Text verification method and device, computer equipment, storage medium and computer program product

Abstract

The invention relates to a text verification method and device, computer equipment, a storage medium and a computer program product. The method comprises the steps of performing segmentation processing on a to-be-analyzed financial text to obtain a first financial text and a second financial text corresponding to the to-be-analyzed financial text; determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text; screening out a first target derivative financial text meeting a first preset condition from the first derivative financial text, and screening out a second target derivative financial text meeting a second preset condition from the second derivative financial text; combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the to-be-analyzed financial text; and determining a target verification result of the to-be-analyzed financial text according to the difference between the to-be-analyzed financial text and the target financial text. By adopting the method, the accuracy of text verification can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a text verification method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] In the insurance field, it is crucial to verify the text in order to ensure its accuracy and consistency.

[0003] In traditional technology, manual verification is generally used in the process of text verification; however, this manual verification method has subjective factors and is prone to errors, resulting in a low accuracy rate of text verification. Summary of the invention

[0004] Based on this, it is necessary to provide a text verification method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of text verification in response to the above technical problems.

[0005] In a first aspect, the present application provides a text verification method, comprising:

[0006] Obtain the financial text to be analyzed;

[0007] Segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string;

[0008] Determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text;

[0009] Screening out a first target derivative financial text that meets a first preset condition from the first derivative financial text, and screening out a second target derivative financial text that meets a second preset condition from the second derivative financial text;

[0010] Combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed;

[0011] According to the difference between the financial text to be analyzed and the target financial text, a target verification result of the financial text to be analyzed is determined.

[0012] In one embodiment, the step of selecting the first target derivative financial text that meets the first preset condition from the first derivative financial text includes:

[0013] respectively determining the difference between each of the first derivative financial texts and the first financial text;

[0014] A first derivative financial text having the smallest difference value with the first financial text is selected from the first derivative financial texts as the first target derivative financial text satisfying the first preset condition.

[0015] In one embodiment, the step of selecting the second target derivative financial text that meets the second preset condition from the second derivative financial text includes:

[0016] Performing feature extraction processing on each of the second derivative financial texts to obtain a first feature vector corresponding to each of the second derivative financial texts, and performing feature extraction processing on the second financial text to obtain a second feature vector corresponding to the second financial text;

[0017] Determining the distance between each of the second derivative financial texts and the second financial text according to the first feature vector corresponding to each of the second derivative financial texts and the second feature vector corresponding to the second financial text;

[0018] A second derivative financial text having the smallest distance from the second financial text is selected from each of the second derivative financial texts as the second target derivative financial text satisfying the second preset condition.

[0019] In one embodiment, determining a target verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text includes:

[0020] Extracting target text information from the financial text to be analyzed;

[0021] Determining a first verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text, and performing verification processing on the target text information to obtain a second verification result of the financial text to be analyzed;

[0022] A target verification result of the financial text to be analyzed is determined according to the first verification result and the second verification result.

[0023] In one embodiment, the segmenting of the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed includes:

[0024] Segmenting the financial text to be analyzed to obtain a target text structure, a first sub-financial text, and a second sub-financial text corresponding to the financial text to be analyzed; the first sub-financial text is used to represent a sub-financial text carrying flag information, and the second sub-financial text is used to represent a sub-financial text not carrying flag information;

[0025] The target text structure is used as the financial text to be analyzed, and the process is skipped to the step of segmenting the financial text to be analyzed until the segmentation result obtained does not contain the target text structure;

[0026] Each of the first sub-financial texts obtained through the segmentation process is used as the first financial text, and each of the second sub-financial texts obtained through the segmentation process is used as the second financial text.

[0027] In one embodiment, combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed includes:

[0028] Determining a target combination method corresponding to the first financial text and the second financial text;

[0029] According to the target combination method, the first target derivative financial text and the second target derivative financial text are combined to obtain a target financial text corresponding to the financial text to be analyzed.

[0030] In a second aspect, the present application also provides a text verification device, comprising:

[0031] A text acquisition module is used to acquire financial texts to be analyzed;

[0032] a text segmentation module, configured to segment the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string;

[0033] a text determination module, configured to determine a first derivative financial text corresponding to the first financial text, and to determine a second derivative financial text corresponding to the second financial text;

[0034] a text screening module, configured to screen out first target derivative financial texts satisfying a first preset condition from the first derivative financial texts, and to screen out second target derivative financial texts satisfying a second preset condition from the second derivative financial texts;

[0035] A text combination module, used for combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed;

[0036] The result determination module is used to determine a target verification result of the financial text to be analyzed according to the difference between the financial text to be analyzed and the target financial text.

