Business information query method and device, electronic equipment, storage medium and program product

Abstract

The invention discloses a business information query method and device, electronic equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring a service search text and a service information knowledge graph; classifying the service search text, and determining a text category of the service search text; when the text category of the service search text is a service query category, searching a corresponding target entity in each alternative entity in the service information knowledge graph according to the service search text; and according to the target entity, generating and feeding back a target answer corresponding to the business search text. According to the technical scheme provided by the embodiment of the invention, answers can be quickly and accurately retrieved for business personnel, information support can be provided for the business personnel to carry out cross-business fields, and efficient and accurate services can be provided for the business personnel.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a business information query method, device, electronic device, storage medium and program product. Background Art

[0002] At present, the business information corresponding to different business scenarios is very rich, which requires a high level of knowledge reserve for business personnel.

[0003] The dissemination of existing business information usually relies on business documents and business presentations, and it is difficult for business personnel to understand cross-business information. Therefore, in cross-business scenarios, it is difficult for business personnel to provide efficient and accurate services. Summary of the invention

[0004] The present invention provides a business information query method, device, electronic device, storage medium and program product, which can quickly and accurately retrieve answers for business personnel, provide information support for business personnel across business fields, and ensure that business personnel provide efficient and accurate services.

[0005] According to one aspect of the present invention, a service information query method is provided, the method comprising:

[0006] Obtain business search text and business information knowledge graph;

[0007] Classifying the business search text to determine the text category of the business search text;

[0008] When the text category of the business search text is a business query class, searching for a corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text;

[0009] According to the target entity, a target answer corresponding to the business search text is generated and fed back.

[0010] According to another aspect of the present invention, a service information query device is provided, the device comprising:

[0011] Business search text acquisition module, used to obtain business search text and business information knowledge graph;

[0012] A text category determination module, used to classify the business search text and determine the text category of the business search text;

[0013] A target entity search module, used to search for a corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text when the text category of the business search text is a business query class;

[0014] The target answer generation module is used to generate a target answer corresponding to the business search text according to the target entity.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the business information query method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the business information query method described in any embodiment of the present invention when executed.

[0020] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the service information query method according to any embodiment of the present invention is implemented.

[0021] The technical solution of the embodiment of the present invention obtains business search text and business information knowledge graph, classifies the business search text, determines the text category of the business search text, and when the text category of the business search text is a business query class, searches for the corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text, and generates and feeds back the target answer corresponding to the business search text according to the target entity. The business information knowledge graph is introduced. When the text category of the business search text is a business query class, the target answer corresponding to the business search text is generated based on the business information knowledge graph, which can quickly and accurately retrieve answers for business personnel, provide information support for business personnel across business fields, and ensure that business personnel provide efficient and accurate services.

[0022] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0024] Figure 1 is a flow chart of a business information query method provided according to Embodiment 1 of the present invention;

[0025] Figure 2 is a flow chart of a business information query method provided according to Embodiment 2 of the present invention;

[0026] Figure 3 is a schematic diagram of a business information query method provided according to Embodiment 2 of the present invention;

[0027] Figure 4 is a schematic diagram of the structure of a business information query device provided according to Embodiment 3 of the present invention;

[0028] Figure 5 It is a structural schematic diagram of an electronic device for implementing the service information query method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] Embodiment 1

[0032] Figure 1 A flowchart of a business information query method provided in Embodiment 1 of the present invention. The embodiment of the present invention can be applied to the case of querying business information based on a knowledge graph, and the method can be executed by a business information query device, which can be implemented in the form of hardware and / or software, and the business information query device can be configured in an electronic device that carries the business information query function, such as a client or a server.

[0033] See also Figure 1 The business information query method shown includes:

[0034] S110. Obtain business search text and business information knowledge graph.

[0035] The business search text may be a query text of business information input by a business person. Optionally, the device may have a corresponding front-end page input box to obtain the business search text input by the business person. The business information knowledge graph may be used to record business information in the form of a knowledge graph. Optionally, the business information knowledge graph may be pre-stored in the device or graph database. Optionally, the number of business information knowledge graphs may be at least one. The business information knowledge graph may correspond to a business document.

[0036] In an optional embodiment of the present invention, the business information knowledge graph is a financial business information knowledge graph; the candidate entities include business fields, product information, activity information, time information, regional information, business processing procedures, speech information and business experience.

[0037] The knowledge graph of financial business information can be used to record financial business information in the form of a knowledge graph. In comparison, financial business scenarios have richer financial business information and product information, and higher service quality requirements. Business fields can be used to characterize business scenarios to which financial business belongs. Exemplarily, business fields may include credit business and storage business, etc. Product information may be product information of a product in a certain business field. Activity information may be the situation of a certain product having an activity. For similar products in the same business field, products with activity information have a higher recommendation priority. Time information may be the time when an activity of a certain product is carried out. Regional information may be the area where an activity of a certain product is carried out. Business processing procedures may be the handling procedures of a certain business. Script information may be general script information (or preferred script information) for a certain business or a certain product. Business experience may be other business information other than business handling procedures. For example, business introductions or business experience posts.

