Loan contract automatic reconciliation method and device based on new interest bearing mode

Abstract

The invention provides a loan contract automatic reconciliation method and device based on a new interest bearing mode, relates to the field of artificial intelligence, can also be used in the financial field, and comprises the steps: inputting a loan contract into a loan contract recognition model corresponding to the new interest bearing mode according to the new interest bearing mode corresponding to the loan contract, obtaining loan contract keywords and corresponding keyword values; generating loan contract information according to the template corresponding to the new interest bearing mode, the loan contract keyword and the corresponding keyword value; and according to the loan contract information and the acquired reconciliation text, performing loan contract automatic reconciliation based on a new interest bearing mode on the loan contract. According to the method, the new interest-bearing mode loan contract can be identified, the important information set of the new interest-bearing mode loan contract is extracted, and automatic account checking is performed based on the acquired account checking file.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence and can be used in the financial field. Specifically, it is a method and device for automatically reconciling loan contracts based on a new interest calculation model. Background Art

[0002] As the London Interbank Offered Rate (Libor) has completely withdrawn from the historical stage, the new interest calculation model loan business based on the risk-free interest rate has increased day by day. However, compared with the interbank offered rate loan contract, the new interest calculation model loan contract is more complicated and has more contractual provisions. If the loan contract based on the new interest calculation model is to be reconciled, the credit personnel of commercial financial institutions need to learn the professional knowledge and calculation methods related to the new interest calculation model loan in advance, which is quite difficult.

[0003] If professional knowledge is not available, it will be difficult to select the information to be entered into the credit system from hundreds of pages of loan contracts, which increases the probability of misidentification of contract information and reduces the efficiency of credit processing. At the same time, when a bank acts as a participating bank, the syndicate lead bank will send reconciliation and interest calculation documents, and business personnel need to ensure that the interest amount calculated by the system is consistent with that of the lead bank. If, in special circumstances, the interest calculation amount is inconsistent due to calculation errors by the lead bank, the difference amount also needs to be determined. Summary of the invention

[0004] In response to the problems in the prior art, the present application provides a method and device for automatic reconciliation of loan contracts based on a new interest calculation model, which can identify loan contracts with the new interest calculation model, extract important information sets of loan contracts with the new interest calculation model, and perform automatic reconciliation based on the obtained reconciliation files.

[0005] In order to solve the above technical problems, this application provides the following technical solutions:

[0006] In a first aspect, the present application provides a method for automatically reconciling loan contracts based on a new interest calculation model, comprising:

[0007] According to the new interest calculation mode corresponding to the loan contract, the loan contract is input into the loan contract recognition model corresponding to the new interest calculation mode to obtain the loan contract keywords and the corresponding keyword values;

[0008] Generate loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0009] The loan contract is automatically reconciled based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0010] Furthermore, before inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract to obtain the loan contract keywords and the corresponding keyword values, the method further includes:

[0011] If the loan contract is an image, performing image enhancement processing on the loan contract to obtain a clear image of the loan contract;

[0012] Performing character segmentation processing on the clear image of the loan contract to segment continuous characters into independent character regions;

[0013] Language character recognition is performed on the independent character area to obtain the full text information of the loan contract.

[0014] Furthermore, the step of pre-training the loan contract recognition model includes:

[0015] Perform data preprocessing on the full text training information corresponding to the loan training contract to obtain the corresponding word vector;

[0016] Classify the word vectors according to the new interest calculation mode corresponding to the loan training contract to obtain basic training samples corresponding to the new interest calculation mode;

[0017] The basic training samples are used to train a document topic generation model to obtain a loan contract recognition model corresponding to the new interest calculation model.

[0018] Furthermore, the step of inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtaining loan contract keywords and corresponding keyword values, includes:

[0019] Determine a corresponding loan contract identification model according to the new interest calculation model corresponding to the loan contract;

[0020] Inputting the full text information of the loan contract into the corresponding loan contract recognition model to obtain the number of topics of the loan contract and the corresponding topic probability distribution;

[0021] The loan contract keywords and corresponding keyword values ​​are determined based on a preset consistency index, the number of topics and the corresponding topic probability distribution.

[0022] Furthermore, the generating of loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values ​​includes:

[0023] Generate basic information of the loan contract according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0024] The interest calculation result of the loan contract is determined based on the obtained risk-free interest rate quotes of each currency and the basic information of the loan contract, as the loan contract information.