[0037] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0038] Obtain the financial text to be analyzed;

[0039] Segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string;

[0040] Determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text;

[0041] Screening out a first target derivative financial text that meets a first preset condition from the first derivative financial text, and screening out a second target derivative financial text that meets a second preset condition from the second derivative financial text;

[0042] Combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed;

[0043] According to the difference between the financial text to be analyzed and the target financial text, a target verification result of the financial text to be analyzed is determined.

[0044] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0045] Obtain the financial text to be analyzed;

[0046] Segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string;

[0047] Determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text;

[0048] Screening out a first target derivative financial text that meets a first preset condition from the first derivative financial text, and screening out a second target derivative financial text that meets a second preset condition from the second derivative financial text;

[0049] Combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed;

[0050] According to the difference between the financial text to be analyzed and the target financial text, a target verification result of the financial text to be analyzed is determined.

[0051] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0052] Obtain the financial text to be analyzed;

[0053] Segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string;

[0054] Determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text;

[0055] Screening out a first target derivative financial text that meets a first preset condition from the first derivative financial text, and screening out a second target derivative financial text that meets a second preset condition from the second derivative financial text;

[0056] Combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed;

[0057] According to the difference between the financial text to be analyzed and the target financial text, a target verification result of the financial text to be analyzed is determined.

[0058] The above-mentioned text verification method, device, computer equipment, storage medium and computer program product first obtain a financial text to be analyzed, and segment the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed, then determine a first derivative financial text corresponding to the first financial text, and determine a second derivative financial text corresponding to the second financial text, then screen out a first target derivative financial text that meets a first preset condition from the first derivative financial text, and screen out a second target derivative financial text that meets a second preset condition from the second derivative financial text, then combine the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed, and finally determine a target verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text. In this way, in the process of text verification, the financial text to be analyzed is subjected to a series of processing such as segmentation, combination and screening, so that a more accurate target financial text can be obtained, and then a more accurate target verification result can be determined according to the difference between the financial text to be analyzed and the target financial text, which is conducive to improving the accuracy of text verification; moreover, the whole process does not require human intervention, avoiding the defect of subjective factors and easy errors in manual verification, which leads to low accuracy of text verification, thereby improving the accuracy of text verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 A flowchart of a text verification method in one embodiment;

[0061] Figure 2 A flowchart of a text verification method in another embodiment;

[0062] Figure 3 A schematic diagram of a flow chart of a weak structure text proofreading method based on a rule engine in one embodiment;

[0063] Figure 4 is a structural block diagram of a text verification device in an embodiment;

[0064] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0067] In an exemplary embodiment, Figure 1 As shown, a text verification method is provided. This embodiment uses the method applied to a server as an example for illustration. It can be understood that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be but is not limited to various personal computers, laptops, smart phones and tablet computers; the server can be implemented as an independent server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0068] Step S101, obtaining financial text to be analyzed.

[0069] The financial text to be analyzed refers to the financial text that needs to be verified.

[0070] Exemplarily, the server obtains multiple candidate financial texts from a database; then, the server determines the complexity corresponding to each candidate financial text, and selects candidate financial texts whose corresponding complexity is greater than a preset complexity from the multiple candidate financial texts as the financial texts to be analyzed.

[0071] Step S102, segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent the financial text containing the target string, and the second financial text is used to represent the financial text not containing the target string.

[0072] The first financial text is used to represent the financial text containing the target character string. In actual scenarios, the first financial text refers to the text with a logo.

[0073] The target string is a special string used to determine whether the financial text belongs to a certain category (such as a disclaimer).

[0074] The second financial text is used to represent the financial text that does not contain the target character string. In actual scenarios, the second financial text refers to the text without a logo.

[0075] Exemplarily, the server segments the financial text to be analyzed to obtain financial text containing a target string and financial text not containing the target string; then, the server uses the financial text containing the target string as the first financial text corresponding to the financial text to be analyzed, and uses the financial text not containing the target string as the second financial text corresponding to the financial text to be analyzed.

[0076] Step S103: determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text.

[0077] Among them, the first derivative financial text refers to the multiple derivative forms corresponding to the first financial text, also known as multiple replaceable contents corresponding to the first financial text.

[0078] Among them, the second derivative financial text refers to a plurality of derivative forms corresponding to the second financial text, also known as a plurality of replaceable contents corresponding to the second financial text.

[0079] It should be noted that both the first derivative financial text and the second derivative financial text are specific texts. It should be noted that the various forms of specific texts are replaceable contents of specific texts. For example, specific text A has three forms: a1, a2, and a3. Then the specific text A here can accept a1, a2, or a3.

