CN113157939A - Information processing method and device - Google Patents
Information processing method and device Download PDFInfo
- Publication number
- CN113157939A CN113157939A CN202110324895.4A CN202110324895A CN113157939A CN 113157939 A CN113157939 A CN 113157939A CN 202110324895 A CN202110324895 A CN 202110324895A CN 113157939 A CN113157939 A CN 113157939A
- Authority
- CN
- China
- Prior art keywords
- slot
- knowledge
- entity
- graph data
- path
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 30
- 238000003672 processing method Methods 0.000 title claims abstract description 23
- 238000000034 method Methods 0.000 claims abstract description 44
- 230000008569 process Effects 0.000 abstract description 16
- 239000000047 product Substances 0.000 description 20
- 230000006870 function Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000013507 mapping Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Machine Translation (AREA)
Abstract
The application discloses an information processing method and device. When the method is used for configuring the intention, whether the slot position configuration performed by the administrator meets the set conditions or not can be judged according to the knowledge graph corresponding to the intention, and prompt information is given to help the administrator to correct the slot position configuration when the slot position configuration does not meet the set conditions. By the method, the intention and the slot position can be managed finely, the situations of configuration errors, configuration information omission or redundant slot position configuration and the like are reduced, and various problems caused by unreasonable slot position configuration in the man-machine conversation process can be reduced.
Description
Technical Field
The present application relates to the field of computer information processing, and in particular, to an information processing method and apparatus.
Background
At present, in a dialog system, especially a dialog system for business processing, a corresponding answer set is usually found according to a conversation intention determined by a user question, and then an accurate answer for a certain question is filtered out by a conversation slot.
The configuration of the session slot value in the above-mentioned session system is particularly important, and if the configuration of the session slot in the southwest and west is omitted or wrong, the problem that a matching answer cannot be found or an answer is omitted can be caused.
Furthermore, if the user question is generic to a session slot value, then continuing to extract each session slot value to query results in over-querying and a poor user experience.
Disclosure of Invention
The applicant creatively provides an information processing method and apparatus.
According to a first aspect of embodiments of the present application, there is provided an information processing method, including: acquiring a first intention and at least one slot position configured according to the first intention; determining a path between the first intent and the at least one slot based on the knowledge-graph data corresponding to the first intent; and determining whether the configuration of at least one slot position meets the set condition or not according to the knowledge graph data and the path, and if not, outputting prompt information.
According to an embodiment of the present application, determining a path between the first intent and the at least one slot from the knowledge-graph data corresponding to the first intent comprises: performing entity identification on the first intention to obtain at least one first entity; according to the knowledge graph data corresponding to the first intention, performing entity linkage on at least one first entity to obtain at least one second entity; according to the knowledge graph data corresponding to the first intention, performing entity link on at least one slot position to obtain at least one third entity; a path between the at least one second entity and the at least one third entity is determined from the knowledge-graph data corresponding to the first intent.
According to an embodiment of the present application, the entity identifying the first intention to obtain at least one first entity includes: and performing entity identification on the session intention based on the named entity identification model to obtain at least one first entity.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: determining the category of at least one slot according to the knowledge graph data; acquiring all entities under the category to obtain a fourth entity; and determining whether paths exist between all the fourth entities and the first intention, if so, the configuration of at least one slot position does not meet the set condition.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: and determining whether the path meets the association mode of the first intention and the at least one slot position, if not, the configuration of the at least one slot position does not meet the set condition.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: and determining whether a fifth entity which is matched with the at least one slot exists according to the knowledge graph data and the path, and if so, outputting the fifth entity.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: determining whether a reachable path exists between the first intent and the at least one slot based on the knowledge-graph data and the path.
According to an embodiment of the present application, prior to determining the path between the first intent and the at least one slot from the knowledge-graph data corresponding to the first intent, the method further comprises: knowledge-graph data corresponding to the first intent is established.
According to an embodiment of the application, establishing knowledge-graph data corresponding to the first intent includes: establishing a knowledge map database of a business field to which the first intention belongs; and acquiring knowledge graph data corresponding to the first intention according to the knowledge graph database.