[0038] This solution introduces an application scenario of financial business information query, and concretizes the alternative entities into business fields, product information, activity information, time information, regional information, business processing procedures, speech information and business experience. It takes into account the characteristics of richer business information products and higher service quality requirements in financial business scenarios, improves the service quality of financial information business information query, and improves the applicability of the technical solution of the embodiment of the present invention.

[0039] Specifically, the business search text entered by the business personnel and the pre-stored business information knowledge graph can be obtained.

[0040] S120: Classify the business search text to determine the text category of the business search text.

[0041] Text categories can be used to characterize the search intent of business search texts. Exemplarily, text categories may include business processing categories and business query categories. Among them, the business processing category can be used to characterize business search texts with the intention of handling business. Exemplarily, business search texts of the business processing category may include "upload treasury bond business report" or "add SMS marketing do not disturb" and the like. The business query category may be other business search intents except the business processing category. Exemplarily, business search texts of the business query category may include "amount", "report" or "what is the maximum amount of a certain product" and the like.

[0042] Optionally, before classifying the business search text and determining the text category of the business search text, the method further includes: matching the corresponding target historical question among the alternative historical questions of the alternative historical question-answer pair according to the business search text. Determine the target historical answer corresponding to the target historical question as the third target answer, and provide feedback to the business personnel. When there is no target historical question matching the business search text among the alternative historical questions, classify the business search text and determine the text category of the business search text.

[0043] Among them, the alternative historical question and answer pairs may be pre-stored historical question and answer pairs. The alternative historical question and answer pairs may be used to quickly generate corresponding answers. The alternative historical question and answer pairs may include alternative historical questions and alternative historical answers. Optionally, the alternative historical question and answer pairs may be historical questions and historical answers that have been evaluated and approved by technical personnel. Optionally, the alternative question and answer pairs may be pre-stored in a relational database. The target historical question may be an alternative historical question that matches the business search text. The target historical answer may be a historical answer corresponding to the target historical question. The third target answer may be a target answer generated in the form of a question and answer pair. When there is no target historical question that matches the business search text among the alternative historical questions, the business search text is classified to determine the text class of the business search text. It can be understood that the question and answer pair method can be preferentially selected to determine the target answer. When the question and answer pair method cannot determine the target answer, the scheme recorded in the embodiment of the present invention is used to determine the target answer.

[0044] Specifically, before classifying the business search text and determining the text category of the business search text, the business search text may be vectorized to obtain a business search vector. Each alternative historical question of the alternative historical question-answer pair may be vectorized to obtain each alternative historical question vector. Similarity calculation may be performed on the business search vector and each alternative historical question vector to obtain a first similarity between each alternative historical question vector and the business search vector. Each first similarity may be compared with a first preset similarity threshold, and when there is a first similarity greater than or equal to the first preset similarity threshold, the alternative historical question corresponding to the maximum value of the first similarity is determined as the target historical question matching the business search text. The target historical answer corresponding to the target historical question is determined as the third target answer, and feedback is given to the business personnel. When each first similarity is less than the first preset similarity threshold, it is determined that each alternative historical question does not have a target historical question that matches the business search text, and the target historical question is fed back to the business personnel. When each alternative historical question does not have a target historical question that matches the business search text, the business search text is classified to determine the text category of the business search text. Among them, the business search vector can be the vectorized result of the business search text. The alternative historical question vector can be the vectorized result of the alternative historical question. By vectorizing the text, it is convenient to quickly query the target historical question that matches the business search text. The first similarity can be used to characterize the degree of proximity between the alternative historical question vector and the business search vector. The first preset similarity threshold can be the lower limit value of the similarity between the alternative historical question vector and the business search vector. It can be understood that the first preset similarity threshold can be a threshold value for measuring whether the alternative historical question can be used as a target historical question that matches the business search text. The alternative historical question corresponding to the maximum value of the first similarity can be the alternative historical question that is closest to the business search text.

[0045] This solution first filters the target answer based on the alternative historical question and answer pairs. When there is no target historical question matching the business search text among the alternative historical questions, the technical solution of the embodiment of the present invention is used to query the business information, and the alternative historical question and answer pairs are introduced to realize a quick search for the corresponding target answer based on the business search text.

[0046] Optionally, after classifying the business search text and determining the text category of the business search text, it also includes: when the text category of the business search text is a business processing category, obtaining each alternative business processing flow; and matching and feeding back the target business processing flow in each alternative business processing flow based on the business search text.

[0047] Among them, the business handling category can be used to represent the business search text of the intention to handle the business. Exemplarily, the business search text of the business handling category may include "upload treasury bond business report" or "add SMS marketing do not disturb" and the like. The alternative business processing flow may be a known business processing flow that is predetermined and stored. The target business flow may be an alternative business handling flow matched based on the business search text.