[0080] Exemplarily, the server determines multiple derivative forms corresponding to the first financial text as first derivative financial texts corresponding to the first financial text, and determines multiple derivative forms corresponding to the second financial text as second derivative financial texts corresponding to the second financial text.

[0081] Step S104 , selecting first target derivative financial texts that meet a first preset condition from the first derivative financial texts, and selecting second target derivative financial texts that meet a second preset condition from the second derivative financial texts.

[0082] The first preset condition refers to a preset condition for screening the first derivative financial text. It should be noted that the first preset condition depends on the circumstances.

[0083] The first target derivative financial text refers to the first derivative financial text that meets the first preset condition.

[0084] The second preset condition refers to a preset condition for screening the second derivative financial text. It should be noted that the second preset condition depends on the circumstances.

[0085] The second target derivative financial text refers to a second derivative financial text that meets the second preset condition.

[0086] Exemplarily, the server selects first derivative financial texts that meet a first preset condition from the first derivative financial texts, and uses these first derivative financial texts as first target derivative financial texts, and selects second derivative financial texts that meet a second preset condition from the second derivative financial texts, and uses these second derivative financial texts as second target derivative financial texts.

[0087] Step S105 : combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed.

[0088] The target financial text refers to a financial text obtained by combining the first target derivative financial text and the second target derivative financial text.

[0089] Exemplarily, the server combines the first target derivative financial text and the second target derivative financial text to obtain a combined financial text; then, the server uses the combined financial text as the target financial text corresponding to the financial text to be analyzed.

[0090] Step S106: determining a target verification result of the financial text to be analyzed according to the difference between the financial text to be analyzed and the target financial text.

[0091] The target verification result is used to indicate the verification status of the financial text to be analyzed.

[0092] Exemplarily, the server determines abnormal text in the financial text to be analyzed (such as a difference block between the financial text to be analyzed and the target financial text) according to the difference between the financial text to be analyzed and the target financial text as the target verification result of the financial text to be analyzed.

[0093] In the above text verification method, the financial text to be analyzed is first obtained, and the financial text to be analyzed is segmented to obtain the first financial text and the second financial text corresponding to the financial text to be analyzed, and then the first derivative financial text corresponding to the first financial text is determined, and the second derivative financial text corresponding to the second financial text is determined, then the first target derivative financial text that meets the first preset condition is screened out from the first derivative financial text, and the second target derivative financial text that meets the second preset condition is screened out from the second derivative financial text, and then the first target derivative financial text and the second target derivative financial text are combined to obtain the target financial text corresponding to the financial text to be analyzed, and finally, according to the difference between the financial text to be analyzed and the target financial text, the target verification result of the financial text to be analyzed is determined. In this way, in the process of text verification, the financial text to be analyzed is subjected to a series of processing such as segmentation, combination and screening, so that a more accurate target financial text can be obtained, and then a more accurate target verification result can be determined according to the difference between the financial text to be analyzed and the target financial text, which is conducive to improving the accuracy of text verification; moreover, the whole process does not require human intervention, avoiding the defect of subjective factors and easy errors in manual verification, which leads to low accuracy of text verification, thereby improving the accuracy of text verification.

[0094] In an exemplary embodiment, the above step S104, screening out the first target derivative financial text that meets the first preset condition from the first derivative financial texts, specifically includes the following contents: respectively determining the difference value between each first derivative financial text and the first financial text; and screening out the first derivative financial text with the smallest difference value between the first financial text and the first derivative financial text from each first derivative financial text as the first target derivative financial text that meets the first preset condition.

[0095] The difference value is used to indicate the difference between each first derivative financial text and the first financial text.

[0096] Exemplarily, the server preprocesses the first derivative financial text and the first financial text to obtain the preprocessed first derivative financial text and the preprocessed first financial text; for example, the server converts the parameter reserved characters of the first derivative financial text and the first financial text into the character '*' to obtain the preprocessed first derivative financial text and the preprocessed first financial text; then, the server respectively determines the redundant characters, missing characters, and erroneous characters between each preprocessed first derivative financial text and the preprocessed first financial text, all of which are used as the difference value between each preprocessed first derivative financial text and the preprocessed first financial text; then, the server selects the preprocessed first derivative financial text with the smallest difference value with the preprocessed first financial text from each preprocessed first derivative financial text, and uses the preprocessed first derivative financial text as the first target derivative financial text that meets the first preset condition.

[0097] For example, the server can determine the difference value by the following formula:

[0098] Difference value = extra characters + missing characters + wrong characters * 2, formula (1)

[0099] In this embodiment, by determining the difference value between each first derivative financial text and the first financial text, and selecting the one with the smallest difference value as the first target derivative financial text, those derivative contents that deviate greatly from the original first financial text can be effectively eliminated, thereby improving the accuracy and reliability of subsequent analysis based on the first target derivative financial text.