According to a second aspect of embodiments of the present application, there is provided an information processing apparatus including: the slot position configuration information acquisition module is used for acquiring a first intention and at least one slot position configured according to the first intention; a knowledge graph path determination module for determining a path between the first intent and the at least one slot based on knowledge graph data corresponding to the first intent; and the slot position configuration prompting module is used for determining whether the configuration of at least one slot position meets the set condition or not according to the knowledge graph data and the path, and outputting prompting information if the configuration of at least one slot position does not meet the set condition.
The embodiment of the application provides an information processing method and device, and when an intention is configured, the method can judge whether slot configuration performed by an administrator meets set conditions according to a knowledge graph corresponding to the intention, and when the slot configuration does not meet the set conditions, prompt information is given to help the administrator to correct the slot configuration. By the method, the intention and the slot position can be managed finely, the situations of configuration errors, configuration information omission or redundant slot position configuration and the like are reduced, and various problems caused by unreasonable slot position configuration in the man-machine conversation process can be reduced.
It is to be understood that the implementation of the present application does not require all of the above-described advantages to be achieved, but rather that certain technical solutions may achieve certain technical effects, and that other embodiments of the present application may also achieve other advantages not mentioned above.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a schematic diagram illustrating an implementation flow of an embodiment of an information processing method according to the present application;
FIG. 2 is a schematic view of a session flow of an intelligent customer service system to which another embodiment of the information processing method of the present application is applied;
FIG. 3 is a schematic view of an implementation flow of another embodiment of the information processing method of the present application;
fig. 4 is a specific example of another embodiment of the information processing method of the present application;
fig. 5 shows a session intention as a composition structure of an embodiment of the information processing apparatus according to the present application.
Detailed Description
In order to make the objects, features and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
Fig. 1 shows an implementation flow of an embodiment of the information processing method of the present application. Referring to fig. 1, the method includes: an operation 110 of obtaining a first intent and at least one slot configured according to the first intent; an operation 120 of determining a path between the first intent and the at least one slot based on the knowledge-graph data corresponding to the first intent; operation 130 determines whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path, and if not, outputs a prompt message.
Generally, in a task-based dialog system such as smart customer service, the input and output of each task are enumerable, each slot of each task is input, and after a given slot value is output, various answers or processing branches are determined according to the slot value. When some slot values are ambiguous, the system may ask the user back to obtain the slot values. The process of the conversation is essentially a process of filling the slot position, namely understanding the user intention according to the user input, filling the slot position value contained in the user input into the slot position corresponding to the intention, and entering the next link of the conversation according to the slot position value. Wherein the user intent and corresponding slot position are pre-defined based on domain knowledge and expert experience.
For example, the user inputs: "how to use NFC? ", what the user expresses at this time is to use NFC, which is an intention. Some models may not have NFC functionality, and different models may have different instructions for using NFC. Thus, a "model" may be configured as a slot with the intent of "using NFC".
In operation 110, the first intention is that the slot needs to be configured, and one intention may correspond to one slot or a plurality of slots. For example, the intent "ticket" may correspond to a plurality of slots such as "date", "departure place", "destination", etc. The configuration slot position mainly comprises a setting slot position, a possible value corresponding to the configuration slot position and the like.
In operation 120, the knowledge-graph data corresponding to the first intention mainly refers to the knowledge-graph data constructed according to the domain knowledge and the expert opinions of the business domain where the first intention is located. Typically, the first intent, the slot set for the first intent, and the corresponding slot value correspond to an entity node in the knowledge-graph data, respectively. And determining a path between the first intention and the at least one slot position, namely mapping the first intention and the slot position to entity nodes in the knowledge-graph data, and calculating the path between corresponding nodes through the path of the knowledge-graph, namely obtaining the path between the first intention and the slot position.
In this way, the intent, slot position, and slot position value can be mapped into the knowledge-graph data corresponding to the first intent, and the slot position configuration can be established within the domain of domain knowledge. Therefore, comprehensive field knowledge and profound expert experience can be used for making up the defects of personal knowledge and personal experience of a slot configuration manager, and the value of slot configuration is limited in a reasonable range.
