CN115719258A

CN115719258A - Method and device for automatically generating quotation and electronic equipment

Info

Publication number: CN115719258A
Application number: CN202211457872.1A
Authority: CN
Inventors: 王礼文; 陈宇珽; 邬桐; 王一宁
Original assignee: Cfets Information Technology Shanghai Co ltd
Current assignee: Cfets Information Technology Shanghai Co ltd
Priority date: 2022-11-21
Filing date: 2022-11-21
Publication date: 2023-02-28

Abstract

The embodiment of the invention discloses a method and a device for automatically generating a quotation and electronic equipment. The embodiment of the invention obtains the transaction instruction text in the dialog box of the instant messaging tool; processing the trading instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese pre-trained model, an intention classification model and an entity recognition model; and displaying the quotation on an interface where a dialog box of the instant messaging tool is located. By the method, the trading instruction text can be automatically acquired in the trading process, and then the quotation is automatically generated through the pre-trained model and the entity normalization, so that a user does not need to manually generate the quotation after searching the trading instruction text in a dialog box one by one, the waste of time can be reduced, and the working efficiency is improved.

Description

Method and device for automatically generating quotation and electronic equipment

Technical Field

The invention relates to the technical field of natural language processing, in particular to a method and a device for automatically generating a quotation and electronic equipment.

Background

With the development of the financial market, the number of financial institutions such as banks and securities is increased, and in the financial market among the financial institutions, various types of transactions, such as common repurchase transactions, foreign currency borrowing transactions, and the like, need to be performed among different financial institutions.

In the prior art, as the professionality of the financial market is higher and the transaction timeliness is also very high, in order to save communication time and accelerate transaction efficiency, a trader engaged in transaction can communicate with a potential trading opponent through an instant communication tool, usually a trading instruction text is used for expressing the intentions of quotation, bargain and the like of the transaction, for example, in foreign currency borrowing transaction of the foreign exchange market, when a trader A sends a trading instruction text "USD 3W Borrow20K @2.6%" to a trader B, the trader A represents that the trader A initiates a quotation request of "borrowing 2 ten thousand dollars for 3 weeks at an interest rate of 2.6%) to an institution of the trader B; however, in a high-frequency period of the transaction, the trader needs to screen and pay attention to the transaction information from a large amount of chat text information in each chat window all the time, contact the opponent after selecting the opponent willing to trade with the trader, then generate the quotation and confirm the transaction, so that a large amount of time needs to be wasted in the process of generating the quotation, and the working efficiency is low.

In summary, how to reduce the waste of time and improve the work efficiency in the transaction process is a problem to be solved.

Disclosure of Invention

In view of this, the embodiment of the present invention provides a method and an apparatus for automatically generating a quotation, and an electronic device, which can reduce time waste and improve work efficiency during a transaction.

In a first aspect, an embodiment of the present invention provides a method for automatically generating a quotation, where the method includes: acquiring a transaction instruction text in a dialog box of the instant messaging tool; processing the trading instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese pre-trained model, an intention classification model and an entity recognition model; and displaying the quotation on an interface where a dialog box of the instant messaging tool is located.

Optionally, the method further includes:

receiving a trading instruction, wherein the trading instruction is used for confirming the quotation or rejecting the quotation.

Optionally, the method further includes:

sending the transaction instruction text in a dialog box of the instant messaging tool; alternatively, the first and second liquid crystal display panels may be,

and receiving the transaction instruction text in a dialog box of the instant messaging tool.

Optionally, the processing the transaction instruction text through a pre-trained model and entity normalization to generate a quotation, specifically including;

processing the transaction instruction text through a pre-trained model and entity normalization, and outputting an output result expressed by json;

and generating the quotation according to the output result expressed by the json.

Optionally, the chinese pre-training model is a FinBERT model, wherein the FinBERT model is a financial field pre-training model.

