Information processing method and device

Abstract

The present invention discloses an information processing method and device. The method is applied to a dialogue system, and includes: first, receiving a sentence input by a user, and analyzing the intention of the sentence; then, obtaining a target slot to be filled with a response sentence according to the intention of the sentence; then, extracting information corresponding to the target slot from the received sentence to obtain known slot information; then, judging whether the known slot information corresponds to a certain response information according to a predetermined slot information hierarchy and the correspondence between slot information at each level and response information, if so, generating a second sentence for responding to the first sentence according to the response information without asking for unknown slot information in the target slot. In this way, when the known slot information corresponds to a certain response information, there is no need to perform a subsequent slot information filling process, avoiding unnecessary calculation processes, thereby greatly simplifying the interaction between man and machine, and providing a response sentence more quickly.

Description

Technical Field

[0001] The present invention relates to the field of computer information processing, and in particular to an information processing method and device. Background Art

[0002] In the dialogue system, after the intent is recognized based on the user input, the dialogue will be advanced step by step according to the intent and the preset process corresponding to the intent, and finally the answer to the user's problem will be found. In this process, the user needs to input relevant information and operate according to the preset process, and each input and operation of the user is actually selecting a strategy branch until the answer is obtained.

[0003] In existing solutions, for the same intent, the process is the same every time, and you have to go through the decision tree from beginning to end. But sometimes the user has already given some specific information when inputting, or when reaching a certain strategy branch, all subsequent branches correspond to the same answer. It would be cumbersome or too rigid to interact step by step according to the decision tree.

[0004] Therefore, how to flexibly simplify the steps in the preset process based on user input and get the corresponding answers with as little interaction as possible is a technical problem that has yet to be solved in the current dialogue system. Summary of the invention

[0005] In view of the above problems, embodiments of the present invention provide a method, an apparatus and a storage medium.

[0006] According to a first aspect of an embodiment of the present invention, an information processing method is applied to a dialogue system, the method comprising: obtaining a first sentence; determining the intent of the first sentence; determining a target slot to be filled with response information based on the intent; obtaining known slot information in the target slot from the first sentence; judging whether the known slot information corresponds to a determined response information based on a predetermined hierarchical relationship of the slot information and a correspondence between the slot information at each level and the response information, and if so, generating a second sentence for responding to the first sentence based on the response information without inquiring about the unknown slot information in the target slot.

[0007] According to an implementation mode of an embodiment of the present invention, the method also includes: judging whether the known slot information corresponds to a determined response information based on a predetermined hierarchical relationship of slot information and the correspondence between slot information at each level and response information; if not, further judging whether the known slot information corresponds to a finite number of determined response information; if so, generating a second statement for responding to the first statement based on the finite number of determined response information without inquiring about unknown slot information in the target slot; if not, further determining the slot to be filled, and generating a third statement for inquiring about the corresponding slot information based on the slot to be filled.

[0008] According to an implementation manner of an embodiment of the present invention, obtaining known slot information in a target slot from a first statement includes: obtaining a machine learning model corresponding to the target slot and used to identify slot information; and obtaining known slot information in the target slot from the first statement according to the machine learning model.

[0009] According to an implementation of an embodiment of the present invention, obtaining known slot information in a target slot from a first statement includes: obtaining relevant rules corresponding to the target slot for identifying slot information; and obtaining known slot information in the target slot from the first statement according to the relevant rules.

[0010] According to an embodiment of the present invention, based on a predetermined hierarchical relationship of slot information and the correspondence between slot information at each level and response information, before determining whether known slot information corresponds to a determined response information, the method further includes: obtaining a hierarchical relationship between known slot information in a target slot and unknown slot information in a target slot based on a pre-established first database, wherein the hierarchical relationship is stored in a tree storage structure, and the slot information stored in a leaf node in the tree storage structure is the final slot information; obtaining response information corresponding to the final slot information based on a pre-established second database; and determining the correspondence between slot information at each level and response information based on the hierarchical relationship and the response information corresponding to the final slot information.

[0011] According to an embodiment of the present invention, before obtaining the hierarchical relationship between known slot information in the target slot and unknown slot information in the target slot according to a pre-established first database, the method also includes: creating and maintaining the first database based on all slot information corresponding to the target slot and the hierarchical relationship between all slot information.

[0012] According to an implementation manner of an embodiment of the present invention, before obtaining the response information corresponding to the final slot information according to the pre-established second database, the method further includes: creating and maintaining the second database according to the final slot information and the response information corresponding to the final slot information.

[0013] According to an embodiment of the present invention, after determining the correspondence between slot information at each level and response information based on the hierarchical relationship and response information corresponding to the final slot information, the method further includes: after changes occur in the first database or the second database, making corresponding changes to the hierarchical relationship of the slot information and the correspondence between slot information at each level and response information.

[0014] According to an implementation manner of an embodiment of the present invention, based on a predetermined hierarchical relationship of slot information and a correspondence between slot information at each level and response information, it is determined whether known slot information corresponds to a determined response information, including: obtaining all response information corresponding to the known slot information to obtain a response information set based on a predetermined hierarchical relationship of slot information and a correspondence between slot information at each level and response information; and determining whether all response information in the information set are the same, if so, the known slot information corresponds to a determined response information.

[0015] According to a second aspect of an embodiment of the present invention, an information processing device is applied to a dialogue system, and the device includes: a first sentence acquisition module, used to acquire a first sentence; an intention determination module, used to determine the intention of the first sentence; a slot determination module, used to determine a target slot to be filled with response information according to the intention; a slot information acquisition module, used to acquire known slot information in the target slot from the first sentence; a second sentence generation module, used to determine whether the known slot information corresponds to a determined response information based on a predetermined hierarchical relationship of the slot information and a correspondence between the slot information at each level and the response information, and if so, generate a second sentence for responding to the first sentence based on the response information without inquiring about the unknown slot information in the target slot.

[0016] According to an implementation mode of an embodiment of the present invention, the second statement generation module is also used to determine whether the known slot information corresponds to a determined response information based on a predetermined slot information hierarchy relationship and the correspondence between each level of slot information and the response information; if not, further determine whether the known slot information corresponds to a finite number of determined response information; if so, generate a second statement for responding to the first statement based on the finite number of determined response information without inquiring about the unknown slot information in the target slot; if not, further determine the slot to be filled, and generate a third statement for inquiring about the corresponding slot information based on the slot to be filled.

[0017] According to an embodiment of the present invention, the slot information acquisition module includes: a machine learning model acquisition sub-module, which is used to establish a machine learning model for identifying slot information based on the target slot; and a slot information acquisition sub-module, which is used to obtain known slot information in the target slot from the first statement based on the machine learning model.

[0018] According to an embodiment of the present invention, the slot information acquisition module includes: a rule acquisition submodule, which is used to obtain relevant rules corresponding to the target slot and used to identify slot information; and obtain known slot information in the target slot from the first statement according to the relevant rules.

[0019] According to an embodiment of the present invention, the device also includes: a hierarchical relationship acquisition module, which is used to acquire the hierarchical relationship between known slot information in the target slot and unknown slot information in the target slot based on a pre-established first database, wherein the hierarchical relationship is stored in a tree storage structure, and the slot information stored in the leaf nodes in the tree storage structure is the final slot information; a response information acquisition module, which is used to acquire response information corresponding to the final slot information based on a pre-established second database; and a correspondence acquisition module, which is used to determine the correspondence between slot information at each level and response information based on the hierarchical relationship and the response information corresponding to the final slot information.

[0020] According to an implementation manner of an embodiment of the present invention, the device further includes: a first database maintenance module, configured to create and maintain a first database according to all slot information corresponding to the target slot and the hierarchical relationship between all slot information.

[0021] According to an implementation manner of an embodiment of the present invention, the device further includes: a second database maintenance module, configured to create and maintain a second database according to the final-stage slot information and the response information corresponding to the final-stage slot information.

[0022] According to an implementation manner of an embodiment of the present invention, the device further includes: a data synchronization module, which is used to make corresponding changes to the hierarchical relationship of slot information and the correspondence between slot information at each level and response information after changes occur in the first database or the second database.

[0023] According to an embodiment of the present invention, the second statement generation module includes: a response information set acquisition sub-module, which is used to obtain all response information corresponding to the known slot information according to a predetermined slot information hierarchy relationship and the correspondence between the slot information at each level and the response information to obtain a response information set; a response information comparison sub-module, which is used to determine whether all the response information in the information set is the same. If they are the same, the known slot information corresponds to a certain response information.

[0024] The embodiment of the present invention provides an information processing method and device, which includes: first, receiving a sentence input by a user, and analyzing the intention of the sentence; then, obtaining a target slot to be filled by a response sentence according to the intention of the sentence; then, extracting information corresponding to the target slot from the received sentence to obtain known slot information; then, judging whether the known slot information corresponds to a certain response information according to a predetermined slot information hierarchy and the correspondence between slot information at each level and response information, if so, generating a second sentence for responding to the first sentence according to the response information without asking for unknown slot information in the target slot. In this way, when the known slot information corresponds to a certain response information, there is no need to perform a subsequent slot information filling process, avoiding unnecessary calculation processes, thereby greatly simplifying the interaction between man and machine, and providing a response sentence more quickly.

[0025] It should be understood that the teachings of the present invention are not required to achieve all of the beneficial effects described above, but specific technical solutions can achieve specific technical effects, and other embodiments of the present invention can also achieve beneficial effects not mentioned above. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0027] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.

[0028] Figure 1 A schematic diagram of the implementation flow of the information processing method according to an embodiment of the present invention;

[0029] Figure 2 Schematic diagram of the intent analysis, slot information hierarchy relationship, and the corresponding relationship between slot information at each level and response information of the first embodiment of the present invention

[0030] Figure 3 This is a schematic diagram of a specific implementation flow of an application information processing method according to Embodiment 1 of the present invention;

[0031] Figure 4 The figure is a schematic diagram of the structure of an information processing device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0033] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0034] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0035] Figure 1 The implementation process of the information processing method according to the embodiment of the present invention is shown. Figure 1 The method is applied to a dialogue system, and includes: operation 110, obtaining a first sentence; operation 120, determining the intent of the first sentence; operation 130, determining a target slot to be filled with response information according to the intent; operation 140, obtaining known slot information in the target slot from the first sentence; operation 150, judging whether the known slot information corresponds to a determined response information according to a predetermined hierarchical relationship of the slot information and a correspondence between the slot information at each level and the response information, and if so, generating a second sentence for responding to the first sentence according to the response information without asking for unknown slot information in the target slot.

[0036] The information processing method of the embodiment of the present invention is mainly applied to a dialogue system, and is particularly suitable for a dialogue system that needs to obtain all required information through a slot filling process and finally give a response statement.

[0037] In operation 110, the first sentence mainly refers to the sentence input by the user in the human-computer dialogue, which may be a question sentence used to obtain certain information, such as "How big is my computer screen?"; it may also be a declarative sentence to initiate a conversation, such as "I want to buy a new computer"; or an operation instruction to instruct the robot to perform a certain operation, such as "Help me open the configuration options", etc.

[0038] In operation 120, the intent refers to a data representation that can represent the true meaning or purpose of a sentence and can be recognized by a computer, and any applicable data structure can be used. For example, the intent of "how big is my computer screen" can be represented by the string "screen size".

[0039] Usually, the dialogue system will execute different processing branches according to different intentions. The processing branch corresponding to each intention will gradually obtain various information required to achieve the intention in steps.

[0040] Figure 2 FIG. 1 shows a schematic diagram of the intent analysis, slot information hierarchy relationship, and the corresponding relationship between slot information at each level and response information of an application in an embodiment of the present invention. Figure 2As shown, each intent (e.g., intent 201, intent 202) corresponds to a different processing branch, and the slot information to be obtained on these processing branches (e.g., slot information 2011, slot information 20111, slot information 201112, etc.) and the final corresponding response statement (e.g., answer A, answer B, answer C, etc.) form a tree structure. The process of obtaining the response statement often starts from the intent node, determines the value of each level of slot information node one by one, and enters the corresponding next level branch until reaching the leaf node to obtain the answer.

[0041] For example, if the intention identified from the received sentence 20 is intention 201, the first-level slot information is determined first, assuming it is slot information 2011; then the second-level slot information is determined, assuming it is slot information 20111; then the third-level slot information is determined, assuming it is slot information 201112; at this time, the corresponding response sentence can be obtained as "Answer B".

[0042] In operation 130, the target slot to be filled in the response information is obtained, that is, the slot information to be obtained in the process of obtaining the final response statement. Assuming that the received sentence is "How big is my computer screen?", after semantic analysis, it is obtained that the intention of the sentence is "screen size". According to the pre-set processing procedure, if you want to obtain the screen size of a certain computer, you need to obtain the first-level slot information "product series", the second-level slot information "product model" and the third-level slot information "device number", among which "product series", "product model" and "device number" are the target slots to be obtained for the response information corresponding to the intention "screen size".

[0043] In operation 140, obtaining the known slot information in the target slot mainly refers to extracting the slot information corresponding to the target slot from the first sentence. The target slot from which the corresponding information can be extracted is a known slot, and the extracted corresponding information is the known slot information in the target slot. Assuming that the received sentence is "My computer is a computer with product series slot information 2011, how big is my computer screen", the slot information corresponding to the target slot "product series" can be extracted as "slot information 2011", where "slot information 2011" is the known slot information in the target slot.

[0044] In operation 150, the slot information hierarchical relationship mainly refers to a tree data structure used to describe the hierarchical relationship between slot information, such as inclusion, dependency, and causality. Figure 2 As shown, the next level of the first-level slot information 2012 is provided with a second-level slot information, namely slot information 20121; the next level of the second-level slot information 20121 is provided with two third-level slot information: slot information 201211 and slot information 201212. The next-level slot information may be the next-level classification of the current-level slot information, possible values, different results generated under different conditions, etc.

[0045] The response information mainly refers to the final answer or conclusion obtained to achieve the intention. For example, the response information corresponding to the intention "screen size" is the specific value of the screen size, such as "55 inches".

[0046] The correspondence between slot information at each level and response information mainly refers to the correspondence between slot information at each level and the response information ultimately pointed to, that is, the response information that may be determined by each slot information of each level of slot information.

[0047] As in Figure 2 In the correspondence between the slot information at each level and the response information shown, the response information corresponding to slot information 2011 includes answer A, answer B and answer C; the response information corresponding to slot information 2012 includes answer D; the response information corresponding to slot information 20111 includes answer A and answer B; the response information corresponding to slot information 201111 includes answer A.

[0048] When the known slot information corresponds to a certain response information, such as Figure 2 The slot information 2012 and slot information 201111 shown can give a second statement in response to the first statement based on the determined response information without asking for other slot information. For example, when the known slot information is slot information 2012, a response statement containing the response information "Answer D" can be given directly without reproducing the query statement to obtain the secondary slot information (e.g., slot information 20121) and the tertiary slot information (e.g., slot information 201211, slot information 201212).

[0049] It can be seen that when the known slot information corresponds to a certain response information, there is no need to perform the subsequent slot information filling process, which avoids unnecessary calculation processes, thereby greatly simplifying the interaction between humans and machines and giving response statements more quickly.

[0050] According to an implementation mode of an embodiment of the present invention, the method also includes: judging whether the known slot information corresponds to a determined response information based on a predetermined hierarchical relationship of slot information and the correspondence between slot information at each level and response information; if not, further judging whether the known slot information corresponds to a finite number of determined response information; if so, generating a second statement for responding to the first statement based on the finite number of determined response information without inquiring about unknown slot information in the target slot; if not, further determining the slot to be filled, and generating a third statement for inquiring about the corresponding slot information based on the slot to be filled.

[0051] Sometimes, although the known slot information does not correspond to a certain response information, it may only correspond to a limited number of certain response information. At this time, the response statement can be given through these limited number of certain response information to reduce the interaction process with the user.

[0052] In this embodiment, when it is determined that the known slot information does not correspond to a certain response information, it is further determined whether the known slot information corresponds to a finite number of certain response information. If so, the second statement is directly generated according to the finite number of certain response information.

[0053] For example, in Figure 2 In the correspondence between the slot information at each level and the response information shown, the response information corresponding to the first-level slot information 2011 includes answers A, B and C. After determining that the corresponding response information is limited, a third sentence containing all the answers can be generated to list all three response information, for example, "If your product model is slot information 20111 and the device number is slot information 201111, the screen size is the answer A; if your product model is slot information 20111 and the device number is slot information 201112, the screen size is the answer B; if your product model is slot information 20112 and the device number is slot information 201121, the screen size is the answer C." In this way, there is no need to ask for the second-level slot information or the third-level slot information again, and a second sentence for answering the first sentence can be generated, thereby greatly simplifying the interaction process with the user.

[0054] According to an implementation manner of an embodiment of the present invention, obtaining known slot information in a target slot from a first statement includes: obtaining a machine learning model corresponding to the target slot and used to identify slot information; and obtaining known slot information in the target slot from the first statement according to the machine learning model.

[0055] When obtaining known slot information in the target slot, information corresponding to the target slot can usually be found through semantic comparison. Therefore, in this embodiment, the known slot information in the target slot is obtained from the received first sentence more accurately and quickly by mainly using a machine learning model corresponding to the target slot and training with a large amount of corpus.

[0056] According to an implementation of an embodiment of the present invention, obtaining known slot information in a target slot from a first statement includes: obtaining relevant rules corresponding to the target slot for identifying slot information; and obtaining known slot information in the target slot from the first statement according to the relevant rules.

[0057] In some scenarios of actual applications, such as task-driven scenarios such as ordering food or buying airline tickets, the received sentence patterns are relatively fixed and do not change much. In this case, relevant rules corresponding to the target slots, such as regular expressions, can be pre-defined for identifying slot information; then, these rules can be used to obtain the information corresponding to the target slots, for example, by matching strings using regular expressions to determine the information corresponding to the target slots.

[0058] According to an embodiment of the present invention, based on a predetermined hierarchical relationship of slot information and the correspondence between slot information at each level and response information, before determining whether known slot information corresponds to a determined response information, the method further includes: obtaining a hierarchical relationship between known slot information in a target slot and unknown slot information in a target slot based on a pre-established first database, wherein the hierarchical relationship is stored in a tree storage structure, and the slot information stored in a leaf node in the tree storage structure is the final slot information; obtaining response information corresponding to the final slot information based on a pre-established second database; and determining the correspondence between slot information at each level and response information based on the hierarchical relationship and the response information corresponding to the final slot information.

[0059] It should be noted that the database mentioned here does not only refer to the database of a certain brand of Xiayi, but also refers to data storage media in a broad sense. The first database is mainly used to store information that can determine the hierarchical relationship between slot information, such as commodity classification information, product model list, business processing logic, etc. Usually, the response information often corresponds to the final slot information, and the second database is a database used to store the correspondence between the final slot information and the response information, such as a device configuration information table, etc.

[0060] In this embodiment, the hierarchical relationship between known slot information in the target slot and unknown slot information in the target slot is obtained from a pre-established first database, and the response information corresponding to the final slot information is obtained from a pre-established second database. The hierarchical relationship of the slot information and the correspondence between the slot information at each level and the response information can be determined, and thereby it can be determined whether the known slot information corresponds to a certain response information.

[0061] According to an embodiment of the present invention, before obtaining the hierarchical relationship between known slot information in the target slot and unknown slot information in the target slot according to a pre-established first database, the method also includes: creating and maintaining the first database based on all slot information corresponding to the target slot and the hierarchical relationship between all slot information.

[0062] In this embodiment, the creation and maintenance of the first database is conducive to conveniently and synchronously modifying the hierarchical relationship between slot information according to changes in information such as product information, commodity classification information or business logic.

[0063] According to an implementation manner of an embodiment of the present invention, before obtaining the response information corresponding to the final slot information according to the pre-established second database, the method further includes: creating and maintaining the second database according to the final slot information and the response information corresponding to the final slot information.

[0064] In this embodiment, the creation and maintenance of the second database is conducive to conveniently and synchronously modifying the final slot information and the response information corresponding to the final slot information according to changes in the device information configuration table and other information.

[0065] According to an embodiment of the present invention, after determining the correspondence between slot information at each level and response information based on the hierarchical relationship and response information corresponding to the final slot information, the method further includes: after changes occur in the first database or the second database, making corresponding changes to the hierarchical relationship of the slot information and the correspondence between slot information at each level and response information.

[0066] When according to an implementation manner of an embodiment of the present invention, based on a predetermined hierarchical relationship of slot information and a correspondence between slot information at each level and response information, it is determined whether known slot information corresponds to a determined response information, including: obtaining all response information corresponding to the known slot information to obtain a response information set according to a predetermined hierarchical relationship of slot information and a correspondence between slot information at each level and response information; and determining whether all response information in the information set are the same, if so, the known slot information corresponds to a determined response information.

[0067] Assuming that the known slot information is "slot information 2012", then in the following example Figure 2 In the slot information hierarchy relationship and the correspondence between slot information at each level and response information shown, the set of all response information corresponding to slot information 2012 is {"Answer D", "Answer D"}. At this time, it can be determined that the known slot information corresponds to a certain response information.