[0025] Furthermore, the automatic reconciliation of the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text includes:

[0026] Determine the interest calculation setting result of the loan contract according to the obtained reconciliation text;

[0027] Automatically reconcile the loan contract based on the new interest calculation model according to the interest calculation setting result and the interest calculation result to determine the interest calculation difference.

[0028] In a second aspect, the present application provides a loan contract automatic reconciliation device based on a new interest calculation model, comprising:

[0029] A word and value determination unit, used to input the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtain loan contract keywords and corresponding keyword values;

[0030] A contract information generating unit, configured to generate loan contract information according to a template corresponding to the new interest calculation mode, the loan contract keywords and corresponding keyword values;

[0031] An automatic reconciliation unit is used to automatically reconcile the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0032] Furthermore, the automatic reconciliation device for loan contracts based on the new interest calculation model further includes:

[0033] A clear image processing unit, configured to perform image enhancement processing on the loan contract to obtain a clear image of the loan contract if the loan contract is an image;

[0034] A character region division unit, used for performing character segmentation processing on the clear image of the loan contract to segment continuous characters into independent character regions;

[0035] The full information generating unit is used to perform language character recognition on the independent character area to obtain the full text information of the loan contract.

[0036] Furthermore, the word and value determination unit includes:

[0037] The vector generation module is used to perform data preprocessing on the full text training information corresponding to the loan training contract to obtain the corresponding word vector;

[0038] A training sample generation module, used to classify the word vectors according to the new interest calculation mode corresponding to the loan training contract, and obtain basic training samples corresponding to the new interest calculation mode;

[0039] The recognition model building module is used to use the basic training samples to train the document topic generation model to obtain the loan contract recognition model corresponding to the new interest calculation model.

[0040] Furthermore, the word and value determination unit includes:

[0041] An identification model matching module, used to determine a corresponding loan contract identification model according to a new interest calculation mode corresponding to the loan contract;

[0042] A topic probability calculation module, used to input the full text information of the loan contract into the corresponding loan contract recognition model to obtain the number of topics of the loan contract and the corresponding topic probability distribution;

[0043] The word and value determination module is used to determine the loan contract keywords and corresponding keyword values ​​based on a preset consistency index, the number of topics and the corresponding topic probability distribution.

[0044] Furthermore, the contract information generating unit includes:

[0045] A basic information generation module, used to generate the basic information of the loan contract according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0046] The contract information generation module is used to determine the interest calculation result of the loan contract based on the obtained risk-free interest rate quotes of various currencies and the basic information of the loan contract as the loan contract information.

[0047] Furthermore, the automatic reconciliation unit comprises:

[0048] A setting result acquisition module, used to determine the interest calculation setting result of the loan contract according to the acquired reconciliation text;

[0049] The interest difference determination module is used to automatically reconcile the loan contract based on the new interest calculation model according to the interest calculation setting result and the interest calculation result to determine the interest difference.

[0050] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for automatic reconciliation of loan contracts based on the new interest calculation model are implemented.

[0051] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for automatic reconciliation of loan contracts based on the new interest calculation model.

[0052] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for automatic reconciliation of loan contracts based on the new interest calculation model.

[0053] In response to the problems in the prior art, the automatic reconciliation method and device for loan contracts based on the new interest calculation model provided by this application can combine optical symbol recognition with the document theme generation model to perform intelligent recognition and theme training on loan contracts under the new interest calculation model, obtain the number of contract themes, keywords and corresponding values, and then form an important information set of the loan contract under the new interest calculation model, which serves as the basis for subsequent automatic interest calculation and reconciliation. Among them, the document theme generation model can dynamically train the basic training samples to better cope with the continuous changes in the terms of loan contracts under the new interest calculation model in the international market, complete the full process of reconciliation, and help business personnel save time in contract analysis, keyword extraction and reconciliation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 This is one of the flow charts of the automatic reconciliation method for loan contracts based on the new interest calculation model in an embodiment of the present application;

[0056] Figure 2 This is the second flow chart of the automatic reconciliation method for loan contracts based on the new interest calculation model in the embodiment of the present application;

[0057] Figure 3 This is a flowchart of training a loan contract recognition model in an embodiment of the present application;

[0058] Figure 4 A flowchart of obtaining loan contract keywords and corresponding keyword values ​​in an embodiment of the present application;