Optionally, the entity normalization is used to represent different words with the same meaning in a uniform manner.

Optionally, the method further includes:

receiving an information display instruction, wherein the information display instruction is used for displaying detail information of the quotation, and the detail information is displayed in an interface where the dialog box is located or is suspended on the interface where the dialog box is located.

In a second aspect, an embodiment of the present invention provides an apparatus for automatically generating a quotation, where the apparatus includes: the acquisition unit is used for acquiring a transaction instruction text in a dialog box of the instant messaging tool; the processing unit is used for processing the transaction instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese and pre-trained model, an intention classification model and an entity recognition model; and the display unit is used for displaying the quotation on the interface where the dialog box of the instant messaging tool is positioned.

Optionally, the apparatus further comprises: the receiving unit is used for receiving a trading instruction, wherein the trading instruction is used for confirming the quotation or rejecting the quotation.

Optionally, the apparatus further comprises: the sending unit is used for sending the transaction instruction text in a dialog box of the instant messaging tool; or, the receiving unit is further configured to receive the transaction instruction text in a dialog box of the instant messaging tool.

Optionally, the processing unit is specifically configured to:

Optionally, the receiving unit is further configured to:

In a third aspect, an embodiment of the present invention provides a computer-readable storage medium on which computer program instructions are stored, which when executed by a processor implement the method according to the first aspect or any one of the possibilities of the first aspect.

In a fourth aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, the memory being configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to the first aspect or any one of the possibilities of the first aspect.

The embodiment of the invention obtains the transaction instruction text in the dialog box of the instant messaging tool; processing the trading instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese pre-trained model, an intention classification model and an entity recognition model; and displaying the quotation on an interface where a dialog box of the instant messaging tool is located. By the method, the transaction instruction text can be automatically acquired in the transaction process, and then the quotation is automatically generated through the pre-trained model and the entity normalization, so that a user does not need to manually generate the quotation after searching the transaction instruction text in a dialog box one by one, the waste of time can be reduced, and the working efficiency is improved.

Drawings

The above and other objects, features and advantages of the present invention will become more apparent from the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

FIG. 1 is a diagram illustrating a dialog box interface of an instant messenger tool according to the prior art;

FIG. 2 is a schematic diagram of a dialog box interface of another instant messenger in the prior art;

FIG. 3 is a flow chart of a method for automatic generation of a quotation according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a dialog box interface of an instant messenger in accordance with an embodiment of the present invention;

FIG. 5 is a schematic diagram of a dialog box interface of another instant messenger according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of an intent classification and entity recognition model architecture in accordance with an embodiment of the present invention;

FIG. 7 is a schematic diagram of a dialog box interface of another instant messenger according to an embodiment of the present invention;

FIG. 8 is a flow diagram of another method for automatic generation of a quotation according to an embodiment of the present invention;

FIG. 9 is a flow chart of yet another method for automatic generation of a quotation according to an embodiment of the present invention;

FIG. 10 is a schematic diagram of a dialog box interface of an instant messenger in accordance with an embodiment of the present invention;

FIG. 11 is a schematic diagram of an apparatus for automatically generating a quotation according to an embodiment of the present invention;

fig. 12 is a schematic diagram of an electronic device according to an embodiment of the invention.

Detailed Description

The present disclosure is described below based on examples, but the present disclosure is not limited to only these examples. In the following detailed description of the present disclosure, certain specific details are set forth. It will be apparent to those skilled in the art that the present disclosure may be practiced without these specific details. Well-known methods, procedures, components, and circuits have not been described in detail so as not to obscure the present disclosure.

Furthermore, those of ordinary skill in the art will appreciate that the drawings provided herein are for illustrative purposes and are not necessarily drawn to scale.

Unless the context clearly requires otherwise, throughout this application, the words "comprise", "comprising", and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is, what is meant is "including, but not limited to".