At operation 130, a relationship between the first intent and the at least one slot in the knowledgegraph data may be obtained according to a path between the first intent and the at least one slot, and a preliminary determination may be made as to whether the configuration of the slot is reasonable according to the relationship.
For example, if neither a slot nor a slot position value is a node in the data map, it may be problematic to state that the setting of the slot is out of the scope of the business domain; if no reachable path exists between the first intention and the slot position, the first intention and the slot position are not related, and the configuration of the slot position is not reasonable; if the relationship between the first intent and the slot is one-to-one or equivalent, then the slot setting has no meaning; if the slot set for the first intent has no decision role for subsequent processing either, the slot set also has no meaning. In addition, when other nodes are found to also conform to the path mode according to the knowledge graph and the first intention and the slot position, the corresponding slot position or the slot position value can be recommended to the slot position configuration administrator.
The set conditions can be that slot position configuration is limited in a reasonable scope, and an implementer can flexibly determine the conditions to be set according to specific implementation conditions and implementation requirements and adjust the conditions according to implementation effects; or a condition of a recommended slot position or slot position value, etc.
The prompt information is the information output according to the judgment conclusion after the judgment is carried out. For example, when the configuration of the slot does not meet the set condition, it indicates that the slot configuration is not reasonable, and a corresponding error may be prompted; and when recommendable slots or slot values are found, they may be prompted.
Therefore, according to the information processing method, firstly, a first intention and at least one slot position configured according to the first intention are obtained; then, mapping the first intention and the slot position into entity nodes in the knowledge graph according to knowledge graph data corresponding to the first intention so as to determine a path between the first intention and the slot position; then, whether the slot configuration is reasonable or not is judged according to the knowledge graph data and the path, or more slots or slot values are recommended, and prompt information is given to help an administrator to correct or supplement the slot configuration, so that fine management of intentions and slots is achieved, situations of configuration errors, missing configuration information, redundant slots and the like are reduced, and various problems caused by unreasonable slot configuration in the man-machine conversation process can be reduced.
It should be noted that the embodiment shown in fig. 1 is only one of the most basic embodiments of the information processing method of the present application, and further refinements and extensions may be made on the basis of the embodiment.
According to an embodiment of the present application, determining a path between the first intent and the at least one slot from the knowledge-graph data corresponding to the first intent comprises: performing entity identification on the first intention to obtain at least one first entity; according to the knowledge graph data corresponding to the first intention, performing entity linkage on at least one first entity to obtain at least one second entity; according to the knowledge graph data corresponding to the first intention, performing entity link on at least one slot position to obtain at least one third entity; a path between the at least one second entity and the at least one third entity is determined from the knowledge-graph data corresponding to the first intent.
Sometimes, the intention is not a single noun, but a phrase consisting of a plurality of nouns, or a predicate phrase consisting of actions and nouns or a predicate phrase, etc. Thus, in mapping the first intent to an entity in the knowledge-graph, entity identification is first performed to obtain at least one entity, the first entity.
Since the identified entities may not be completely consistent with the names of the entities in the knowledge-graph due to different cases or names, but actually have the same content, an association or correspondence may be established between the first entity and the entities in the knowledge-graph through entity links. Thus, the search range can be expanded as much as possible to avoid missing, and the path between the first intention and the at least one slot determined by the embodiment is more complete.
According to an embodiment of the present application, the entity identifying the first intention to obtain at least one first entity includes: and performing entity identification on the session intention based on the named entity identification model to obtain at least one first entity.
A Named Entity Recognition (NER) model is obtained by training through a supervised or semi-supervised training method by using a technology based on language grammar and a statistical model, and the accuracy of Entity Recognition is high. In addition, when named entity recognition is used, the named entity recognition can be combined with the knowledge-graph data corresponding to the first intention, so that subsequent entity linkage is easier and quicker.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: determining the category of at least one slot according to the knowledge graph data; acquiring all entities under the category to obtain a fourth entity; and determining whether paths exist between all the fourth entities and the first intention, if so, the configuration of at least one slot position does not meet the set condition.