[0100] In an exemplary embodiment, the above-mentioned step S104, screening out second target derivative financial texts that meet the second preset condition from the second derivative financial texts, specifically includes the following contents: performing feature extraction processing on each second derivative financial text to obtain a first feature vector corresponding to each second derivative financial text, and performing feature extraction processing on the second financial text to obtain a second feature vector corresponding to the second financial text; determining the distance between each second derivative financial text and the second financial text according to the first feature vector corresponding to each second derivative financial text and the second feature vector corresponding to the second financial text; and screening out, from each second derivative financial text, the second derivative financial text having the smallest distance to the second financial text as the second target derivative financial text that meets the second preset condition.

[0101] The first feature vector refers to the feature vector corresponding to each second derivative financial text.

[0102] The second feature vector refers to the feature vector corresponding to the second financial text.

[0103] The distance is used to indicate the difference between each second derivative financial text and the second financial text. It should be noted that the distance may refer to the edit distance.

[0104] Exemplarily, the server inputs each second derivative financial text into a feature extraction model, performs feature extraction processing on each second derivative financial text through the feature extraction model, and obtains a first feature vector corresponding to each second derivative financial text; then, the server inputs the second financial text into the feature extraction model, performs feature extraction processing on the second financial text through the feature extraction model, and obtains a second feature vector corresponding to the second financial text; then, the server determines the difference between each second derivative financial text and the second financial text according to the first feature vector corresponding to each second derivative financial text and the second feature vector corresponding to the second financial text, as the distance between each second derivative financial text and the second financial text; then, the server selects the second derivative financial text with the smallest distance to the second financial text from each second derivative financial text, and uses the second derivative financial text as the second target derivative financial text that meets the second preset condition.

[0105] In this embodiment, the second target derivative financial text is screened out through feature extraction and distance calculation, so that the text can be processed efficiently. Compared with the traditional method of reading and comparing the text content one by one, the screening and evaluation of a large amount of text can be completed in a shorter time, which greatly improves the overall processing efficiency of financial text processing and saves time and labor costs.

[0106] In an exemplary embodiment, the above step S106, determining a target verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text, specifically includes the following contents: extracting target text information in the financial text to be analyzed; determining a first verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text, and performing verification processing on the target text information to obtain a second verification result of the financial text to be analyzed; determining a target verification result of the financial text to be analyzed based on the first verification result and the second verification result.

[0107] The target text information refers to the parameter information (such as date) in the financial text to be analyzed.

[0108] The first verification result refers to a verification result of the financial text to be analyzed obtained according to the difference between the financial text to be analyzed and the target financial text.

[0109] The second verification result refers to the verification result of the financial text to be analyzed obtained by performing verification processing on the target text information.

[0110] Exemplarily, the server extracts parameter information from the financial text to be analyzed as target text information; then, the server determines abnormal text in the financial text to be analyzed (such as a difference block between the financial text to be analyzed and the target financial text) based on the difference between the financial text to be analyzed and the target financial text as a first verification result of the financial text to be analyzed; then, the server performs verification processing on the target text information according to preset verification information to obtain a second verification result of the financial text to be analyzed; then, the server combines the first verification result and the second verification result to obtain a target verification result of the financial text to be analyzed.

[0111] Furthermore, when it is identified that parameter information has not been extracted, the server generates an alarm message and sends the alarm message to a terminal associated with the financial text to be analyzed.

[0112] In this embodiment, by performing verification processing on the target text information based on the difference between the financial text to be analyzed and the target financial text, a target verification result of the financial text to be analyzed is comprehensively obtained, thereby making the target verification result of the financial text to be analyzed more comprehensive and accurate, which is conducive to improving the determination accuracy of the target verification result of the financial text to be analyzed.

[0113] In an exemplary embodiment, the above step S102, segmenting the financial text to be analyzed to obtain the first financial text and the second financial text corresponding to the financial text to be analyzed, specifically includes the following contents: segmenting the financial text to be analyzed to obtain the target text structure, the first sub-financial text and the second sub-financial text corresponding to the financial text to be analyzed; the first sub-financial text is used to represent the sub-financial text carrying the flag information, and the second sub-financial text is used to represent the sub-financial text not carrying the flag information; taking the target text structure as the financial text to be analyzed, and jumping to the step of segmenting the financial text to be analyzed until the segmentation result does not contain the target text structure; taking each first sub-financial text obtained by the segmentation as the first financial text, and taking each second sub-financial text obtained by the segmentation as the second financial text.

[0114] The target text structure refers to a financial text structure that contains certain specific characters and can be used as identification marks, also known as a marked structure.