In the description of the present disclosure, it is to be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In addition, in the description of the present disclosure, "a plurality" means two or more unless otherwise specified.

In the prior art, because the professionality of the financial market is high and the transaction timeliness is very high, in order to save communication time and accelerate transaction efficiency, a trader engaged in the transaction can communicate with a potential trading opponent through an instant communication tool in a single chat mode, and the intentions of quotation and bargaining of the transaction are usually expressed by using a transaction instruction text, for example, in a foreign currency borrowing transaction of the foreign currency market, when a trader A sends a transaction instruction text "USD 3W Borrow 2K @2.6%" to a trader B in a dialog box of the instant communication tool, as shown in fig. 1 in particular, fig. 1 is a dialog box interface of the trader A, and in fig. 1, the trader A comprises a multi-sentence dialog in the dialog box with the trader B, wherein the above-mentioned "USD 3W Borrow 2K @2.6%" is only one of the sentences, and represents that the trader A initiates a loan request for one institution to the institution of the trader B for "3 ten thousand dollars at an interest rate of 2.6"; however, during high frequency periods of the transaction, the trader may be communicating with multiple trading objects simultaneously, for example, as shown in fig. 2, trader a is shown on the instant messenger interface, and trader a is shown at 13:00 to 13: the method comprises the following steps of simultaneously communicating with a trader 1, a trader 2, a trader 3, a trader 4, a trader 5, a trader 6, a trader 7, a trader 8, a trader 9 and a trader 10 within 30 time periods, wherein 10 traders in total communicate, therefore, a trader A needs to screen and pay attention to trading information in a large amount of chatting text information of a chatting window of each trading object at any moment, contact the other party after selecting an opponent party willing to trade with the trader, then generate a quotation sheet to confirm the trading, and waste a large amount of time in the process of generating the quotation sheet, so that the working efficiency is low; therefore, how to reduce the waste of time and improve the work efficiency in the transaction process is a problem to be solved.

In the embodiment of the present invention, fig. 3 is a flowchart of a method for automatically generating a quotation according to the embodiment of the present invention. As shown in fig. 3, the method specifically includes the following steps:

step S300, obtaining a transaction instruction text in a dialog box of the instant messaging tool.

Specifically, the transaction instruction text may be sent by the trader a to the trader B in a dialog box of the instant messenger, for example, as shown in fig. 4, fig. 4 is a dialog box interface of the instant messenger of the trader a, the trader a sends a transaction instruction text "[ AA/AAA ] PPN, africa, nanjing siwye new city building (group) limited company, 10 billion (5 billion released in this period), 5 years, 3.6% -5.5%, middle-debt credit-accrual investment limited company guarantee (AAA), 9 months" to the trader B in the dialog box of the instant messenger.

Alternatively, the transaction instruction text may also be sent by the trader a in a dialog box of the instant messenger, for example, as shown in fig. 5, fig. 5 is a dialog box interface of the instant messenger of the trader a, and the trader a receives the transaction instruction text "[ AA/AAA ] PPN, africa, nanjing sijon new city building (group) limited company, 10 billion yuan (5 billion yuan issued in this period), 5 years, 3.6% -5.5%, and the medium-debt credit accrual investment portfolio limited company guarantee (AAA), 9 months, sent by the trader B in the dialog box of the instant messenger.

In the embodiment of the present invention, the source of the transaction instruction text is not limited, and the transaction instruction text may be sent by a trader or received by the trader, and any transaction instruction text may be obtained, and the specific transaction instruction text is determined according to actual conditions.

Step S301, processing the transaction instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese pre-trained model, an intention classification model and an entity recognition model.

Specifically, the transaction instruction text is processed through a pre-trained model and entity normalization, and an output result expressed by json is output; and generating the quotation according to an output result expressed by the json.