If a similar path exists between all entities of the same category as the slot and the first intention, which indicates that the slot or the corresponding slot value has no influence on the subsequent processing, the slot may not be set. Therefore, on one hand, computing resources can be saved, on the other hand, user input in the conversation process and the number of conversation rounds with the user can be reduced, the conversation efficiency is higher, information really needed by the user can be provided more quickly, and the user experience is better.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: and determining whether the path meets the association mode of the first intention and the at least one slot position, if not, the configuration of the at least one slot position does not meet the set condition.
The path between the intent and the slot typically also represents a relationship. If the relationship that the intent represents with the path between the slots is "equivalent" or "alias," the setting of the slot may not make sense. These can all be filtered by defining the association pattern that the path should conform to. For example, a path cannot be "null", a path cannot be ". x-equal-,". x-alias- "or the like.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: and determining whether a fifth entity which is matched with the at least one slot exists according to the knowledge graph data and the path, and if so, outputting the fifth entity.
From the knowledge-graph data and the path, other entities that also have similar relationships to the first intent, i.e., the fifth entity, can also be found through the path, which are likely to be other slots or slot values that are equally applicable. Thus, the slot position can be configured more thoroughly and completely, so that the probability of extracting a valid slot position value from the user input is increased, and the conversation of the user is further reduced. In addition, when the slot position is configured completely, the path and the calculation amount of the final answer can be reduced, and the conversation efficiency is higher.
According to an embodiment of the present application, determining whether the configuration of the at least one slot meets the set condition according to the knowledge-graph data and the path includes: determining whether a reachable path exists between the first intent and the at least one slot based on the knowledge-graph data and the path.
If the intention and the slot position or no reachable path exists at all, the intention and the slot position are not related, the configuration is likely to be wrong, prompt information can be output so that a slot position configuration manager can check the configuration, and the problems that the dialog cannot be pushed or a corresponding answer cannot be found in the dialog process due to the configuration mistake are avoided.
According to an embodiment of the present application, prior to determining the path between the first intent and the at least one slot from the knowledge-graph data corresponding to the first intent, the method further comprises: knowledge-graph data corresponding to the first intent is established.
As described above, the knowledge graph data corresponding to the first intention mainly refers to knowledge graph data constructed from domain knowledge and expert opinions of the business domain where the first intention is located, and thus may be constructed by collecting domain knowledge and expert opinions of the business domain, for example, entity information and an association relationship between entities may be extracted from a professional dictionary, an industry standard or a business process specification to construct knowledge graph data.
According to an embodiment of the application, establishing knowledge-graph data corresponding to the first intent includes: establishing a knowledge map database of a business field to which the first intention belongs; and acquiring knowledge graph data corresponding to the first intention according to the knowledge graph database.
Generally speaking, the business field to which the first intention belongs may be large or small, and when the business field to which the first intention belongs is large and knowledge map data established according to domain knowledge and expert opinions is huge, the computational complexity for executing the information processing method of the application also grows exponentially. Thus, only the knowledge-graph data associated with the first intent may be extracted and calculated accordingly based on the knowledge-graph data associated with the first intent. Therefore, the computing resources can be greatly saved, and the computing speed is improved.
Fig. 2 to fig. 4 show another embodiment of the information processing method of the present application, which is applied to an intelligent customer service session system, and the above various embodiments are applied in a comprehensive manner, so as to finally form an embodiment with a better implementation effect.
The embodiments shown in fig. 2 to 4 are applied to an intelligent customer service system of a certain mobile phone product, and are particularly applied to a slot configuration process of the intelligent customer service system.
The dialog processing flow of the intelligent customer service system is shown in fig. 2:
first, user input is received 201;
then, an intent 202 is identified from the user input 201;
then, the corresponding answer set is found according to the intention 202, and then a specific answer for a certain value (e.g., a certain model) is filtered from the answer set by the slot value (slot value 2031 or slot value 2032), for example, answer 1 or answer 2.