[0115] The first sub-financial text is used to represent the sub-financial text carrying the flag information.

[0116] It should be noted that the target text structure is similar to the first sub-financial text in form, but the target text structure presents structural information, while the first sub-financial text presents unstructured information.

[0117] Herein, the mark information refers to the identification mark.

[0118] Among them, the sub-financial text refers to a part of the financial text to be analyzed.

[0119] Among them, the second sub-financial text is used to represent the sub-financial text without flag information.

[0120] Among them, the segmentation processing result is used to represent the processing situation corresponding to the segmentation processing of the financial text to be analyzed.

[0121] Exemplarily, the server performs segmentation processing on the financial text to be analyzed according to the formatting segmentation symbols of the financial text (such as "1.", "1.", "(1)"), and obtains the target text structure corresponding to the financial text to be analyzed, the sub-financial text carrying flag information, and the sub-financial text without flag information; then, the server uses the sub-financial text carrying flag information as the first sub-financial text, and the sub-financial text without flag information as the second sub-financial text; then, the server uses the target text structure as the financial text to be analyzed, and jumps to the step of performing segmentation processing on the financial text to be analyzed until the obtained segmentation processing result does not contain the target text structure; then, the server uses each first sub-financial text obtained by the segmentation processing as the first financial text, and each second sub-financial text obtained by the segmentation processing as the second financial text.

[0122] In this embodiment, by repeatedly performing segmentation processing on the financial text to be analyzed, when there is still a target text structure in the financial text to be analyzed after the segmentation processing, the target text structure is segmented again, so that the complex financial text can be gradually disassembled into sub-financial texts at different levels, and then the target content with different characteristics can be accurately extracted from the financial text to be analyzed, which is convenient for the subsequent text processing process.

[0123] In an exemplary embodiment, in step S105 above, the first target derivative financial text and the second target derivative financial text are combined to obtain the target financial text corresponding to the financial text to be analyzed, which specifically includes the following content: determining the target combination method corresponding to the first financial text and the second financial text; combining the first target derivative financial text and the second target derivative financial text according to the target combination method to obtain the target financial text corresponding to the financial text to be analyzed.

[0124] Among them, the target combination method refers to the combination method corresponding to the first financial text and the second financial text.

[0125] Exemplarily, the server determines a target combination method corresponding to the first financial text and the second financial text based on the formatting segmentation symbol of the financial text; then, the server combines the first target derivative financial text and the second target derivative financial text according to the target combination method to obtain a combined financial text; then, the server uses the combined financial text as the target financial text corresponding to the financial text to be analyzed.

[0126] In this embodiment, by determining the target combination method corresponding to the first financial text and the second financial text, the first target derived financial text and the second target derived financial text can be combined according to specific rules or logic, which is beneficial to improving the accuracy of determining the target financial text corresponding to the financial text to be analyzed.

[0127] In an exemplary embodiment, Figure 2 As shown, another text verification method is provided, which is described by taking the method applied to a server as an example, and includes the following steps:

[0128] Step S201, obtaining financial text to be analyzed.

[0129] Step S202, segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent the financial text containing the target string, and the second financial text is used to represent the financial text not containing the target string.

[0130] Step S203: determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text.

[0131] Step S204: respectively determine the difference between each first derivative financial text and the first financial text.

[0132] Step S205 , selecting the first derivative financial text having the smallest difference with the first financial text from the first derivative financial texts as the first target derivative financial text satisfying the first preset condition.

[0133] Step S206 , performing feature extraction processing on each second derivative financial text to obtain a first feature vector corresponding to each second derivative financial text, and performing feature extraction processing on the second financial text to obtain a second feature vector corresponding to the second financial text.

[0134] Step S207: determining the distance between each second derivative financial text and the second financial text according to the first feature vector corresponding to each second derivative financial text and the second feature vector corresponding to the second financial text.

[0135] Step S208 : selecting, from the second derivative financial texts, the second derivative financial text having the smallest distance to the second financial text as the second target derivative financial text satisfying the second preset condition.

[0136] Step S209, determining a target combination method corresponding to the first financial text and the second financial text; combining the first target derived financial text and the second target derived financial text according to the target combination method to obtain a target financial text corresponding to the financial text to be analyzed.

[0137] Step S210, extracting target text information from the financial text to be analyzed; determining a first verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text, and performing verification processing on the target text information to obtain a second verification result of the financial text to be analyzed.

[0138] Step S211, determining a target verification result of the financial text to be analyzed according to the first verification result and the second verification result.