[0059] Figure 5 A flowchart for generating loan contract information in an embodiment of the present application;

[0060] Figure 6 This is a flowchart of automatic reconciliation of loan contracts in an embodiment of the present application;

[0061] Figure 7 This is one of the structural diagrams of the automatic reconciliation device for loan contracts based on the new interest calculation model in the embodiment of the present application;

[0062] Figure 8 This is the second structural diagram of the automatic reconciliation device for loan contracts based on the new interest calculation model in the embodiment of the present application;

[0063] Fig. 9 This is a structural diagram of a word and value determination unit in an embodiment of the present application;

[0064] Fig.10 This is a structural diagram of a word and value determination unit in an embodiment of the present application;

[0065] Fig.11 This is a structural diagram of a contract information generating unit in an embodiment of the present application;

[0066] Fig.12 This is a structural diagram of the automatic reconciliation unit in an embodiment of the present application;

[0067] Fig.13 A schematic diagram of the structure of an electronic device in an embodiment of the present application;

[0068] Fig.14 This is a flow chart of the automatic reconciliation method for loan contracts based on the new interest calculation model in an embodiment of the present application. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0070] It should be noted that the automatic reconciliation method and device for loan contracts based on the new interest calculation model provided in the present application can be used in the financial field, and can also be used in any field other than the financial field. The application field of the automatic reconciliation method and device for loan contracts based on the new interest calculation model provided in the present application is not limited.

[0071] The acquisition, storage, use and processing of data in the technical solution of this application comply with the relevant provisions of laws and regulations.

[0072] In one embodiment, see Figure 1 In order to identify the loan contract with the new interest calculation model, extract the important information set of the loan contract with the new interest calculation model, and automatically reconcile based on the obtained reconciliation file, the present application provides an automatic reconciliation method for loan contracts based on the new interest calculation model, including:

[0073] S101: inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtaining loan contract keywords and corresponding keyword values;

[0074] S102: Generate loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0075] S103: Automatically reconcile the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0076] It is understandable that in order to improve the recognition efficiency and accuracy of the elements of the new interest calculation model loan contract and automatically perform interest calculation reconciliation, the present application provides an automatic recognition and reconciliation method for the new interest calculation model loan contract that supports multiple languages.

[0077] The method aims to automatically identify information related to the new interest calculation model in multilingual loan contracts through optical character recognition (OCR) technology, automatically generate a set of key contract information elements, and automatically fill the identification information into the interest calculation template of the new interest calculation model. Then, the information in the template is compared with the information in the interest calculation reconciliation file issued by the lead bank, that is, the loan contract is automatically reconciled, which improves the efficiency and accuracy of the bank's extraction of contract information elements, thereby improving credit efficiency.

[0078] Among them, optical character recognition technology, as an important branch of computer vision technology, can obtain document images through devices with camera functions, and then use many algorithms to identify and analyze the document content.

[0079] See also Fig.14 When it is necessary to use the method provided by this application to automatically reconcile a loan contract with a new interest calculation model, it is first necessary to check whether the file carrier storing the loan contract with the new interest calculation model is an image; if so, it is necessary to first perform optical symbol recognition and other processing on it to obtain the text therein. In fact, the new interest calculation models corresponding to loan contracts are different, including: simple interest model, compound principal model, compound principal and interest model, and term interest rate model, etc. Therefore, in order to identify the elements (information) in the loan contract, it is necessary to pre-train the loan contract recognition model, and different new interest calculation models correspond to different loan contract recognition models. In specific implementation, it is necessary to train the document topic generation model (Latent Dirichlet Allocation, LDA for short) according to the new interest calculation model corresponding to the loan contract to obtain the loan contract recognition model corresponding to each new interest calculation model.

[0080] When using it, the loan contract needs to be input into the loan contract recognition model corresponding to its new interest calculation mode to obtain the loan contract keywords and corresponding keyword values. Next, the loan contract information (i.e., elements) is generated according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values. Finally, the loan contract is automatically reconciled based on the new interest calculation mode according to the loan contract information and the obtained reconciliation text.

[0081] The document topic generation model is also called the three-layer Bayesian probability model, which includes three layers of structures: words, topics, and documents. The so-called generation model means that each word in the article is usually considered to be obtained through the process of "selecting a topic with a certain probability, and selecting a word from this topic with a certain probability". The document to topic follows a multinomial distribution, and the topic to word follows a multinomial distribution.