[0048] Specifically, after classifying the business search text and determining the text category of the business search text, when the text category of the business search text is a business processing class, obtain each alternative business processing flow. The business search text can be vectorized to obtain a business search vector, and the alternative business processing flow identifier of each alternative business processing flow can be vectorized to obtain an alternative business processing flow vector. The second similarity between each alternative business processing flow vector and the business search vector can be measured. The second similarity can be compared with the second preset similarity threshold, and when there is a second similarity greater than or equal to the second preset similarity threshold, the alternative business processing flow corresponding to the maximum value of the second similarity is determined as the target business processing flow that matches the business search text, and the third target processing flow is fed back to the business personnel. When each second similarity is less than the second preset similarity threshold, it is determined that each alternative business processing flow does not have a target business processing flow that matches the business search text. When there is no target business process flow that matches the business search text in each alternative business process flow, a target business process work order is established according to the business search text, so that the operation and maintenance personnel can generate the target business process flow and feedback the target business process flow to the business personnel. Among them, the alternative business process flow vector can be the vectorization result of the alternative business process flow. By vectorizing the text, it is convenient to quickly query the target business process flow that matches the business search text. The second similarity can be used to characterize the degree of proximity between the alternative business process flow vector and the business search vector. The second preset similarity threshold can be the lower limit value of the similarity between the alternative business process flow vector and the business search vector. It can be understood that the second preset similarity threshold can be a threshold value for measuring that the alternative business process flow can be used as a target business process flow that matches the business search text. The alternative historical question corresponding to the maximum value of the second similarity can be the alternative business process flow that is closest to the business search text. The target business process work order can be used to prompt the operation and maintenance personnel to manually feedback the target business process flow corresponding to the business search text. When there is no target business process flow matching the business search text in each candidate business process flow, a target business process work order is established according to the business search text, which can improve the fault tolerance of the solution.

[0049] Optionally, when the text category of the business search text is a business processing category, the various alternative business processing flows in the business information knowledge graph can be matched according to the business search text to determine and feedback the target business processing flow.

[0050] When the text category of the business search text is business processing type, this solution introduces various alternative business processing processes, and matches and feeds back the target business processing process in each alternative business processing process based on the business search text, thereby achieving rapid feedback on the business processing process.

[0051] S130. When the text category of the business search text is a business query category, search for a corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text.

[0052] The candidate entity may be an entity in the business information knowledge graph. The business information knowledge graph has at least one candidate entity. Through the candidate entities and the candidate relationships between the candidate entities, the business information and the relevance between the business information may be intuitively expressed in the form of a graph. The target entity may be a candidate entity that matches the business search text. Optionally, the number of target entities may be at least one.

[0053] Specifically, when the text category of the business search text is a business query class, the business search text can be vectorized to obtain a business search vector. Each candidate entity can be vectorized to obtain a candidate entity vector. The similarity between each candidate entity vector and the business search vector can be calculated to obtain a third similarity. The third similarity between each candidate entity vector and the business search vector can be compared. The candidate entity corresponding to the maximum value of the third similarity can be determined as the target entity matching the business search text. Among them, the candidate entity vector can be the vectorization result of the candidate entity. By vectorizing the text, it is convenient to quickly query the target entity matching the business search text. The third similarity can be used to characterize the degree of proximity between the candidate entity vector and the business search vector. The candidate entity corresponding to the maximum value of the third similarity can be the candidate entity closest to the business search text.

[0054] S140: Generate and feed back a target answer corresponding to the business search text according to the target entity.

[0055] The target answer can be the feedback answer of the business search text generated based on the target entity. Compared with directly viewing the business document, the content is complicated and it is impossible to quickly filter the information required by the business personnel. The target answer corresponding to the business search text generated by the technical solution of the embodiment of the present invention has a higher degree of pertinence and matching with the business search text, and is more convenient for business personnel to view and use.

[0056] Specifically, natural language processing technology can be used to summarize the content of the target entity, obtain the target answer corresponding to the business search text, and feed back the target answer to the business personnel.

[0057] Optionally, after generating the target answer corresponding to the business search text, the target answer can be quality evaluated. After passing the evaluation, the business search text can be used as an alternative historical question and the target answer can be used as an alternative historical answer to generate and store alternative historical question-answer pairs.

[0058] The technical solution of the embodiment of the present invention obtains business search text and business information knowledge graph, classifies the business search text, determines the text category of the business search text, and when the text category of the business search text is a business query class, searches for the corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text, and generates and feeds back the target answer corresponding to the business search text according to the target entity. The business information knowledge graph is introduced. When the text category of the business search text is a business query class, the target answer corresponding to the business search text is generated based on the business information knowledge graph, which can quickly and accurately retrieve answers for business personnel, provide information support for business personnel across business fields, and ensure that business personnel provide efficient and accurate services.

[0059] In an optional embodiment of the present invention, before obtaining the business search text and the business information knowledge graph, it also includes: obtaining the business document and the original text identifier of the business document; performing named entity recognition and relationship extraction on the business document to obtain at least one alternative entity and the alternative relationship between each alternative entity; generating a business information knowledge graph corresponding to the business document based on each alternative entity and the alternative relationship between each alternative entity, and adding the original text identifier of the business document to the business information knowledge graph; generating and feeding back the target answer corresponding to the business search text, including: generating the target answer corresponding to the business search text, and feeding back the target answer corresponding to the business search text and the original text identifier of the business document.

[0060] A business document can be used to record business information. Optionally, the number of business documents can be at least one. Optionally, a single business document can be used to record product information of a product in a certain business field. A business document can also record product information of each product in a certain business field. The business document original text identifier can be used to identify the original location of the business document. Exemplarily, the business document original text identifier can be the business document original text URL or the business document original text ID (identity identifier).