In one possible implementation manner, the chinese pre-training model is a FinBERT model, wherein the FinBERT model is a financial field pre-training model; the FinBERT model is used for solving the problem that a universal pre-training model is relatively strange due to the fact that many proprietary namewords are contained in the language materials among banks, the FinBERT model is based on three types of financial domain language materials including financial and financial news, research and newspaper, company announcements on the market and financial encyclopedia entries, important parts are screened and preprocessed to obtain the language materials finally used for the FinBERT model training, the language materials totally contain 30 hundred million language symbols (Tokens), and the number of the language materials exceeds the scale of native Chinese BERT pre-training, so that the FinBERT pre-training model obtains remarkable performance improvement in downstream tasks of multiple financial domains, and F1-score (score) is directly improved by at least 2-5.7 percentage points without any additional adjustment; in the embodiment of the invention, the pre-training model can be directly used on a Natural Language Processing (NLP) data set in the financial field, a model does not need to be built from the beginning to solve the similar NLP problem, and a large amount of time and computing resources can be saved by using the pre-training model.

In one possible implementation, the intention classification and entity recognition model is used for intention classification and entity recognition, wherein the intention in the intention classification is a transaction intention, which may also be referred to as a transaction purpose, and includes a currency cash note issue, a currency cash note deal, a currency cash note offer, a currency repurchase deal, a currency interest rate interchange offer, a currency fund offer, a foreign exchange swap deal, a foreign exchange forward deal, a foreign exchange loan deal, a foreign exchange option deal, and the like; entities in the entity identification comprise bonds, issuers, subject ratings, enterprise types, issuing modes, scales, periods, yield rates, dates, directions, total amount of securities faces, clearing speed, states, transaction amount, prices and the like; the method is determined according to actual conditions.

In the embodiment of the present invention, the structure of the intention classification and Entity recognition (DIET) model is shown in fig. 6, the intention (Intent) of the DIET model is "index", the Entity of the DIET model is "beauty", "meta", "middle", "price", and "between", "Feed-forward (FFW)" are shared weights, the input sentence is regarded as a sequence of tokens, the tokens may be words or sub-words according to different pipelines (pipeline), and a special classification tag u _ CLS _ is added at the end of each sentence; wherein each input logogram has sparse features which are one-hot or multi-hot codes of n-grams (n < = 5), and dense features which contain many redundant features from a pre-trained model. ConveRT is used as the encoder for sentences, and the _ CLS vector of ConveRT is used as the initial input of DIET as a sentence information feature in addition to word information. The sparse feature is calculated through full connection, wherein the weight of the full connection is shared with the full connection of other sequences at time steps, the purpose is to make the dimension of the sparse feature consistent with the dimension of the dense feature, and then the output of the FFW of the sparse feature and the vector of the dense feature are subjected to concat splicing, because the vector dimension required to be input by the transform needs to be consistent, the FFW is connected after splicing, and the weight of the FFW is also shared; the transformer module, using a 2-layer, uses token's relative position encoding.

The entity identification classification label is calculated according to a Conditional Random Field (CRF), and a vector of an output of a transform is adopted, so that a corresponding input token position can be found according to the output vector; the intention classification is to calculate the similarity after CLS is output through transformations, then the intention label vectorization, dot-product loss is adopted during loss calculation, so that the similarity and the similarity of a real intention are the highest, and the similarity of negative sampling calculation and other intentions is reduced. Inspired by a Masking language modeling task, an additional training target is added to predict an input token of random masked, then after the token of the part is calculated by a transducer, the token and the original word are subjected to vector sum FFN, and then click loss is calculated. Finally, calculating Total Loss of the Total Loss, wherein the Total Loss is the sum of three Loss parts, namely Entity Loss (Entity Loss), intention Loss (Total Loss) and Mask Loss (Mask Loss); the DIET model further comprises an Embedding Layer (Embedding Layer), similarity (similarity) and the like, intent classification and entity identification are carried out through the DIET model transaction instruction text, in the implementation of the invention, other models can be adopted to achieve the purpose, and the embodiment of the invention is not limited.