Before the information processing method is not used, the slot configuration process of the intelligent customer service system mainly comprises the following steps:
first, an intent 202 is obtained;
then, a slot 203 is set for the intent 202;
thereafter, the slot 203 is configured with a slot value, such as slot value 2031, slot value 2032, or the like.
And then, the configured slot position value is corresponding to answer 1 or answer 2 through system calculation.
Because the configuration of the slot position, that is, the slot position is set for the purpose and the slot position value is configured for the slot position manually and manually by the slot position configuration administrator, the situation that the slot position configuration is unreasonable or the slot position value configuration is wrong is inevitable. Therefore, the daily processing flow is influenced, and the problems that answers cannot be found or the number of conversation rounds is unnecessarily increased are caused.
After using the information processing method of the present application, a slot configuration process of the intelligent customer service system is shown in fig. 3, and mainly includes:
the product knowledge graph is formed by integrating public information such as a product configuration table and the like, and comprises information such as configuration, functions and components of the product.
3030, configuring the slot position to enable the component/function in the intention to be associated with the product;
3050, judging whether other slot positions conforming to the path exist, if so, continuing to 3080;
3080, recommending and prompting other slot positions;
3090, prompting that a slot position is not required to be configured;
Specifically, taking the example given in fig. 4 as an example, when a slot is configured for the intent "using NFC":
in step 3010, the intention "use NFC" is physically linked using the entity identification model, identifying this component "NFC" from the intention;
in step 3020, the intent "NFC in" NFC used "is mapped to the component" NFC "in the product knowledge graph 40;
meanwhile, the slot configuration administrator performs slot configuration for the intention:
assume that, in step 3030, the slot configuration administrator sets a slot "model" for the intent and gives a slot bit value "model C"; upon proceeding to step 3040, the product knowledge-graph 40 is found to be absent this entity; thus, performing a "determination that the product in the tank does not conform to the path" in step 3070 results in a "yes" result; further, entering step 3100, and prompting that the slot configuration may be wrong;
if the configuration administrator gives up the slot position value 'model C' later, and gives a slot position value 'application'; in proceeding to step 3040, it is found that while the product knowledge graph 40 has this entity, there is no reachable path between entity "NFC" and entity "application"; thus, a "yes" result is obtained by performing a "determination that the product in the slot does not conform to the path" in step 3070; entering into step 3100, prompting that slot configuration may be incorrect;
suppose that the slot configuration administrator gives up the slot value "application" and gives a slot value "model a"; upon proceeding to step 3040, it is found that a path "model a-with-NFC" exists in the product knowledge map 40; then, step 3050 is entered, and an entity node similar to "x-has-NFC" is searched in the product knowledge graph 40 shown in fig. 4, and a "model B" is found as a result; subsequently, step 3080 is carried out, and model B is recommended to a slot configuration administrator; in addition, when the step 3060 is entered, only two models, namely the model A and the model B, are found in the product knowledge map 40, which indicates that all models have the NFC component, and then the step 3090 is entered, which indicates that the model does not need to be set with the model slot.
Therefore, when the information processing method is applied to the slot position configuration process, other slot position values which can be filled in can be prompted and recommended after the slot position is configured for a certain intention once, or the rationality of the configured slot position is checked, and the condition that an answer cannot be found in the conversation process caused by configuration errors and missing slot position values is reduced. In addition, the intention irrelevant to the slot position can be prompted, unnecessary slot positions are avoided being set, the user is excessively inquired, or multiple rounds of conversations are triggered. Moreover, by virtue of the product knowledge graph, problems and slots are butted together in a more elaborate manner, i.e., by components or functions being linked together, so that the intent is to be able to perceive the function of the product, thereby assisting the slot configuration administrator in better configuring the slots.
It should be noted that, although steps 3010 to 3020 are performed before step 3030 in the embodiment shown in fig. 2 to 4, in practice, steps 3010 to 3020 may be performed simultaneously with step 3030, and may even be performed after step 3030, as long as the steps are completed before step 3040.