[0061] Specifically, before obtaining the business search text and the business information knowledge graph, the business document and the original text identifier of the business document may be obtained. A named entity recognition algorithm may be used to perform named entity recognition on the business document to obtain at least one candidate entity. A relationship extraction algorithm may be used to perform relationship extraction on the business document to obtain alternative relationships between the candidate entities. A business information knowledge graph corresponding to the business document may be generated based on the alternative entities and the alternative relationships between the candidate entities, and the original text identifier of the business document may be added to the business information knowledge graph. After generating the target answer corresponding to the business search text, the original text identifier of the business document may be fed back at the same time as the target answer corresponding to the business search text is fed back.

[0062] This solution pre-generates a business information knowledge graph and binds the business information knowledge graph with the original text identifier of the business document. When feeding back the target answer, the original text identifier of the business document is fed back, which enriches the feedback content to the business personnel, makes it easier for the business personnel to select the content they need to view, and can further improve the service quality of the business personnel.

[0063] Embodiment 2

[0064] Figure 2 This is a flow chart of a business information query method provided in Embodiment 2 of the present invention. On the basis of the above-mentioned embodiments, the embodiment of the present invention specifies "searching for the corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text" as "performing sentence detection on the business search text to determine the sentence category of the business search text; when the sentence category of the business search text is a question sentence, searching for the corresponding at least one first target entity and the first target relationship between each target entity in each candidate entity in the business information knowledge graph according to the business search text; when the sentence category of the business search text is a declarative sentence, searching for the corresponding at least one second target entity in each candidate entity in the business information knowledge graph according to the business search text", and specifies "generating and feeding back the target answer corresponding to the business search text according to the target entity" as "generating and feeding back the first target answer corresponding to the business search text according to each first target entity and the first target relationship between each first target entity when the sentence category of the business search text is a question sentence; generating and feeding back the second target answer corresponding to the business search text according to each second target entity when the sentence category of the business search text is a declarative sentence", thereby improving the flexibility and accuracy of answer generation and further improving the service quality of business personnel. It should be noted that for the parts not described in detail in the embodiments of the present invention, reference may be made to the descriptions of other embodiments.

[0065] See also Figure 2 The business information query method shown includes:

[0066] S210. Obtain business search text and business information knowledge graph.

[0067] S220: Classify the business search text to determine the text category of the business search text.

[0068] S230: When the text category of the business search text is a business query category, sentence pattern detection is performed on the business search text to determine the sentence pattern category of the business search text.

[0069] Sentence categories can be used to characterize the query content structure of the business search text. Exemplarily, sentence categories can include question sentences and declarative sentences. The amount of information contained in the query content of different sentence categories may be different. For example, the business search text of the question sentence can be "What is the maximum amount of product B?", wherein the query content contained in the business search text of the question sentence includes product B and the maximum amount. For another example, the business search text of the declarative sentence can be "Product B", wherein the query content contained in the business search text of the declarative sentence is product B.

[0070] Specifically, when the text category of the business search text is a business query category, a natural language processing technology may be used to perform sentence category detection on the business search text to obtain the sentence category of the business search text.

[0071] S240. When the sentence type of the business search text is a question sentence, search for at least one corresponding first target entity and a first target relationship between each first target entity in each candidate entity in the business information knowledge graph according to the business search text.

[0072] The first target entity may be an alternative entity matched when the business search text is a question sentence. The first target relationship may be a relationship between the first target entities. For example, the first target entity may be business information and product information; the first target relationship may be related. For another example, the first target entity may be product information and activity information; the first target relationship may be related. For another example, the first target entity may be activity information and time information; the first target relationship may be start time or end time. For another example, the first target entity may be activity information and customer group information; the first target relationship may be oriented. For another example, the first target entity may be activity information and regional information; the first target relationship may be oriented. For another example, the first target entity may be product information and speech information; the first target relationship may be related. For another example, the first target entity may be product information and business processing flow; the first target relationship may be related. For another example, the first target entity may be product information and business experience; the first target relationship may be related.

[0073] Specifically, when the sentence category of the business search text is a question sentence, the business search text can be vectorized to obtain a business search vector. Each candidate entity can be vectorized to obtain a candidate entity vector. The similarity between each candidate entity vector and the business search vector can be calculated to obtain a fourth similarity. The fourth similarity between each candidate entity vector and the business search vector can be calculated. The fourth similarity can be compared with the fourth preset similarity threshold, and the candidate entity greater than or equal to the fourth preset similarity threshold can be determined as the first target entity matching the business search text. The relationship between each first target entity can be determined as the first target relationship between each first target entity. Among them, the fourth similarity can be used to characterize the degree of proximity between the candidate entity vector and the business search vector when the sentence category of the business search text is a question sentence. The fourth preset similarity threshold can be the lower limit value of the similarity between the alternative entity vector and the business search vector when the sentence category of the business search text is a question sentence. It can be understood that the fourth preset similarity threshold can be the threshold value for measuring whether the alternative business processing flow can be used as a target entity matching the business search text when the sentence category of the business search text is a question sentence.