[0082] The training process of the document topic generation model is an unsupervised machine learning process that can be used to identify the hidden topic information in a large document collection or corpus. It uses the bag of words method. This method treats each document as a word frequency vector, thereby converting text information into digital information that is easy to model. Among them, each document represents a probability distribution composed of some topics, and each topic represents a probability distribution composed of many words.

[0083] From the above description, it can be seen that the automatic reconciliation method for loan contracts based on the new interest calculation model provided by this application can combine optical symbol recognition with the document theme generation model to intelligently identify and train the loan contracts under the new interest calculation model, obtain the number of contract themes, keywords and corresponding values, and then form an important information set of the loan contract under the new interest calculation model, which serves as the basis for subsequent automatic interest calculation and reconciliation. Among them, the document theme generation model can dynamically train the basic training samples to better cope with the continuous changes in the terms of loan contracts under the new interest calculation model in the international market, complete the full process of reconciliation, and help business personnel save time in contract analysis, keyword extraction and reconciliation.

[0084] In one embodiment, see Figure 2 , before inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract to obtain the loan contract keywords and the corresponding keyword values, it also includes:

[0085] S201: If the loan contract is an image, perform image enhancement processing on the loan contract to obtain a clear image of the loan contract;

[0086] S202: performing character segmentation processing on the clear image of the loan contract to segment continuous characters into independent character regions;

[0087] S203: Performing language character recognition on the independent character region to obtain full text information of the loan contract.

[0088] It is understandable that the present device comprises four functional modules: an optical character recognition module, a document theme generation model training module, a new interest-bearing loan calculation module and a reconciliation module. Steps S201 to S203 correspond to the functions of the optical character recognition module.

[0089] The following is an example of a Chinese contract:

[0090] This module effectively identifies the language of the loan contract through OCR character recognition technology, and extracts text information from the new interest calculation model loan contract and reconciliation file, including text and numbers. This technology can support electronic images and paper contracts. For document types that can be directly edited for subsequent processing, this module can be skipped directly and the subsequent steps can be entered.

[0091] The processing of this module includes:

[0092] ① Image preprocessing. Since the loan contract image may have watermarks or uneven backgrounds, it is necessary to use image enhancement technology to eliminate image noise. The embodiment of the present application can use shadow removal technology for image enhancement.

[0093] ② Character segmentation: Segment the continuous characters in the loan contract image into independent character regions.

[0094] ③ Contract language and character recognition. First, the contract language is recognized. The GMM-UBM language recognition method is used to determine the language used in the loan contract, and the text in the image is converted into computer-recognizable text to obtain the full text information of the loan contract as the input item for subsequent steps.

[0095] ④ Reconciliation documents are usually issued by the lead bank of the syndicate. After optical symbol recognition, the corresponding text information can be obtained and saved for subsequent reconciliation use.

[0096] From the above description, it can be seen that the automatic reconciliation method of loan contracts based on the new interest calculation model provided in this application can obtain the full text information of the loan contract.

[0097] In one embodiment, see Figure 3 , the step of pre-training the loan contract recognition model comprises:

[0098] S301: Perform data preprocessing on the full text training information corresponding to the loan training contract to obtain the corresponding word vector;

[0099] S302: Classifying the word vectors according to the new interest calculation mode corresponding to the loan training contract to obtain basic training samples corresponding to the new interest calculation mode;

[0100] S303: Using the basic training samples to train a document topic generation model to obtain a loan contract recognition model corresponding to the new interest calculation model.

[0101] It can be understood that steps S301 to S303 correspond to the functions of the document topic generation model training module.

[0102] By using the previous steps to perform optical symbol recognition, we can obtain the full text information of the loan contract for training as the basic training sample. We perform data preprocessing, including word segmentation and word form restoration, and remove words that are not related to the key information of the contract to obtain word vectors.

[0103] Specifically, the following execution steps are included:

[0104] ① Data preprocessing. Perform data processing on the full text training information corresponding to the loan training contract. Since there may be no clear distinction between the characters and words in the contract, it is necessary to perform word segmentation on the training samples to form word vectors, filter out stop words, and perform word form restoration on individual words. Stop words refer to words that appear frequently but have low importance. The embodiment of the present application achieves the filtering of stop words in the loan contract under the new interest calculation model by combining a general stop word list with a special stop word list.