In addition, the embodiments shown in fig. 2 to 4 are only exemplary illustrations of the information processing method of the present application, and are not intended to limit the embodiments and application scenarios of the information processing method of the present application. The implementer can adopt any applicable implementation mode and be applied to any applicable application scene according to specific implementation conditions.
Further, the embodiment of the application also provides an information processing device. As shown in fig. 5, the apparatus 50 includes: a slot position configuration information obtaining module 501, configured to obtain a first intention and at least one slot position configured according to the first intention; a knowledge-graph path determination module 502 to determine a path between the first intent and the at least one slot based on knowledge-graph data corresponding to the first intent; the slot configuration prompting module 503 is configured to determine whether configuration of at least one slot meets a set condition according to the knowledge graph data and the path, and if not, output a prompting message.
According to an embodiment of the present application, the knowledge-graph path determining module 502 includes: the entity identification submodule is used for carrying out entity identification on the first intention to obtain at least one first entity; the entity linking submodule is used for carrying out entity linking on at least one first entity according to the knowledge graph data corresponding to the first intention to obtain at least one second entity; the entity linking submodule is further used for carrying out entity linking on at least one slot position according to the knowledge graph data corresponding to the first intention to obtain at least one third entity; a path computation submodule for determining a path between the at least one second entity and the at least one third entity based on the knowledge-graph data corresponding to the first intent.
According to an embodiment of the application, the entity identification submodule is specifically configured to perform entity identification on the session intention based on a named entity identification model to obtain at least one first entity.
According to an embodiment of the present application, the slot configuration prompting module 503 is specifically configured to: determining the category of at least one slot according to the knowledge graph data; acquiring all entities under the category to obtain a fourth entity; and determining whether paths exist between all the fourth entities and the first intention, if so, the configuration of at least one slot position does not meet the set condition.
According to an embodiment of the present application, the slot configuration prompting module 503 is specifically configured to: and determining whether the path meets the association mode of the first intention and the at least one slot position, if not, the configuration of the at least one slot position does not meet the set condition.
According to an embodiment of the present application, the slot configuration prompting module 503 is specifically configured to: and determining whether a fifth entity which is matched with the at least one slot exists according to the knowledge graph data and the path, and if so, outputting the fifth entity.
According to an embodiment of the present application, the slot configuration prompting module 503 is specifically configured to: determining whether a reachable path exists between the first intent and the at least one slot based on the knowledge-graph data and the path.
According to an embodiment of the present application, the apparatus 50 further comprises a knowledge-graph data establishing module for establishing knowledge-graph data corresponding to the first intention.
According to an embodiment of the application, the knowledge-graph data establishing module comprises: a knowledge map database establishing submodule for establishing a knowledge map database of a business field to which the first intention belongs; and the knowledge graph data acquisition sub-module is used for acquiring knowledge graph data corresponding to the first intention according to the knowledge graph database.
According to a third aspect of embodiments herein, there is provided a computer-readable storage medium comprising a set of computer-executable instructions for performing any one of the above-mentioned information processing methods when the instructions are executed.
Here, it should be noted that: the above description on the embodiment of the information processing apparatus and the above description on the embodiment of the computer readable storage medium are similar to the description on the foregoing method embodiments, and have similar beneficial effects to the foregoing method embodiments, and therefore, the description is not repeated. For technical details that have not been disclosed in the description of the embodiment of the information processing apparatus and the embodiment of the computer-readable storage medium, please refer to the description of the foregoing method embodiments of the present application for understanding, and therefore will not be described again for brevity.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of a unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another device, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media capable of storing program codes, such as a removable storage medium, a Read Only Memory (ROM), a magnetic disk, and an optical disk.
Alternatively, the integrated units described above in the present application may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially implemented or portions thereof that contribute to the prior art may be embodied in the form of a software product stored in a storage medium, and including several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a removable storage medium, a ROM, a magnetic disk, an optical disk, or the like, which can store the program code.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. An information processing method, the method comprising:
acquiring a first intention and at least one slot configured according to the first intention;
determining a path between the first intent and the at least one slot based on knowledge-graph data corresponding to the first intent;
and determining whether the configuration of the at least one slot position meets a set condition or not according to the knowledge graph data and the path, and if not, outputting prompt information.