[0074] Optionally, when the sentence type of the business search text is a question sentence type, keywords can be extracted from the business search text to obtain business search text keywords. The business search text keywords can be vectorized to obtain business search keyword vectors. Each candidate entity can be vectorized to obtain a candidate entity vector. The similarity between each candidate entity vector and the business search keyword vector can be calculated to obtain a fifth similarity. The fifth similarity between each candidate entity vector and the business search keyword vector can be calculated. The fifth similarity can be compared with the fifth preset similarity threshold, and the candidate entity greater than or equal to the fifth preset similarity threshold can be determined as the first target entity matching the business search text keyword. The relationship between each first target entity can be determined as the first target relationship between each first target entity. Among them, the business search keyword vector can be the vectorization result of the business search text keyword. By vectorizing the text, it is convenient to quickly query the target entity matching the business search text. The fifth similarity can be used to characterize the closeness between the candidate entity vector and the business search keyword vector. The fifth preset similarity threshold may be a lower limit of the similarity between the candidate entity vector and the business search keyword vector. It can be understood that the fifth preset similarity threshold may be a threshold value for measuring whether the candidate entity can be used as a target entity matching the business search text.

[0075] By introducing a process of extracting keywords from business search texts, performing entity matching with business search text keywords and candidate entities, and determining the first target entity, the efficiency and accuracy of entity matching can be further improved.

[0076] S250: Generate and feed back a first target answer corresponding to the business search text according to each first target entity and the first target relationship between the first target entities.

[0077] The first target answer may be a target answer generated based on the first target entity and the first target relationship between the first target entities. In comparison, the first target answer takes into account the first target entity and the first target relationship between the first target entities, and the content of the first target answer is richer.

[0078] Specifically, natural language processing technology can be used to summarize the first target entity and the first target relationship between each first target entity, obtain the first target answer corresponding to the business search text, and feed back the target answer to the business personnel.

[0079] In an optional embodiment of the present invention, when the sentence category of the business search text is a question sentence, a first target answer corresponding to the business search text is generated and fed back based on each first target entity and the first target relationship between each first target entity, including: when the sentence category of the business search text is a question sentence, obtaining a preset answer template; filling the first target relationship between each first target entity and each first target entity into the preset answer template, and generating and feeding back the first target answer corresponding to the business search text.

[0080] The preset answer template can be a template for generating a preset answer. The preset answer template can be used to quickly and standardizedly generate a target answer. The preset answer template can have a filling position for the first target entity and the first target relationship. The preset answer template can be pre-generated and stored in the device or database.

[0081] Specifically, when the sentence type of the business search text is a question sentence type, a pre-stored preset answer template can be obtained. According to the corresponding filling position in the preset answer template, each first target entity and the first target relationship between each first target entity can be filled into the preset answer template to generate and feedback the first target answer corresponding to the business search text.

[0082] This solution introduces a preset answer template to generate the first target answer for the business search text in the question sentence format, which can quickly generate the first target answer and improve the service efficiency of business personnel.

[0083] S260: When the sentence type of the business search text is a declarative sentence, search for at least one corresponding second target entity in each candidate entity in the business information knowledge graph according to the business search text.

[0084] The second target entity may be a candidate entity matched when the business search text is a declarative sentence.

[0085] Specifically, when the sentence type of the business search text is a declarative sentence type, the business search text can be vectorized to obtain a business search vector. Each candidate entity can be vectorized to obtain a candidate entity vector. The similarity between each candidate entity vector and the business search vector can be calculated to obtain a sixth similarity. The sixth similarity between each candidate entity vector and the business search vector can be calculated. The sixth similarity can be compared with the sixth preset similarity threshold, and the candidate entity greater than or equal to the sixth preset similarity threshold can be determined as the second target entity matching the business search text. Among them, the sixth similarity can be used to characterize the degree of proximity between the candidate entity vector and the business search vector when the sentence type of the business search text is a declarative sentence. The sixth preset similarity threshold can be the lower limit value of the similarity between the candidate entity vector and the business search vector when the sentence type of the business search text is a declarative sentence. It can be understood that the sixth preset similarity threshold can be a threshold value for measuring the target entity that the candidate business processing flow can be matched with the business search text when the sentence type of the business search text is a declarative sentence.

[0086] S270: Generate and feed back a second target answer corresponding to the business search text according to each second target entity.

[0087] The second target answer may be a target answer generated based on each second target entity. In comparison, the second target answer takes into account the content of each second target entity.

[0088] Specifically, natural language processing technology can be used to summarize the content of each second target entity, obtain the second target answer corresponding to the business search text, and feed back the target answer to the business personnel.

[0089] In an optional embodiment of the present invention, when the sentence category of the business search text is a declarative sentence, a second target answer corresponding to the business search text is generated and fed back according to each second target entity, including: when the sentence category of the business search text is a declarative sentence, a pre-trained answer generation model is used to summarize the content of each second target entity, and the second target answer corresponding to the business search text is generated and fed back.