[0105] ②Build a loan contract recognition model (that is, train the loan contract recognition model).

[0106] The basic training samples are trained, and the number of topics with the largest consistency index is used as the number of topics. The embodiment of the present application uses the method proposed by Stevens K to obtain the optimal number of topics using the consistency index, and the calculation formula is:

[0107]

[0108] score(v i , v j ,ε)=log[p(v i v j )+ε / p(v i )p(v j )](2)

[0109] Among them, v is the word describing the topic, v i and v j are two words in v, p(v i v j) is v i and v j The probability of co-occurrence, ε is a smoothing factor equal to 1. The higher the similarity between the word and the sentence, the higher the consistency index will be. p(v i ) is v i The probability of occurrence, p(v j ) is v j The probability of occurrence. Score is the score of the vocabulary describing the topic. Coherence is the consistency of the vocabulary describing the topic.

[0110] From the above description, it can be seen that the automatic reconciliation method for loan contracts based on the new interest calculation model provided in this application can pre-train the loan contract recognition model.

[0111] In one embodiment, see Figure 4 , the step of inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtaining loan contract keywords and corresponding keyword values, includes:

[0112] S401: Determine a corresponding loan contract identification model according to the new interest calculation mode corresponding to the loan contract;

[0113] S402: Inputting the full text information of the loan contract into the corresponding loan contract recognition model to obtain the number of topics of the loan contract and the corresponding topic probability distribution;

[0114] S403: Determine the loan contract keywords and corresponding keyword values ​​based on a preset consistency index, the number of topics and the corresponding topic probability distribution.

[0115] It can be understood that steps S401 to S403 correspond to the functions of the document topic generation model training module.

[0116] ① The consistency index can be used to obtain the number of topics of the new interest-bearing loan contract, thereby obtaining the vocabulary probability distribution of each topic, and then classifying the keywords of the new interest-bearing loan contract, and finally obtaining topics such as contract issuance date, contract withdrawal date, maturity date, interest accrual date, interest calculation period, new interest calculation model type, term, currency, contract agreed amount, quotation, credit adjustment spread, floating rate, observation period transfer, accuracy, interest rate bottom, interest rate top, interest rate upper limit and interest rate lower limit.

[0117] ② After obtaining the keywords (topics) in the previous step, it is necessary to identify and obtain the data and units corresponding to each topic, such as the period is 5 years.

[0118] ③The model can be dynamically trained by continuously importing incremental new interest-bearing loan contract text data, and the number of topics of new interest-bearing loan contracts can be continuously updated over time.

[0119] From the above description, it can be seen that the automatic reconciliation method for loan contracts based on the new interest calculation model provided in this application can input the loan contract into a loan contract recognition model corresponding to the new interest calculation model according to the new interest calculation model corresponding to the loan contract, and obtain loan contract keywords and corresponding keyword values.

[0120] In one embodiment, see Figure 5 , the generating of loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values ​​includes:

[0121] S501: Generate basic information of the loan contract according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0122] S502: Determine the interest calculation result of the loan contract according to the obtained risk-free interest rate quotes of each currency and the basic information of the loan contract as the loan contract information.

[0123] It can be understood that steps S501 to S502 correspond to the functions of the new interest-bearing loan calculation module.

[0124] There is a special calculation method for the interest calculation of new interest calculation models. Currently, the new interest calculation models currently in use internationally include simple interest model, compound principal model, compound principal and interest model, and term interest rate model. The calculation formula for each model is different. This module will extract the keywords and corresponding data output in the previous step, and match the keywords (topics) obtained in the previous step with the calculation template input items through the manual definition window, so as to obtain the interest calculation result through the calculation model formula.

[0125] The following takes the compound interest rate model as an example to illustrate the specific processing process:

[0126] ① After the previous step is completed, the keywords and corresponding data of the new interest-bearing loan contract will be output. Whether these data need to be substituted into the formula calculation needs to be manually defined. As the new interest calculation model in the international market continues to be enriched and optimized, the input items of the formula may change in the future. Therefore, it is necessary to set up a manual definition window to match the keywords and corresponding data obtained in the previous step with the input items of the new interest calculation model. At the same time, the input items in the calculation model also include content that must be entered in advance, such as the start date and end date of the interest calculation period. See Table 1.