In one possible implementation, the entity normalization is used to represent different words with the same meaning in a uniform manner.

For example, the entity normalization process includes: firstly, determining a current entity, then determining an entity relationship, and finally determining a target entity; <xnotran> , " 7 ", " 7 ", " 1 ", " ", " ", " ", " 1 ", " 1 ", " ", " ", " 1 ", " 1 " " ", " ", " ", , , ; </xnotran> The specific entity normalization is determined according to actual conditions, and is only exemplary.

In one possible implementation, the transaction instruction text "[ AA/AAA ] PPN, not published, nanjing sihe new city construction (group) limited, 10 billion yuan (5 billion yuan released at this stage), 5 years, 3.6% -5.5%, middle debt credit accrual investment equity limited guarantee (AAA), 9 months" output results expressed in json after text error correction, intent classification and entity identification are as follows: "Intent": "ask _ coupon issuance", "Confidence": 1.0, "ErrorMsg (error)": null (none), "NLU Result (natural language understanding)" { "Domain (range)", "home currency", "layer (Buyer)", "receiver id" ], "Seller)": [ "sender id" ], "Trade Category": null, "Business scene": "issues", "Entities (Entities)": "rating", "Start (Start)": 1, "End (End)": 3,

"Role": "publisher body rating", "Raw Value": "AA" }, { "Entity": "rating", "Start (Start)": 4, "End (End)": 7, "Role": "debt rating", "Raw Value (Raw Value)": "AAA" }, { "Entity": "release method", "Start (Start)": 12, "End": 15, "Role": "release method", "Raw Value": "unpublished" }, { "Entity": "issuer", "Start": 16,

"End": 32, "Role": "issuer", "Raw Value": "Nanjing Liuhe New City construction (group) Co., ltd.",

{ "Entity": "scale value", "Start (Start)": 42, "End (End)": 43, "Role": "scale Value", "Raw Value": "5" }, { "Entity": "scale unit",

"Start": 43, "End (End)": 45, "Role":

"scale unit", "Raw Value": "one hundred million yuan",

"Normalized Value": "hundred million" }, { "Entity":

"deadline", "Start": 47, "End": 49, "Role": "term", "Raw Value": the number of the 5 years is increased,

{ "Entity": "rate of return", "Start (Start)": 51, "End (End)": 55, "Role": "yield (lowest)", "Raw Value": "3.6%" }, { "Entity": "rate of return", "Start": 56, "End": 60, "Role":

"yield (highest)", "Raw Value": "5.5%", which is not described again, this time is merely an exemplary description of the output result, and is determined according to the actual situation.

In an embodiment of the present invention, the numerical values 1,3,4,7, 12, 15, etc. are sorting values of characters in the transaction instruction text.

In the embodiment of the invention, a final output result is obtained through the links of text error correction, intention classification, entity identification, entity normalization and result output, and then a quotation is generated according to the output result.

And step S302, displaying the quotation on an interface where a dialog box of the instant messaging tool is located.

Specifically, the quotation can be displayed on the right side of an interface where a dialog box of an instant messaging tool used by a trader is located, specifically as shown in fig. 7, where fig. 7 is an interface schematic diagram of a trader a, where the quotation includes: the distributor: nanjing Liuhe new city construction (group) Co., ltd., term: 5 years, release scale: 5 hundred million, yield: 3.6% -5.5%, and the quotation further comprises option boxes of 'confirm' and 'reject'; the description is given for illustrative purposes only and is determined according to actual conditions.

In the embodiment of the invention, the quotation can be automatically generated after the trading instruction text in the dialog box is acquired, the working efficiency is improved, the time is saved, and the quotation and trading instruction texts of various trading products in the home currency market and the foreign exchange market can be supported, so that the application range is wide.

In a possible implementation manner, after step S302, other steps are included, and fig. 8 is a flowchart of a method for automatically generating a quotation according to an embodiment of the present invention. As shown in fig. 8, the method specifically includes the following steps:

step S303, receiving a trading instruction, wherein the trading instruction is used for confirming the quotation or rejecting the quotation.

Specifically, in fig. 7, while the quotation is displayed on the right side of the interface where the dialog box of the instant messaging tool is located, the option boxes of "confirm" and "reject" are also displayed, and when the trader triggers the option box of "confirm", that is, the trader confirms the transaction; when the trader triggers the "decline" option box, i.e., the trader declines the transaction; the quotation corresponding to the confirmed or rejected transaction can be deleted from the right side of the interface where the dialog box of the instant messaging tool is located.

In a possible implementation manner, before step S303, other steps are further included, and fig. 9 is a flowchart of a method for automatically generating a quotation according to an embodiment of the present invention. As shown in fig. 9, the method specifically includes the following steps:

step S304, receiving an information display instruction, wherein the information display instruction is used for displaying detail information of the quotation, and the detail information is displayed in an interface where the dialog box is located or is suspended on the interface where the dialog box is located.

In a possible implementation manner, in fig. 7, the quotation list displayed on the right side of the interface where the dialog box of the instant messaging tool is located only displays important information, and due to the limitation of the size of the interface, all detail information cannot be displayed, and if the trader needs to confirm the detail information before trading, an information display instruction is issued to call the detail information of the quotation list, and the detail information may be suspended on the interface where the dialog box is located, specifically as shown in fig. 10, the method includes: a distributor: nanjing Liuhe New City construction (group) Co., ltd., term: 5 years, release scale: 5 hundred million, yield: 3.6% -5.5%, subject assessment: AA, debt rating: AAA, issue method: non-publicly, the warranty agency: the medium debt credit promotes the guarantee of investment shares company, etc.; the method comprises the steps that the confirmation and rejection option boxes are determined according to actual conditions, in addition, the 'confirmation' and 'rejection' option boxes are also displayed in a quotation suspending on an interface where a dialog box of the instant messaging tool is located, and when a trader triggers the 'confirmation' option box, the trader confirms the transaction; when the trader triggers the "decline" option box, the trader declines the transaction.

Fig. 11 is a schematic diagram of an apparatus for automatically generating a quotation according to an embodiment of the present invention. As shown in fig. 11, the apparatus of the present embodiment includes an acquisition unit 1101, a processing unit 1102, and a display unit 1103.

The obtaining unit 1101 is configured to obtain a transaction instruction text in a dialog box of an instant messaging tool;

the processing unit 1102 is configured to process the transaction instruction text through a pre-trained model and entity normalization to generate a quotation, where the pre-trained model includes a chinese and pre-trained model, an intention classification model, and an entity recognition model;

a display unit 1103, configured to display the quotation on the interface where the dialog box of the instant messaging tool is located.

Further, the apparatus further comprises: the receiving unit is used for receiving a trading instruction, wherein the trading instruction is used for confirming the quotation or rejecting the quotation.

Further, the apparatus further comprises: the sending unit is used for sending the transaction instruction text in a dialog box of the instant messaging tool; or, the receiving unit is further configured to receive the transaction instruction text in a dialog box of the instant messaging tool.

Further, the processing unit is specifically configured to:

and generating the quotation according to an output result expressed by the json.

Further, the Chinese pre-training model is a FinBERT model, wherein the FinBERT model is a financial field pre-training model.

Further, the entity normalization is used to represent different words with the same meaning in a uniform manner.

Further, the receiving unit is further configured to:

FIG. 12 is a schematic view of an electronic device of an embodiment of the invention. As shown in fig. 12, the apparatus for automatically generating a quotation for an electronic device shown in fig. 12 comprises a general computer hardware structure including at least a processor 1201 and a memory 1202. The processor 1201 and the memory 1202 are connected by a bus 1203. The memory 1202 is adapted to store instructions or programs executable by the processor 1201. The processor 1201 may be a stand-alone microprocessor or a collection of one or more microprocessors. Thus, the processor 1201 implements processing of data and control of other devices by executing instructions stored by the memory 1202 to perform the method flows of embodiments of the present invention as described above. The bus 1203 connects the above components together, as well as connecting the above components to a display controller 1204 and a display device and input/output (I/O) device 1205. Input/output (I/O) devices 1205 may be a mouse, keyboard, modem, network interface, touch input device, motion-sensing input device, printer, and other devices known in the art. Typically, the input/output devices 1205 are connected to the system through an input/output (I/O) controller 1206.

Wherein the instructions stored by the memory 1202 are executable by the at least one processor 1201 to implement: acquiring a transaction instruction text in a dialog box of the instant messaging tool; processing the trading instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese pre-trained model, an intention classification model and an entity recognition model; and displaying the quotation on the interface of the dialog box of the instant messaging tool.

Specifically, the electronic device includes: one or more processors 1201 and a memory 1202, with one processor 1201 being illustrated in fig. 12. The processor 1201 and the memory 1202 may be connected by a bus or other means, and fig. 12 illustrates an example of the bus connection. Memory 1202, which is a non-volatile computer-readable storage medium, may be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The processor 1201 implements the above-described method of automatically generating a quotation by executing the non-volatile software programs, instructions, and modules stored in the memory 1202 to execute various functional applications and data processing of the device.

The memory 1202 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store a list of options, etc. Further, the memory 1202 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, memory 1202 may optionally include memory located remotely from processor 1201, which may be connected to an external device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

One or more modules are stored in the memory 1202, which when executed by the one or more processors 1201 perform the method of automatic generation of a quotation in any of the method embodiments described above.

The product can execute the method provided by the embodiment of the application, has corresponding functional modules and beneficial effects of the execution method, and can refer to the method provided by the embodiment of the application without detailed technical details in the embodiment.

Embodiments of the present invention relate to a non-transitory storage medium storing a computer-readable program for causing a computer to perform some or all of the above-described method embodiments.

That is, as can be understood by those skilled in the art, all or part of the steps in the method for implementing the embodiments described above may be implemented by a program instructing related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific to implementations of the invention, and that various changes in form and detail may be made therein without departing from the spirit and scope of the invention in practice.

Claims

1. A method for automatic generation of a quotation, the method comprising:

acquiring a transaction instruction text in a dialog box of the instant messaging tool;

processing the trading instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese pre-trained model, an intention classification model and an entity recognition model;

and displaying the quotation on an interface where a dialog box of the instant messaging tool is located.

2. The method of claim 1, further comprising:

3. The method of claim 1, further comprising:

sending the transaction instruction text in a dialog box of the instant messaging tool; alternatively, the first and second electrodes may be,

4. The method of claim 1, wherein said processing said trade instruction text through a pre-trained model and entity normalization generates a quotation, comprising;

5. The method of claim 1, wherein the chinese pre-training model is a FinBERT model, wherein the FinBERT model is a financial domain pre-training model.

6. The method of claim 1, wherein the entity normalization is used to represent different words with the same meaning in a unified manner.

7. The method of claim 1, further comprising:

8. An apparatus for automatic generation of a quotation, the apparatus comprising:

the acquisition unit is used for acquiring a transaction instruction text in a dialog box of the instant messaging tool;

the processing unit is used for processing the transaction instruction text through a pre-trained model and entity normalization to generate a quotation, wherein the pre-trained model comprises a Chinese and pre-trained model, an intention classification model and an entity recognition model;

and the display unit is used for displaying the quotation on the interface where the dialog box of the instant messaging tool is positioned.

9. A computer-readable storage medium on which computer program instructions are stored, which, when executed by a processor, implement the method of any one of claims 1-7.

10. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method of any of claims 1-7.