[0090] The answer generation model can be used to summarize the content of each second target entity and generate a second target answer. The input data of the answer generation model is each second target entity, and the output result is the second target answer. Exemplarily, the answer generation model can be a large language model or a neural network model. Wherein, when the answer generation model is a neural network model, the training sample of the answer generation model includes at least one historical target entity and a corresponding historical target answer. The historical target entity can be a target entity searched when performing a business information query at a historical moment. The historical target answer can be a target answer obtained based on each historical target entity. The historical target answer can be generated after verification by a technician.

[0091] Specifically, when the sentence type of the business search text is a declarative sentence, each second target entity can be input into a pre-trained answer generation model, the content of each second target entity can be summarized, and the second target answer corresponding to the business search text can be generated and fed back.

[0092] This solution introduces an answer generation model to generate a second target answer for a business search text in a declarative sentence format, which can improve the generation efficiency and accuracy of the second target answer and further improve the service quality and efficiency of business personnel.

[0093] The technical solution of the embodiment of the present invention determines the sentence category of the business search text by performing sentence detection on the business search text. When the sentence category of the business search text is a question sentence, based on the business search text, each candidate entity in the business information knowledge graph is searched for at least one corresponding first target entity and a first target relationship between each first target entity. Based on each first target entity and the first target relationship between each first target entity, a first target answer corresponding to the business search text is generated and fed back. When the sentence category of the business search text is a declarative sentence, based on the business search text, each candidate entity in the business information knowledge graph is searched for at least one corresponding second target entity. Based on each second target entity, a second target answer corresponding to the business search text is generated and fed back. Sentence category detection of the business search text is introduced. For different sentence categories, target answers are generated based on different reference information, thereby improving the flexibility and accuracy of target answer generation and further improving the service quality of business personnel.

[0094] At present, there are a wide variety of financial products in different business lines, which require business personnel to have high business knowledge reserves. It is difficult to provide efficient and accurate services for cross-business scenarios. Knowledge dissemination relies on layers of lectures and documents, and lacks a platform for fast and accurate answer retrieval. When serving customers, it is too dependent on the experience and knowledge reserves of business personnel, and there may be loopholes in the collection, recording and updating of information, resulting in opaque and inaccurate information. Existing business information relies on business documents and lectures for dissemination, and lacks a unified integration platform. When business personnel encounter problems or get solutions, they cannot quickly share knowledge. The core of this solution is to build a business information knowledge base, and provide question-and-answer services based on the knowledge base for business personnel to query. Each business-related information is used as an alternative entity in the business information knowledge graph, and the alternative entity has multiple types of attributes and relationships. The maintainer of the business information knowledge graph can upload business documents (or product documents) for knowledge extraction by logging into the system, or directly maintain question-and-answer knowledge to expand the content of the knowledge base. Among them, the construction process of the business information knowledge graph is as follows: according to the entity category, the existing business document (or product document) corpus is annotated with entities and relationships, and the knowledge extraction model is trained. Named entity recognition algorithms and relationship extraction algorithms can be used to widely mine the attributes in various text materials, different business scenarios, and the relationship between products and various business scenarios, and record the original text ID in the business information knowledge graph for knowledge traceability. After review by business personnel, these extracted triples (i.e., alternative entities and alternative relationships between alternative entities) can be used to construct a business relationship knowledge graph and stored in the graph database. When business personnel search and ask questions, the business search text can be identified by text category, and the target entity can be found in the business information knowledge graph based on the semantic similarity between the business search text and the alternative entities in the business information knowledge graph. The preset answer template can be used to organize the generation of the target answer and mark the source of the answer for reference by business personnel.