[0127] Table 1 Schematic diagram of manual window customization

[0128] Keyword data Template Input Contract Amount principal Interest rate ceiling Interest rate ceiling Interest rate bottom Interest rate bottom Interest rate cap Interest rate cap Interest rate floor Interest rate floor Observation period transfer days Observation period transfer days - Interest Calculation Period Start Date - Interest accrual period end date ...... ......

[0129] ② Automatically obtain risk-free interest rate quotes for each currency from the syndicate lead bank.

[0130] ③ Automatically bring the input items and manually customized items into the calculation model to obtain the interest calculation result. The formula is:

[0131]

[0132] Among them, i is the i-th working day, d b is the number of working days in the interest calculation period, Principle is the principal, NCR is the non-cumulative compound interest rate, CAS is the credit adjustment spread, Margin is the spread, n is the number of interest calculation days, and N is the number of interest rate conversion days.

[0133] From the above description, it can be seen that the automatic reconciliation method for loan contracts based on the new interest calculation model provided in this application can generate loan contract information according to the template corresponding to the new interest calculation model, the loan contract keywords and the corresponding keyword values.

[0134] In one embodiment, see Figure 6 , the automatic reconciliation of the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text includes:

[0135] S601: Determine the interest calculation setting result of the loan contract according to the obtained reconciliation text;

[0136] S602: Automatically reconcile the loan contract based on the new interest calculation model according to the interest calculation setting result and the interest calculation result to determine the interest calculation difference.

[0137] It can be understood that steps S601 to S602 correspond to the functions of the reconciliation module.

[0138] The reconciliation text (including the interest calculation setting result) obtained in the optical character recognition module is automatically compared with the interest calculation result calculated in the previous step, and the comparison result and the comparison difference number (also called the interest calculation difference) can be displayed. This process can be a simple mathematical calculation.

[0139] From the above description, it can be seen that the automatic reconciliation method of loan contracts based on the new interest calculation model provided in this application can automatically reconcile the loan contracts based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0140] Based on the same inventive concept, the embodiment of the present application also provides an automatic reconciliation device for loan contracts based on a new interest calculation model, which can be used to implement the method described in the above embodiment, as described in the following embodiment. Since the principle of solving the problem by the automatic reconciliation device for loan contracts based on the new interest calculation model is similar to the automatic reconciliation method for loan contracts based on the new interest calculation model, the implementation of the automatic reconciliation device for loan contracts based on the new interest calculation model can refer to the implementation of the method based on software performance benchmark determination, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0141] In one embodiment, see Figure 7 In order to identify the loan contract with the new interest calculation model, extract the important information set of the loan contract with the new interest calculation model, and perform automatic reconciliation based on the obtained reconciliation file, the present application provides an automatic reconciliation device for loan contracts based on the new interest calculation model, including: a word and value determination unit 701, a contract information generation unit 702 and an automatic reconciliation unit 703.

[0142] The word and value determination unit 701 is used to input the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtain the loan contract keywords and corresponding keyword values;

[0143] A contract information generating unit 702, configured to generate loan contract information according to a template corresponding to the new interest calculation mode, the loan contract keywords and corresponding keyword values;

[0144] The automatic reconciliation unit 703 is used to automatically reconcile the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0145] In one embodiment, see Figure 8 The automatic reconciliation device for loan contracts based on the new interest calculation model also includes: a clear image processing unit 801, a character area division unit 802 and a full information generation unit 803.

[0146] A clear image processing unit 801 is used for performing image enhancement processing on the loan contract to obtain a clear image of the loan contract if the loan contract is an image;

[0147] The character region division unit 802 is used to perform character segmentation processing on the clear image of the loan contract to segment continuous characters into independent character regions;

[0148] The full information generating unit 803 is used to perform language character recognition on the independent character area to obtain the full text information of the loan contract.

[0149] In one embodiment, see Fig. 9 The word and value determination unit 701 includes: a vector generation module 901, a training sample generation module 902 and a recognition model construction module 903.

[0150] The vector generation module 901 is used to perform data preprocessing on the full text training information corresponding to the loan training contract to obtain the corresponding word vector;

[0151] The training sample generation module 902 is used to classify the word vectors according to the new interest calculation mode corresponding to the loan training contract to obtain basic training samples corresponding to the new interest calculation mode;

[0152] The recognition model building module 903 is used to use the basic training samples to train the document topic generation model to obtain the loan contract recognition model corresponding to the new interest calculation model.