[0139] In the above-mentioned text verification method, during the text verification process, the financial text to be analyzed is subjected to a series of processing such as segmentation, combination and screening, so that a more accurate target financial text can be obtained, and then a more accurate target verification result can be determined based on the difference between the financial text to be analyzed and the target financial text, which is conducive to improving the accuracy of text verification; moreover, the entire process does not require human intervention, avoiding the defect of subjective factors and prone to errors in manual verification, resulting in a low accuracy rate of text verification, thereby improving the accuracy of text verification.

[0140] In an exemplary embodiment, in order to more clearly illustrate the text verification method provided by the embodiment of the present application, the text verification method is specifically described below with a specific embodiment. Figure 3 As shown, the present application also provides a weak structure text proofreading method based on a rule engine. In the process of text proofreading, the specific text, multiple forms of the specific text, and the rules of the text are first configured according to the needs, and the input text is structurally split and classified to obtain marked text and unmarked text, and the optimal form of the marked text is screened and the optimal form of the unmarked text is matched to obtain the text form closest to the text, and then the text form closest to the text is verified according to the configured rules to obtain the text verification result. Specifically including the following contents:

[0141] 1. Rule Engine:

[0142] (1) Managers configure specific texts, multiple forms of specific texts, and text rules based on their needs.

[0143] Among them, specific text refers to the text content for a specific purpose, such as a disclaimer. Specific text is divided into "text with a marker" and "text without a marker". The difference between them lies in whether there is a special string that can directly determine it as a specific text with a marker.

[0144] Among them, the various forms of specific text are the replaceable content of specific text. For example, if specific text A has three forms: a1, a2, and a3, then specific text A here can accept a1, can also accept a2, and can also accept a3.

[0145] Among them, an example of text rule configuration: the text "{#date} The weather is really nice", and the rule "{#date} > {!set date}" ({!set date} should be before {#date}). "{#xxx}" is the reserved position for parameters, and "{!xxx}" can be a parameter from other text rules or databases and can be defined by oneself.

[0146] Among them, the rules include but are not limited to numerical calculation and comparison, text consistency comparison, etc.

[0147] It should be noted that there can also be an if condition (such as a preset text length, preset keywords, etc.) before each rule, and the corresponding rule verification is performed only when the if condition is met.

[0148] (2) Managers can declare a marker structure and state what specific text / structure must be present / can be present under a certain marker structure.

[0149] Among them, a marker structure refers to a text structure that contains certain specific words and can be used as an identification marker, such as "Adverse reaction: xxxxx". A marker structure can also be specific text, and vice versa. The specific situation can be defined by the manager.

[0150] 2. Structure splitting and classification:

[0151] According to the formatting delimiters of the text (such as "I.", "1.", "(1)"), the text is split, and according to the rules and definitions in the rule engine, the original structure is divided into "marker structure", "text with a marker", and "text without a marker". If the input text itself is defined as a "marker structure" text, it will also be judged whether there is additional "text with a marker". Among them, the "marker structure" can enter step 2 again for further splitting and classification (the original complete text can also be regarded as a special case of the "marker structure").

[0152] Among them, a text with a marker is a text that contains certain specific words and can be used as an identification marker.

[0153] 3. The optimal form of the marked text is screened as follows:

[0154] (1) Convert the reserved words in the text form to the character '*'.

[0155] (2) Calculate the longest common string for each form of the corresponding mark text, and the difference value is calculated as follows: Difference value = extra characters + missing characters + wrong characters * 2.

[0156] (3) Select the shape with the smallest difference value as the optimal shape.

[0157] 4. Optimal morphological matching of unmarked text. The process is as follows:

[0158] (1) In the same manner as the first and second steps in module 3, calculate the difference between all the remaining unmarked texts x1, x2, x3, etc. and all the forms a1, a2, a3, b1, b2, b3, c1, c2, c3 of all the preset unmarked texts A, B, C. It should be noted that all the forms corresponding to the unmarked text A are a1, a2, a3, all the forms corresponding to the unmarked text B are b1, b2, b3, and all the forms corresponding to the unmarked text C are c1, c2, c3.

[0159] (2) The value of f(x,Y) is obtained by taking the smallest difference value among all forms, where x represents any unmarked text and Y represents any preset unmarked text. If x1 is closest to c2 among the forms c1, c2, and c3 of C, and the difference value is v, then f(x1,C)=v.

[0160] (3) Use the KM algorithm to match the optimal unmarked text with the optimal match of the preset unmarked text. The form y with the smallest difference value for x is used as the optimal form match for x. It should be noted that the purpose of using the KM algorithm is to ensure that all x are matched to Y as much as possible, so that the sum of the difference values ​​corresponding to all x is minimized. When the sum of the difference values ​​corresponding to all x is minimized, all x can find the corresponding form in reverse. Assuming that the final matching result includes the match between x1 and C, it can be known that c2 is the optimal result corresponding to x1 in C.

[0161] 5. Parameter extraction and parameter verification:

[0162] The text of each sub-part is processed according to the closest text form and rules obtained in steps 3 and 4. The process is as follows:

[0163] (1) Convert the reserved words in the text form to the character '*'.

[0164] (2) Calculate the longest common substring and divide the difference blocks (based on the characteristics of the longest common substring, we can know where in text B the extra words in text A are located).

[0165] (3) Extract the parameters that can be extracted, and issue an alarm for the parameters that cannot be extracted.

[0166] (4) Verify the parameters according to the rules in the text form. If any parameter cannot be extracted successfully or does not meet the rules, an alarm is issued. It should be noted that in actual situations, when the number of parameters in the form does not match the parameters that can be extracted from the text, the verification of this small part may be terminated (no further rule verification will be performed, but it will not affect the verification of other parts).

[0167] 6. Summary module:

[0168] All parameters, rule-related alarms, and text differences (except those caused by parameter extraction) in the previous steps are summarized and displayed, and missing text and text that should not appear are prompted.

[0169] In the above embodiment, during the text verification process, the financial text to be analyzed is segmented, combined, and screened, so that a more accurate target financial text can be obtained, and then a more accurate target verification result can be determined based on the difference between the financial text to be analyzed and the target financial text, which is conducive to improving the accuracy of text verification; moreover, the entire process does not require manual intervention, avoiding the defects of subjective factors and easy errors in manual verification, resulting in low accuracy of text verification, thereby improving the accuracy of text verification. At the same time, this solution does not need to collect data and train neural networks, and at the same time avoids the problem that the neural network cannot stably output correct results 100%, and combines with the rule engine to flexibly call rules, and can set internal parameters and combine with the database to perform more complex verification, and will not cause verification to be terminated due to some minor errors.

[0170] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0171] Based on the same inventive concept, the embodiment of the present application also provides a text verification device for implementing the above-mentioned text verification method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more text verification device embodiments provided below can refer to the limitations of the text verification method above, and will not be repeated here.

[0172] In an exemplary embodiment, Figure 4 As shown, a text verification device is provided, including: a text acquisition module 401, a text segmentation module 402, a text determination module 403, a text screening module 404, a text combination module 405 and a result determination module 406, wherein:

[0173] The text acquisition module 401 is used to acquire the financial text to be analyzed.

[0174] The text segmentation module 402 is used to segment the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent the financial text containing the target string, and the second financial text is used to represent the financial text not containing the target string.

[0175] The text determination module 403 is used to determine a first derivative financial text corresponding to the first financial text, and to determine a second derivative financial text corresponding to the second financial text.

[0176] The text screening module 404 is used to screen out first target derivative financial texts that meet a first preset condition from the first derivative financial texts, and to screen out second target derivative financial texts that meet a second preset condition from the second derivative financial texts.

[0177] The text combination module 405 is used to combine the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed.

[0178] The result determination module 406 is used to determine a target verification result of the financial text to be analyzed according to the difference between the financial text to be analyzed and the target financial text.

[0179] In an exemplary embodiment, the text screening module 404 is further used to respectively determine the difference values ​​between each first derivative financial text and the first financial text; and screen out the first derivative financial text having the smallest difference value with the first financial text from each first derivative financial text as the first target derivative financial text that meets the first preset condition.

[0180] In an exemplary embodiment, the text screening module 404 is further used to perform feature extraction processing on each second derivative financial text to obtain a first feature vector corresponding to each second derivative financial text, and to perform feature extraction processing on the second financial text to obtain a second feature vector corresponding to the second financial text; determine the distance between each second derivative financial text and the second financial text according to the first feature vector corresponding to each second derivative financial text and the second feature vector corresponding to the second financial text; and screen out the second derivative financial text with the smallest distance to the second financial text from each second derivative financial text as the second target derivative financial text that meets the second preset condition.

[0181] In an exemplary embodiment, the result determination module 406 is also used to extract target text information from the financial text to be analyzed; determine a first verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text, and perform verification processing on the target text information to obtain a second verification result of the financial text to be analyzed; and determine a target verification result of the financial text to be analyzed based on the first verification result and the second verification result.

[0182] In an exemplary embodiment, the text segmentation module 402 is further used to perform segmentation processing on the financial text to be analyzed, and obtain a target text structure, a first sub-financial text, and a second sub-financial text corresponding to the financial text to be analyzed; the first sub-financial text is used to represent a sub-financial text carrying flag information, and the second sub-financial text is used to represent a sub-financial text not carrying flag information; the target text structure is used as the financial text to be analyzed, and the step of performing segmentation processing on the financial text to be analyzed is jumped to until the segmentation processing result obtained does not contain the target text structure; each first sub-financial text obtained by the segmentation processing is used as the first financial text, and each second sub-financial text obtained by the segmentation processing is used as the second financial text.

