial product field updating method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for updating the financial product field according to any one of claims 1 to 3 is implemented.

Images