[0153] In one embodiment, see Fig.10 The word and value determination unit 701 includes: a recognition model matching module 1001, a topic probability calculation module 1002 and a word and value determination module 1003.

[0154] The identification model matching module 1001 is used to determine the corresponding loan contract identification model according to the new interest calculation mode corresponding to the loan contract;

[0155] The topic probability calculation module 1002 is used to input the full text information of the loan contract into the corresponding loan contract recognition model to obtain the number of topics of the loan contract and the corresponding topic probability distribution;

[0156] The word and value determination module 1003 is used to determine the loan contract keywords and corresponding keyword values ​​based on a preset consistency index, the number of topics and the corresponding topic probability distribution.

[0157] In one embodiment, see Fig.11 The contract information generating unit 702 includes: a basic information generating module 1101 and a contract information generating module 1102.

[0158] A basic information generation module 1101 is used to generate basic information of the loan contract according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0159] The contract information generating module 1102 is used to determine the interest calculation result of the loan contract according to the obtained risk-free interest rate quotes of various currencies and the basic information of the loan contract as the loan contract information.

[0160] In one embodiment, see Fig.12 The automatic reconciliation unit 703 includes: a setting result acquisition module 1201 and an interest difference determination module 1202.

[0161] The setting result acquisition module 1201 is used to determine the interest calculation setting result of the loan contract according to the acquired reconciliation text;

[0162] The interest difference determination module 1202 is used to automatically reconcile the loan contract based on the new interest calculation model according to the interest calculation setting result and the interest calculation result to determine the interest difference.

[0163] From the hardware level, in order to be able to identify the loan contract with the new interest calculation model, extract the important information set of the loan contract with the new interest calculation model, and automatically reconcile based on the acquired reconciliation file, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the automatic reconciliation method for loan contracts based on the new interest calculation model, and the electronic device specifically includes the following contents:

[0164] Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the automatic reconciliation device for loan contracts based on the new interest calculation model and related devices such as core business systems, user terminals and related databases; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the automatic reconciliation method for loan contracts based on the new interest calculation model and the embodiment of the automatic reconciliation device for loan contracts based on the new interest calculation model in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0165] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0166] In practical applications, part of the automatic reconciliation method for loan contracts based on the new interest calculation model can be executed on the electronic device side as described above, or all operations can be completed in the client device. The specific selection can be based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0167] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0168] Fig.13 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig.13 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig.13 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0169] In one embodiment, the function of the automatic reconciliation method for loan contracts based on the new interest calculation model can be integrated into the central processor 9100. The central processor 9100 can be configured to perform the following control:

[0170] S101: inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtaining loan contract keywords and corresponding keyword values;

[0171] S102: Generate loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0172] S103: Automatically reconcile the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0173] From the above description, it can be seen that the automatic reconciliation method for loan contracts based on the new interest calculation model provided by this application can combine optical symbol recognition with the document theme generation model to intelligently identify and train the loan contracts under the new interest calculation model, obtain the number of contract themes, keywords and corresponding values, and then form an important information set of the loan contract under the new interest calculation model, which serves as the basis for subsequent automatic interest calculation and reconciliation. Among them, the document theme generation model can dynamically train the basic training samples to better cope with the continuous changes in the terms of loan contracts under the new interest calculation model in the international market, complete the full process of reconciliation, and help business personnel save time in contract analysis, keyword extraction and reconciliation.

[0174] In another embodiment, the automatic reconciliation device for loan contracts based on the new interest calculation model can be configured separately from the central processor 9100. For example, the data composite transmission device can be configured as a chip connected to the central processor 9100, and the function of the automatic reconciliation method for loan contracts based on the new interest calculation model can be realized through the control of the central processor.

[0175] like Fig.13 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig.13 In addition, the electronic device 9600 may also include Fig.13 For components not shown, reference may be made to the prior art.