[0183] In an exemplary embodiment, the text combination module 405 is also used to determine a target combination method corresponding to the first financial text and the second financial text; according to the target combination method, the first target derivative financial text and the second target derivative financial text are combined to obtain a target financial text corresponding to the financial text to be analyzed.

[0184] Each module in the above-mentioned text verification device can be implemented in whole or in part by software, hardware and a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0185] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as a first financial text and a second financial text. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a text verification method is implemented.

[0186] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0187] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0188] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0189] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0190] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0191] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0192] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A text verification method, characterized in that: The method comprises: Obtain the financial text to be analyzed; Segmenting the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string; Determining a first derivative financial text corresponding to the first financial text, and determining a second derivative financial text corresponding to the second financial text; Screening out a first target derivative financial text that meets a first preset condition from the first derivative financial text, and screening out a second target derivative financial text that meets a second preset condition from the second derivative financial text; Combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed; According to the difference between the financial text to be analyzed and the target financial text, a target verification result of the financial text to be analyzed is determined.

2. The method according to claim 1, characterized in that The step of selecting a first target derivative financial text satisfying a first preset condition from the first derivative financial text includes: respectively determining the difference between each of the first derivative financial texts and the first financial text; A first derivative financial text having the smallest difference value with the first financial text is selected from each of the first derivative financial texts as the first target derivative financial text satisfying the first preset condition.

3. The method according to claim 1, characterized in that The step of selecting the second target derivative financial text that meets the second preset condition from the second derivative financial text includes: Performing feature extraction processing on each of the second derivative financial texts to obtain a first feature vector corresponding to each of the second derivative financial texts, and performing feature extraction processing on the second financial text to obtain a second feature vector corresponding to the second financial text; determining the distance between each of the second derivative financial texts and the second financial text according to the first feature vector corresponding to each of the second derivative financial texts and the second feature vector corresponding to the second financial text; A second derivative financial text having the smallest distance from the second financial text is selected from each of the second derivative financial texts as the second target derivative financial text satisfying the second preset condition.

4. The method according to claim 1, characterized in that: The step of determining a target verification result of the financial text to be analyzed based on the difference between the financial text to be analyzed and the target financial text includes: Extracting target text information from the financial text to be analyzed; Determining a first verification result of the financial text to be analyzed according to the difference between the financial text to be analyzed and the target financial text, and performing verification processing on the target text information to obtain a second verification result of the financial text to be analyzed; A target verification result of the financial text to be analyzed is determined according to the first verification result and the second verification result.

5. The method according to claim 1, characterized in that The segmenting process of the financial text to be analyzed to obtain the first financial text and the second financial text corresponding to the financial text to be analyzed includes: Segmenting the financial text to be analyzed to obtain a target text structure, a first sub-financial text, and a second sub-financial text corresponding to the financial text to be analyzed; the first sub-financial text is used to represent a sub-financial text carrying flag information, and the second sub-financial text is used to represent a sub-financial text not carrying flag information; The target text structure is used as the financial text to be analyzed, and the process is skipped to the step of segmenting the financial text to be analyzed until the segmentation result obtained does not contain the target text structure; Each of the first sub-financial texts obtained through the segmentation process is used as the first financial text, and each of the second sub-financial texts obtained through the segmentation process is used as the second financial text.

6. The method according to any one of claims 1 to 5, characterized in that The combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed includes: Determining a target combination method corresponding to the first financial text and the second financial text; According to the target combination method, the first target derivative financial text and the second target derivative financial text are combined to obtain a target financial text corresponding to the financial text to be analyzed.

7. A text verification device, characterized in that: The device comprises: A text acquisition module is used to acquire financial texts to be analyzed; a text segmentation module, configured to segment the financial text to be analyzed to obtain a first financial text and a second financial text corresponding to the financial text to be analyzed; the first financial text is used to represent a financial text containing a target character string, and the second financial text is used to represent a financial text not containing the target character string; a text determination module, configured to determine a first derivative financial text corresponding to the first financial text, and to determine a second derivative financial text corresponding to the second financial text; a text screening module, configured to screen out first target derivative financial texts satisfying a first preset condition from the first derivative financial texts, and to screen out second target derivative financial texts satisfying a second preset condition from the second derivative financial texts; A text combination module, used for combining the first target derivative financial text and the second target derivative financial text to obtain a target financial text corresponding to the financial text to be analyzed; The result determination module is used to determine a target verification result of the financial text to be analyzed according to the difference between the financial text to be analyzed and the target financial text.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images