[0095] Among them, the main task of the named entity recognition algorithm is to identify different types of entities from text data, which is a basic task for building a knowledge graph. Exemplarily, the named entity recognition algorithm can be a BiLSTM-CRF (bidirectional long short-term memory network-conditional random field) algorithm. By using a bidirectional long short-term memory network (BiLSTM) to extract sentence feature vectors, and then using a conditional random field (CRF) to learn the relationship between labels, the sequence labeling task is completed. The relationship extraction algorithm is an algorithm model used to determine the semantic relationship between two entities that appear simultaneously in a text. This solution can use a convolutional neural network (CNN) to extract important local features in a sentence, and obtain the classification results of character relationships through pooling and fully connected layers. When querying business information, business personnel can enter questions described in natural language (i.e., business search text), and through the steps of classifying questions, entity (or attribute) recognition, and semantic disambiguation, find the triple with the strongest relevance to the question from the business information knowledge graph, and answer according to the preset answer template. At the same time, for questions that are not accurately expressed by users, corresponding prompts or responses are given. Figure 3 FIG. 1 is a schematic diagram of a business information query method. Based on the above embodiment, Figure 3 This is a preferred embodiment of the present invention. Figure 3As shown, the business personnel enter the business search text in the system. First, based on the business search text, the target historical answer is searched in the alternative historical question and answer pairs (i.e., the answer data already exists) in the relational database to obtain the third target answer, and the third target answer is fed back to the business personnel. When there is no target historical question that matches the business search text in each alternative historical question (i.e., there is no direct answer data), the classifier is used to divide the business search text into two categories: business processing class (i.e., demand) and business query class (i.e., question). When the text category of the business search text is the business processing class, each alternative business processing flow is obtained. According to the business search text, in each alternative business processing flow, the target business processing flow is matched (i.e., the known business flow is matched), and the target business processing flow is fed back (i.e., the business processing flow is returned). When there is no target business processing flow that matches the business search text in each alternative business processing flow (i.e., the known business flow is not matched), a target business processing work order (i.e., a work order) is established according to the business search text, so that the operation and maintenance personnel can generate the target business processing flow and feed the target business processing flow back to the business personnel. When the text category of the business search text is a business query category, sentence detection is performed on the business search text to determine the sentence category of the business search text. Among them, the sentence category of the business search text includes question sentences (i.e., template sentence questions) and declarative sentences (i.e., fuzzy questions). When the sentence category of the business search text is a question sentence, according to the business search text, at least one corresponding first target entity and the first target relationship between each first target entity are searched in each candidate entity in the business information knowledge graph. Obtain a preset answer template. Fill the first target relationship between each first target entity and each first target entity into the preset answer template, generate and feedback the first target answer corresponding to the business search text (i.e., generated according to the template). When the sentence category of the business search text is a declarative sentence, according to the business search text, at least one corresponding second target entity is searched in each candidate entity in the business information knowledge graph. When the sentence category of the business search text is a declarative sentence, a pre-trained answer generation model is used to summarize the content of each second target entity, and the second target answer corresponding to the business search text is generated and feedback (i.e., large model text generation). For example, "upload treasury bond business report" or "add SMS marketing do not disturb" is a business processing type (ie demand type), which can return relevant information for handling the business based on the matched business processing flow and product information. If the known business processing flow in the system is not matched, the auxiliary business personnel will establish a process for transferring the work order to manual service. For business search texts of the business query type (ie question type), such as "What is the maximum limit of product A?". The keywords "product A" and "limit" can be extracted, and the "limit" type entities related to the "product A" product can be found in the business information knowledge graph, and the target answer is generated according to the preset answer template.If the business personnel input a single candidate entity such as "Product A", "Amount" or "Report", the relevant content will be retrieved from the business information knowledge graph and input into the answer generation model to write the answer. By constructing the business information knowledge graph, the present invention provides a platform for business personnel to quickly and accurately retrieve answers, which can provide strong support when facing unfamiliar business areas.

[0096] Embodiment 3

[0097] Figure 4 A schematic diagram of the structure of a business information query device provided in Embodiment 3 of the present invention. The embodiment of the present invention can be applied to the case where business information is queried based on a knowledge graph, the device can execute a business information query method, the device can be implemented in the form of hardware and / or software, and the device can be configured in an electronic device that carries a business information query function, such as a client or a server.

[0098] See also Figure 4 The business information query device shown includes: a business search text acquisition module 410, a text category determination module 420, a target entity search module 430 and a target answer generation module 440. Among them, the business search text acquisition module 410 is used to acquire the business search text and the business information knowledge graph; the text category determination module 420 is used to classify the business search text and determine the text category of the business search text; the target entity search module 430 is used to search for the corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text when the text category of the business search text is a business query class; the target answer generation module 440 is used to generate the target answer corresponding to the business search text according to the target entity.

[0099] The technical solution of the embodiment of the present invention obtains business search text and business information knowledge graph, classifies the business search text, determines the text category of the business search text, and when the text category of the business search text is a business query class, searches for the corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text, and generates and feeds back the target answer corresponding to the business search text according to the target entity. The business information knowledge graph is introduced. When the text category of the business search text is a business query class, the target answer corresponding to the business search text is generated based on the business information knowledge graph, which can quickly and accurately retrieve answers for business personnel, provide information support for business personnel across business fields, and ensure that business personnel provide efficient and accurate services.

[0100] In an optional embodiment of the present invention, the target entity search module 430 includes: a sentence category detection unit, which is used to perform sentence detection on the business search text to determine the sentence category of the business search text; a first target entity determination unit, which is used to search for at least one corresponding first target entity and a first target relationship between each first target entity in each candidate entity in the business information knowledge graph according to the business search text when the sentence category of the business search text is a question sentence; a second target entity determination unit, which is used to search for at least one corresponding second target entity in each candidate entity in the business information knowledge graph according to the business search text when the sentence category of the business search text is a declarative sentence; the target answer generation module 440 includes: a first target answer generation unit, which is used to generate and feedback a first target answer corresponding to the business search text according to each first target entity and the first target relationship between each first target entity when the sentence category of the business search text is a question sentence; and a second target answer generation unit, which is used to generate and feedback a second target answer corresponding to the business search text according to each second target entity when the sentence category of the business search text is a declarative sentence.

[0101] In an optional embodiment of the present invention, the first target answer generation unit includes: a preset answer template acquisition subunit, used to obtain a preset answer template when the sentence category of the business search text is a question sentence; a first target answer generation subunit, used to fill the first target relationship between each first target entity and each first target entity into the preset answer template, and generate and feedback the first target answer corresponding to the business search text.

[0102] In an optional embodiment of the present invention, the second target answer generation unit includes: a first target answer generation sub-unit, which is used to use a pre-trained answer generation model to summarize the content of each second target entity when the sentence category of the business search text is a declarative sentence, and generate and feedback the second target answer corresponding to the business search text; wherein the training sample of the answer generation model includes at least one historical target entity and a corresponding historical target answer.

[0103] In an optional embodiment of the present invention, the device also includes: a business document acquisition module, which is used to acquire business documents and business document original text identifiers before acquiring the business search text and the business information knowledge graph; an alternative entity recognition module, which is used to perform named entity recognition and relationship extraction on the business document to obtain at least one alternative entity and an alternative relationship between each of the alternative entities; a knowledge graph generation module, which is used to generate a business information knowledge graph corresponding to the business document based on each of the alternative entities and the alternative relationship between each of the alternative entities, and add the business document original text identifier to the business information knowledge graph; the target answer generation module 440 includes: a first target answer feedback unit, which is used to generate a target answer corresponding to the business search text, and to feedback the target answer corresponding to the business search text and the business document original text identifier.

[0104] In an optional embodiment of the present invention, the business information knowledge graph is a financial business information knowledge graph; the candidate entities include business fields, product information, activity information, time information, regional information, business processing procedures, speech information and business experience.

[0105] The business information query device provided in the embodiment of the present invention can execute the business information query method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0106] In the technical solution of the embodiment of the present invention, the collected information is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0107] Embodiment 4

[0108] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.

[0109] Figure 5A schematic diagram of the structure of an electronic device 500 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0110] like Figure 5 As shown, the electronic device 500 includes at least one processor 501, and a memory connected to the at least one processor 501 in communication, such as a read-only memory (ROM) 502, a random access memory (RAM) 503, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 501 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 502 or the computer program loaded from the storage unit 508 to the random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 can also be stored. The processor 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0111] Multiple components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0112] The processor 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 501 executes the various methods and processes described above, such as a business information query method.

[0113] In some embodiments, the business information query method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the processor 501, one or more steps of the business information query method described above may be performed. Alternatively, in other embodiments, the processor 501 may be configured to perform the business information query method in any other appropriate manner (e.g., by means of firmware).

[0114] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0116] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0118] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0119] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS (Virtual Private Server) services.

[0120] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0121] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A business information query method, characterized in that: The method comprises: Obtain business search text and business information knowledge graph; Classifying the business search text to determine the text category of the business search text; When the text category of the business search text is a business query class, searching for a corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text; According to the target entity, a target answer corresponding to the business search text is generated and fed back.

2. The method according to claim 1, characterized in that The step of searching for a corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text includes: Performing sentence detection on the business search text to determine the sentence category of the business search text; When the sentence type of the business search text is a question sentence type, searching, according to the business search text, for at least one corresponding first target entity and a first target relationship between each of the first target entities in the business information knowledge graph; When the sentence type of the business search text is a declarative sentence type, searching for at least one corresponding second target entity in each candidate entity in the business information knowledge graph according to the business search text; The generating and feeding back a target answer corresponding to the business search text according to the target entity includes: When the sentence type of the business search text is a question sentence type, generating and feeding back a first target answer corresponding to the business search text according to each of the first target entities and the first target relationship between each of the first target entities; When the sentence type of the business search text is a declarative sentence, a second target answer corresponding to the business search text is generated and fed back according to each of the second target entities.

3. The method according to claim 2, characterized in that When the sentence type of the business search text is a question sentence type, generating and feeding back a first target answer corresponding to the business search text according to each of the first target entities and the first target relationship between each of the first target entities includes: When the sentence type of the business search text is a question sentence type, obtaining a preset answer template; The first target entities and the first target relationships between the first target entities are filled into the preset answer template, and a first target answer corresponding to the business search text is generated and fed back.

4. The method according to claim 2, characterized in that: When the sentence type of the business search text is a declarative sentence, generating and feeding back a second target answer corresponding to the business search text according to each of the second target entities includes: When the sentence type of the business search text is a declarative sentence, a pre-trained answer generation model is used to summarize the content of each second target entity, and a second target answer corresponding to the business search text is generated and fed back.

5. The method according to claim 1, characterized in that Before obtaining the business search text and the business information knowledge graph, it also includes: Obtain business documents and original business document identifiers; Performing named entity recognition and relationship extraction on the business document to obtain at least one candidate entity and candidate relationships between the candidate entities; Generate a business information knowledge graph corresponding to the business document according to each of the candidate entities and the candidate relationships between the candidate entities, and add the original text identifier of the business document to the business information knowledge graph; The generating and feeding back a target answer corresponding to the business search text includes: Generate a target answer corresponding to the business search text, and feed back the target answer corresponding to the business search text and the original text identifier of the business document.

6. The method according to claim 1, characterized in that The business information knowledge graph is a financial business information knowledge graph; the candidate entities include business fields, product information, activity information, time information, regional information, business processing procedures, speech information and business experience.

7. A business information query device, characterized in that: The device comprises: Business search text acquisition module, used to obtain business search text and business information knowledge graph; A text category determination module, used to classify the business search text and determine the text category of the business search text; A target entity search module, used to search for a corresponding target entity in each candidate entity in the business information knowledge graph according to the business search text when the text category of the business search text is a business query class; The target answer generation module is used to generate a target answer corresponding to the business search text according to the target entity.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the business information query method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the business information query method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the service information query method according to any one of claims 1 to 6 is implemented.

Images