[0176] like Fig.13 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0177] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0178] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0179] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0180] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0181] The communication module 9110 is a transmitter / receiver 9110 that sends and receives signals via an antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0182] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0183] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the automatic reconciliation method for loan contracts based on a new interest calculation model in the above-mentioned embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the automatic reconciliation method for loan contracts based on a new interest calculation model in the above-mentioned embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0184] S101: inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtaining loan contract keywords and corresponding keyword values;

[0185] S102: Generate loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values;

[0186] S103: Automatically reconcile the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

[0187] From the above description, it can be seen that the automatic reconciliation method for loan contracts based on the new interest calculation model provided by this application can combine optical symbol recognition with the document theme generation model to intelligently identify and train the loan contracts under the new interest calculation model, obtain the number of contract themes, keywords and corresponding values, and then form an important information set of the loan contract under the new interest calculation model, which serves as the basis for subsequent automatic interest calculation and reconciliation. Among them, the document theme generation model can dynamically train the basic training samples to better cope with the continuous changes in the terms of loan contracts under the new interest calculation model in the international market, complete the full process of reconciliation, and help business personnel save time in contract analysis, keyword extraction and reconciliation.

[0188] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0189] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0190] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0191] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0192] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for automatic reconciliation of loan contracts based on a new interest calculation model, characterized in that: include: According to the new interest calculation mode corresponding to the loan contract, the loan contract is input into the loan contract recognition model corresponding to the new interest calculation mode to obtain the loan contract keywords and the corresponding keyword values; Generate loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values; The loan contract is automatically reconciled based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

2. The automatic reconciliation method for loan contracts based on the new interest calculation model according to claim 1 is characterized in that: Before inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract to obtain the loan contract keywords and the corresponding keyword values, the method further includes: If the loan contract is an image, performing image enhancement processing on the loan contract to obtain a clear image of the loan contract; Performing character segmentation processing on the clear image of the loan contract to segment continuous characters into independent character regions; Language character recognition is performed on the independent character area to obtain the full text information of the loan contract.

3. The automatic reconciliation method for loan contracts based on the new interest calculation model according to claim 1 is characterized in that: The step of pre-training the loan contract recognition model comprises: Perform data preprocessing on the full text training information corresponding to the loan training contract to obtain the corresponding word vector; Classify the word vectors according to the new interest calculation mode corresponding to the loan training contract to obtain basic training samples corresponding to the new interest calculation mode; The basic training samples are used to train a document topic generation model to obtain a loan contract recognition model corresponding to the new interest calculation model.

4. The automatic reconciliation method for loan contracts based on the new interest calculation model according to claim 2 is characterized in that: The step of inputting the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtaining loan contract keywords and corresponding keyword values, includes: Determine a corresponding loan contract identification model according to the new interest calculation model corresponding to the loan contract; Inputting the full text information of the loan contract into the corresponding loan contract recognition model to obtain the number of topics of the loan contract and the corresponding topic probability distribution; The loan contract keywords and corresponding keyword values ​​are determined based on a preset consistency index, the number of topics and the corresponding topic probability distribution.

5. The automatic reconciliation method for loan contracts based on the new interest calculation model according to claim 1 is characterized in that: The generating of loan contract information according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values ​​includes: Generate basic information of the loan contract according to the template corresponding to the new interest calculation mode, the loan contract keywords and the corresponding keyword values; The interest calculation result of the loan contract is determined based on the obtained risk-free interest rate quotes of each currency and the basic information of the loan contract, as the loan contract information.

6. The automatic reconciliation method for loan contracts based on the new interest calculation model according to claim 5 is characterized in that: The automatic reconciliation of the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text includes: Determine the interest calculation result of the loan contract based on the obtained reconciliation text; Automatically reconcile the loan contract based on the new interest calculation model according to the interest calculation setting result and the interest calculation result to determine the interest calculation difference.

7. An automatic reconciliation device for loan contracts based on a new interest calculation model, characterized in that: include: A word and value determination unit, used to input the loan contract into a loan contract recognition model corresponding to the new interest calculation mode according to the new interest calculation mode corresponding to the loan contract, and obtain loan contract keywords and corresponding keyword values; A contract information generating unit, configured to generate loan contract information according to a template corresponding to the new interest calculation mode, the loan contract keywords and corresponding keyword values; An automatic reconciliation unit is used to automatically reconcile the loan contract based on the new interest calculation model according to the loan contract information and the obtained reconciliation text.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the automatic reconciliation method for loan contracts based on the new interest calculation model as described in 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 automatic reconciliation method for loan contracts based on a new interest calculation model as described in any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the automatic reconciliation method for loan contracts based on a new interest calculation model as described in any one of claims 1 to 6 are implemented.

Images