[0068] Figure 3 The specific implementation process of the application information processing method of the first embodiment of the present invention is shown, including:

[0069] Step 3010, receiving a statement input by a user;

[0070] Step 3020, identifying user intent;

[0071] Step 3030, analyzing the slot information required to achieve the user's intention;

[0072] Step 3040, according to the slot information required to achieve the user's intention, obtain the required slot information from the sentence input by the user;

[0073] Step 3050, judging whether the slot information corresponds to the determined response information according to the predetermined slot information hierarchy relationship and the correspondence relationship between the slot information at each level and the response information, if yes, proceeding to step 3060, if no, proceeding to step 3070;

[0074] Step 3060, generating a response statement containing the answer;

[0075] Step 3070, generate a response statement to inquire for more slot information.

[0076] It should be noted that the above is only an application example of the information processing method of the embodiment of the present invention, and is not a limitation on the application scenario and implementation method of the embodiment of the present invention. The implementer can adopt any applicable implementation method according to the implementation conditions and apply it to any applicable application scenario.

[0077] Furthermore, an embodiment of the present invention also provides an information processing device, which is applied to a dialogue system, such as Figure 4 As shown, the device 40 includes: a first sentence acquisition module 401, used to acquire the first sentence; an intention determination module 402, used to determine the intention of the first sentence; a slot determination module 403, used to determine the target slot to be filled in the response information according to the intention; a slot information acquisition module 404, used to acquire known slot information in the target slot from the first sentence; a second sentence generation module 405, used to determine whether the known slot information corresponds to a determined response information based on a predetermined hierarchical relationship of the slot information and the correspondence between the slot information at each level and the response information, and if so, generate a second sentence for responding to the first sentence based on the response information without inquiring about the unknown slot information in the target slot.

[0078] According to an implementation mode of an embodiment of the present invention, the second statement generation module 405 is also used to determine whether the known slot information corresponds to a determined response information based on a predetermined slot information hierarchy relationship and the correspondence between each level of slot information and the response information; if not, further determine whether the known slot information corresponds to a finite number of determined response information; if so, generate a second statement for responding to the first statement based on the finite number of determined response information without inquiring about the unknown slot information in the target slot; if not, further determine the slot to be filled, and generate a third statement for inquiring about the corresponding slot information based on the slot to be filled.

[0079] According to an embodiment of the present invention, the slot information acquisition module 404 includes: a machine learning model acquisition submodule, which is used to establish a machine learning model for identifying slot information based on the target slot; and a slot information acquisition submodule, which is used to obtain known slot information in the target slot from the first statement based on the machine learning model.

[0080] According to an implementation of an embodiment of the present invention, the slot information acquisition module 404 includes: a rule acquisition submodule, which is used to obtain relevant rules corresponding to the target slot and used to identify slot information; and obtain known slot information in the target slot from the first statement according to the relevant rules.

[0081] According to an embodiment of the present invention, the device 40 also includes: a hierarchical relationship acquisition module, which is used to acquire the hierarchical relationship between known slot information in the target slot and unknown slot information in the target slot based on a pre-established first database, wherein the hierarchical relationship is stored in a tree storage structure, and the slot information stored in the leaf nodes in the tree storage structure is the final slot information; a response information acquisition module, which is used to acquire the response information corresponding to the final slot information based on a pre-established second database; and a correspondence acquisition module, which is used to determine the correspondence between the slot information at each level and the response information based on the hierarchical relationship and the response information corresponding to the final slot information.

[0082] According to an implementation manner of an embodiment of the present invention, the device 40 further includes: a first database maintenance module, which is used to create and maintain a first database according to all slot information corresponding to the target slot and the hierarchical relationship between all slot information.

[0083] According to an implementation of an embodiment of the present invention, the device 40 further includes: a second database maintenance module, configured to create and maintain a second database according to the final slot information and the response information corresponding to the final slot information.

[0084] According to an implementation of an embodiment of the present invention, the device 40 further includes: a data synchronization module, which is used to make corresponding changes to the hierarchical relationship of slot information and the correspondence between slot information at each level and response information after changes occur in the first database or the second database.

[0085] According to an implementation mode of an embodiment of the present invention, the second statement generation module 405 includes: a response information set acquisition submodule, which is used to obtain all response information corresponding to the known slot information according to the predetermined slot information hierarchy relationship and the correspondence between the slot information at each level and the response information to obtain a response information set; a response information comparison submodule, which is used to determine whether all the response information in the information set is the same. If they are the same, the known slot information corresponds to a determined response information.

[0086] It should be noted here that the above description of the information processing device embodiment is similar to the description of the aforementioned method embodiment, and has similar beneficial effects as the aforementioned method embodiment, so it will not be repeated. For the technical details not yet disclosed in the description of the information processing device embodiment of the present invention, please refer to the description of the aforementioned method embodiment of the present invention for understanding, and in order to save space, it will not be repeated.

[0087] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0088] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another device, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0089] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0090] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0091] A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage medium, read-only memory (ROM), disk or optical disk, etc. Various media that can store program codes.

[0092] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage media, ROM, magnetic disks or optical disks.

[0093] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An information processing method, applied to a dialogue system, comprising: Get the first statement; determining the intent of the first statement; Determine a target slot to be filled according to the intent to obtain response information; Acquire known slot information in the target slot from the first statement; According to the predetermined slot information hierarchy relationship and the correspondence between the slot information at each level and the response information, determine whether the one or more response information corresponding to the known slot information through one or more correspondences is a determined response information. If so, generate a second statement for responding to the first statement based on the response information without inquiring about the unknown slot information in the target slot.

2. The method according to claim 1, further comprising: According to the predetermined hierarchical relationship of slot information and the correspondence between slot information at each level and response information, determine whether the known slot information corresponds to a determined response information; if not, further determine whether the known slot information corresponds to a finite number of determined response information; if so, generate a second statement for answering the first statement based on the finite number of determined response information without inquiring about the unknown slot information in the target slot; if not, further determine the slot to be filled, and generate a third statement for inquiring about the corresponding slot information based on the slot to be filled.

3. According to the method of claim 1, the step of obtaining known slot information in the target slot from the first statement comprises: Obtaining a machine learning model corresponding to the target slot and used to identify slot information; Acquire known slot information in the target slot from the first statement according to the machine learning model.

4. The method according to claim 1, wherein obtaining known slot information in the target slot from the first statement comprises: Obtain relevant rules corresponding to the target slot and used to identify slot information; The known slot information in the target slot is obtained from the first statement according to the relevant rule.

5. The method according to claim 1, before determining whether the known slot information corresponds to a determined response information based on the predetermined slot information hierarchy relationship and the correspondence relationship between the slot information at each level and the response information, the method further comprises: Acquire a hierarchical relationship between known slot information in the target slot and unknown slot information in the target slot according to a pre-established first database, wherein the hierarchical relationship is stored in a tree storage structure, and the slot information stored in a leaf node in the tree storage structure is the final slot information; Acquire response information corresponding to the final slot information according to a pre-established second database; The corresponding relationship between the slot information at each level and the response information is determined according to the hierarchical relationship and the response information corresponding to the final slot information.

6. The method according to claim 5, before acquiring the hierarchical relationship between the known slot information in the target slot and the unknown slot information in the target slot according to the pre-established first database, the method further comprises: The first database is created and maintained according to all slot information corresponding to the target slot and the hierarchical relationship between all slot information.

7. The method according to claim 5, before acquiring the response information corresponding to the final slot information according to the pre-established second database, the method further comprises: The second database is created and maintained according to the final slot information and the response information corresponding to the final slot information.

8. The method according to claim 5, after determining the correspondence between slot information at each level and response information according to the hierarchical relationship and the response information corresponding to the final slot information, the method further comprises: After the first database or the second database is changed, the hierarchical relationship of the slot information and the correspondence between the slot information at each level and the response information are changed accordingly.

9. The method according to claim 1, wherein judging whether the known slot information corresponds to a determined response information according to a predetermined slot information hierarchy relationship and a correspondence relationship between slot information at each level and response information comprises: Acquire all the response information corresponding to the known slot information according to the predetermined slot information hierarchy relationship and the correspondence relationship between the slot information at each level and the response information to obtain a response information set; It is determined whether all the response information in the response information set are the same. If they are the same, the known slot information corresponds to a certain response information.

10. An information processing device, applied to a dialogue system, comprising: A first sentence acquisition module, used for acquiring a first sentence; an intention determination module, used to determine the intention of the first sentence; A slot determination module, used to determine a target slot to be filled according to the intent to obtain the response information; A slot information acquisition module, used to acquire known slot information in the target slot from the first statement; The second statement generation module is used to determine whether one or more response information corresponding to the known slot information through one or more correspondences is a determined response information based on the predetermined slot information hierarchy relationship and the correspondence between the slot information at each level and the response information. If so, a second statement for responding to the first statement is generated based on the response information without inquiring about the unknown slot information in the target slot.

Images