2. The method of claim 1, the determining a path between the first intent and the at least one slot from knowledge-graph data corresponding to the first intent, comprising:
performing entity identification on the first intention to obtain at least one first entity;
according to the knowledge graph data corresponding to the first intention, performing entity linkage on the at least one first entity to obtain at least one second entity;
according to the knowledge graph data corresponding to the first intention, performing entity link on the at least one slot position to obtain at least one third entity;
determining a path between the at least one second entity and the at least one third entity from the knowledge-graph data corresponding to the first intent.
3. The method of claim 2, the entity identifying the first intent resulting in at least one first entity, comprising:
and performing entity identification on the conversation intention based on a named entity identification model to obtain at least one first entity.
4. The method of claim 1, determining whether a configuration of the at least one slot meets a set condition based on the knowledge-graph data and the path, comprising:
determining the category of the at least one slot according to the knowledge graph data;
acquiring all entities under the category to obtain a fourth entity;
and determining whether the paths exist between all the fourth entities and the first intention, if so, the configuration of the at least one slot position does not meet the set condition.
5. The method of claim 1, determining whether a configuration of the at least one slot meets a set condition based on the knowledge-graph data and the path, comprising:
and determining whether the path meets an association mode of the first intention and the at least one slot position, if not, the configuration of the at least one slot position does not meet a set condition.
6. The method of claim 1, the determining whether the configuration of the at least one slot meets a set condition based on the knowledge-graph data and the path, comprising:
and determining whether a fifth entity adapted to the at least one slot exists according to the knowledge-graph data and the path, and if so, outputting the fifth entity.
7. The method of claim 1, the determining whether the configuration of the at least one slot meets a set condition based on the knowledge-graph data and the path, comprising:
determining whether a reachable path exists between the first intent and the at least one slot based on the knowledge-graph data and the path.
8. The method of claim 1, prior to said determining a path between the first intent and the at least one slot from the knowledge-graph data corresponding to the first intent, the method further comprising:
establishing knowledge-graph data corresponding to the first intent.
9. The method of claim 8, the establishing knowledge-graph data corresponding to the first intent, comprising:
establishing a knowledge map database of a business field to which the first intention belongs;
and acquiring knowledge graph data corresponding to the first intention according to the knowledge graph database.
10. An information processing apparatus, the apparatus comprising:
the slot position configuration information acquisition module is used for acquiring a first intention and at least one slot position configured according to the first intention;
a knowledge-graph path determination module to determine a path between the first intent and the at least one slot based on knowledge-graph data corresponding to the first intent;
and the slot position configuration prompting module is used for determining whether the configuration of the at least one slot position meets the set condition or not according to the knowledge graph data and the path, and outputting prompting information if the configuration of the at least one slot position does not meet the set condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110324895.4A CN113157939A (en) | 2021-03-26 | 2021-03-26 | Information processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110324895.4A CN113157939A (en) | 2021-03-26 | 2021-03-26 | Information processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113157939A true CN113157939A (en) | 2021-07-23 |
Family
ID=76884913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110324895.4A Pending CN113157939A (en) | 2021-03-26 | 2021-03-26 | Information processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113157939A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110502608A (en) * | 2019-07-05 | 2019-11-26 | 平安科技（深圳）有限公司 | The interactive method and human-computer dialogue device of knowledge based map |
| CN111125309A (en) * | 2019-12-23 | 2020-05-08 | 中电云脑(天津)科技有限公司 | Natural language processing method and device, computing equipment and storage medium |
| JP2020135689A (en) * | 2019-02-25 | 2020-08-31 | 株式会社コトバデザイン | Model learning system, intention interpretation system, method for learning model, and model learning program |
| CN111625633A (en) * | 2020-05-22 | 2020-09-04 | 广东飞企互联科技股份有限公司 | Knowledge graph-based enterprise system question-answer intention identification method and device |
| CN111949787A (en) * | 2020-08-21 | 2020-11-17 | 平安国际智慧城市科技股份有限公司 | Automatic question-answering method, device, equipment and storage medium based on knowledge graph |
| US20210027788A1 (en) * | 2019-07-23 | 2021-01-28 | Baidu Online Network Technology (Beijing) Co., Ltd. | Conversation interaction method, apparatus and computer readable storage medium |
| WO2021017290A1 (en) * | 2019-07-31 | 2021-02-04 | 平安科技（深圳）有限公司 | Knowledge graph-based entity identification data enhancement method and system |
-
2021
- 2021-03-26 CN CN202110324895.4A patent/CN113157939A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020135689A (en) * | 2019-02-25 | 2020-08-31 | 株式会社コトバデザイン | Model learning system, intention interpretation system, method for learning model, and model learning program |
| CN110502608A (en) * | 2019-07-05 | 2019-11-26 | 平安科技（深圳）有限公司 | The interactive method and human-computer dialogue device of knowledge based map |
| US20210027788A1 (en) * | 2019-07-23 | 2021-01-28 | Baidu Online Network Technology (Beijing) Co., Ltd. | Conversation interaction method, apparatus and computer readable storage medium |
| WO2021017290A1 (en) * | 2019-07-31 | 2021-02-04 | 平安科技（深圳）有限公司 | Knowledge graph-based entity identification data enhancement method and system |
| CN111125309A (en) * | 2019-12-23 | 2020-05-08 | 中电云脑(天津)科技有限公司 | Natural language processing method and device, computing equipment and storage medium |
| CN111625633A (en) * | 2020-05-22 | 2020-09-04 | 广东飞企互联科技股份有限公司 | Knowledge graph-based enterprise system question-answer intention identification method and device |
| CN111949787A (en) * | 2020-08-21 | 2020-11-17 | 平安国际智慧城市科技股份有限公司 | Automatic question-answering method, device, equipment and storage medium based on knowledge graph |
Non-Patent Citations (1)
| Title |
|---|
| 陈成;陈跃国;刘宸;吕晓彤;杜小勇;: "意图知识图谱的构建与应用", 大数据, no. 02 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021003819A1 (en) | Man-machine dialog method and man-machine dialog apparatus based on knowledge graph | |
| CN106776544B (en) | Character relation recognition method and device and word segmentation method | |
| CN110008308B (en) | Method and device for supplementing information for user question | |
| CN111177307A (en) | Test scheme and system based on semantic understanding similarity threshold configuration | |
| CN112784066A (en) | Information feedback method, device, terminal and storage medium based on knowledge graph | |
| CN113032862A (en) | Building information model checking method and device and terminal equipment | |
| CN113312468A (en) | Conversation mode-based conversation recommendation method, device, equipment and medium | |
| CN105335466A (en) | Audio data retrieval method and apparatus | |
| CN111402864A (en) | Voice processing method and electronic equipment | |
| CN112860873B (en) | Intelligent response method, device and storage medium | |
| CN110609618A (en) | Man-machine conversation method and device, computer equipment and storage medium | |
| CN111125379B (en) | Knowledge base expansion method and device, electronic equipment and storage medium | |
| CN109857450B (en) | Verification service arrangement method and device | |
| CN113157939A (en) | Information processing method and device | |
| CN112541357B (en) | Entity identification method and device and intelligent equipment | |
| CN114202363A (en) | Artificial intelligence based call method, device, computer equipment and medium | |
| US20220262353A1 (en) | Method and device for Processing Voice Information, Storage Medium and Electronic Apparatus | |
| CN112988986B (en) | Man-machine interaction method, device and equipment | |
| CN114239602A (en) | Session method, apparatus and computer program product | |
| CN116755683B (en) | Data processing method and related device | |
| CN117235236B (en) | Dialogue method, dialogue device, computer equipment and storage medium | |
| CN112597149B (en) | Data table similarity determination method and device | |
| CN117421415A (en) | Data processing method, device, electronic equipment and storage medium | |
| CN117032646A (en) | Index configuration method, index execution method and system | |
| CN114647631A (en) | Log query method, device and equipment based on secure shell protocol terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |