CN112148853A - Query result determination method and device, storage medium and electronic device - Google Patents
Query result determination method and device, storage medium and electronic device Download PDFInfo
- Publication number
- CN112148853A CN112148853A CN202010968765.XA CN202010968765A CN112148853A CN 112148853 A CN112148853 A CN 112148853A CN 202010968765 A CN202010968765 A CN 202010968765A CN 112148853 A CN112148853 A CN 112148853A
- Authority
- CN
- China
- Prior art keywords
- query information
- query
- condition
- target
- entity data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000006243 chemical reaction Methods 0.000 claims abstract description 35
- 238000013145 classification model Methods 0.000 claims description 24
- 238000000605 extraction Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 5
- 238000013135 deep learning Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 244000061456 Solanum tuberosum Species 0.000 description 3
- 235000002595 Solanum tuberosum Nutrition 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 244000017020 Ipomoea batatas Species 0.000 description 2
- 235000002678 Ipomoea batatas Nutrition 0.000 description 2
- 235000020247 cow milk Nutrition 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000004615 ingredient Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 235000013606 potato chips Nutrition 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000013589 supplement Substances 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
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- 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
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/313—Selection or weighting of terms for indexing
-
- 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
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method and a device for determining a query result, a storage medium and an electronic device, wherein the method comprises the following steps: acquiring query information input into a question-answering system by a target object; determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages; and performing component conversion on the structural data of the query information according to the target category to determine a query result of the query information, namely performing component conversion on the query information through the target category to determine the query result of the query information.
Description
Technical Field
The present invention relates to the field of communications, and in particular, to a method and an apparatus for determining a query result, a storage medium, and an electronic apparatus.
Background
With the increase of the scale of the e-commerce, the continuous increase of the business and the increase of the number of orders, the work level of the question answering is multiplied, the labor cost is difficult to control, and a large part of the consultation questions belong to repeated questions, so that a question answering system can be selected for answering.
The current answer-question system reply mode is usually a template-based matching method, namely, certain rules are manually maintained, and then inquiry and feedback answer of user questions are realized in a matching mode, although the accuracy of the method is high, the matching degree is low, the matching fails as long as the rules are not met, in addition, along with the improvement of the answer-question workload, a large amount of manual work is required to continuously maintain the rules, and the transportability to similar questions is poor; the question-answering system can only answer basic questions, and is limited by low template matching degree, and answers on most of the questions are difficult to make except extremely common template questions; in addition, the question-answering system can only strictly execute the content of the matched template, and cannot understand the natural semantic meaning and the deeper intention.
Aiming at the problems that in the related technology, a common question answering system can only strictly execute the content of a matched template, so that the question answering efficiency is reduced and the like, an effective technical scheme is not provided.
Disclosure of Invention
The embodiment of the invention provides a method and a device for determining a query result, a storage medium and an electronic device, which are used for solving the problems that the frequently-encountered question-answering system in the related technology can only strictly execute the content of a matched template, so that the question-answering efficiency is reduced and the like.
According to an embodiment of the present invention, there is provided a method for determining a query result, including: acquiring query information input into a question-answering system by a target object; determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages; and performing component conversion on the structural data of the query information according to the target category to determine a query result of the query information.
In an exemplary embodiment, after obtaining the query information input by the target object to the question-answering system, the method further comprises: determining entity data corresponding to the query information; under the condition that the entity data corresponding to the query information is not included in a pre-established knowledge base, taking a target fixed text as a query result of the query information; determining whether the query information is contained in the preset list or not under the condition that the entity data corresponding to the query information is contained in a pre-established knowledge base, wherein the knowledge base at least comprises one of the following items: a plurality of entity data, and relationships between the entity data.
In an exemplary embodiment, determining the target category corresponding to the query information includes: analyzing the query information according to a classification model, and determining a target category corresponding to the query information, wherein the classification model is trained by deep learning by using multiple groups of data, and each group of data in the multiple groups of data comprises: the query information and a target category corresponding to the query information, wherein the target category comprises at least one of the following: question-answer type category, true-false type category.
In one exemplary embodiment, performing component transformation on the structural data of the query information according to the target category to determine a query result of the query information includes: under the condition that the target category is the authenticity type category, acquiring entity data in the query information according to an extraction mode of the authenticity type category; carrying out component conversion on entity data in the query information to obtain a query condition of the query information; and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
In one exemplary embodiment, performing component transformation on the structural data of the query information according to the target category to determine a query result of the query information includes: under the condition that the target category is a question-answer type category, acquiring entity data in the query information according to an extraction mode of the question-answer type category; carrying out component conversion on entity data in the query information to obtain a query condition of the query information; and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
In one exemplary embodiment, performing component transformation on the structural data of the query information according to the target category to determine a query result of the query information includes: sending a query request to the target object under the condition that the components acquired in the component conversion process are missing; receiving the missing component of the target object feedback in response to the query request.
According to an embodiment of the present invention, there is provided a query result determination apparatus including: the acquisition module is used for acquiring query information input into the question-answering system by the target object; a first determining module, configured to determine a target category corresponding to the query information when the query information is not in a preset list, where the preset list includes: a plurality of query messages; and the second determination module is used for performing component conversion on the structural data of the query information according to the target category so as to determine the query result of the query information.
In an exemplary embodiment, the first determining module is further configured to determine entity data corresponding to the query information; under the condition that the entity data corresponding to the query information is not included in a pre-established knowledge base, taking a target fixed text as a query result of the query information; determining whether the query information is contained in the preset list or not under the condition that the entity data corresponding to the query information is contained in a pre-established knowledge base, wherein the knowledge base at least comprises one of the following items: a plurality of entity data, and relationships between the entity data.
According to another embodiment of the present invention, a computer-readable storage medium is also provided, in which a computer program is stored, wherein the computer program is configured to perform the steps of any of the above-described method embodiments when executed.
According to yet another embodiment of the present invention, there is also provided an electronic device, including a memory in which a computer program is stored and a processor configured to execute the computer program to perform the steps in any of the above method embodiments.
By the method, the query information input into a question-answering system by the target object is acquired; determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages; the technical scheme is adopted, the problems that in the related technology, a common question-answering system can only strictly execute the content of matched templates, the question-answering efficiency is reduced and the like are solved, the query information is classified in a target mode, the query result of the query information is finally determined, and the question-answering efficiency is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
fig. 1 is a block diagram of a hardware configuration of a computer terminal of a method for determining a query result according to an embodiment of the present invention;
FIG. 2 is a flow diagram of a method of determining query results according to an embodiment of the invention;
FIG. 3 is a schematic flow chart illustrating the establishment of a knowledge base in a query result determination method according to an alternative embodiment of the present invention;
FIG. 4 is a schematic flow chart illustrating a question-answer flow in a method for determining a query result according to an alternative embodiment of the present invention;
FIG. 5 is a schematic flow chart illustrating component transformation in a method for determining query results according to an alternative embodiment of the present invention;
fig. 6 is a block diagram of the structure of a query result determination apparatus according to an embodiment of the present invention.
Detailed Description
The invention will be described in detail hereinafter with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
The method provided by the embodiment of the application can be executed in a computer terminal, a mobile terminal or a similar operation device. Taking a computer terminal running on the computer terminal as an example, fig. 1 is a hardware structure block diagram of the computer terminal of the method for determining a query result according to the embodiment of the present invention. As shown in fig. 1, a computer terminal may include one or more (only one shown) processors 102 (the processors 102 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the electronic device. For example, the computer terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 may be used to store a computer program, for example, a software program and a module of application software, such as a computer program corresponding to the method for determining the query result in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to a computer terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
An embodiment of the present invention provides a method for determining a query result, which is applied to the computer terminal, and fig. 2 is a flowchart of the method for determining a query result according to the embodiment of the present invention, as shown in fig. 2, the flowchart includes the following steps:
step S202, acquiring query information input into a question-answering system by a target object;
step S204, determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages;
step S206, performing component conversion on the structural data of the query information according to the target category to determine a query result of the query information.
Through the steps, acquiring query information input into a question-answering system by a target object; determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages; the technical scheme is adopted, the problems that in the related technology, a common question-answering system can only strictly execute the content of matched templates, the question-answering efficiency is reduced and the like are solved, the query information is classified in a target mode, the query result of the query information is finally determined, and the question-answering efficiency is improved.
It should be noted that the query result of the query information may be displayed on a terminal with a screen, such as a computer terminal.
In an exemplary embodiment, after obtaining the query information input by the target object to the question-answering system, the method further comprises: determining entity data corresponding to the query information; under the condition that the entity data corresponding to the query information is not included in a pre-established knowledge base, taking a target fixed text as a query result of the query information; determining whether the query information is contained in the preset list or not under the condition that the entity data corresponding to the query information is contained in a pre-established knowledge base, wherein the knowledge base at least comprises one of the following items: a plurality of entity data, and relationships between the entity data.
The above process may be understood as determining whether query information input by the target object is unimportant, and there is no need to feed back query information of the result, specifically, by determining whether entity data corresponding to the query information is included in a pre-established knowledge base, if the entity data corresponding to the query information is not included in the pre-established knowledge base, using the target fixed text as the query result of the query information, and if the entity data corresponding to the query information is included in the pre-established knowledge base, determining again whether the query information is included in the preset list.
That is, if the query information input by the target object is judged to be unimportant, the query information fed back to the result is not needed, and the query information is directly replied to the target object according to the preset fixed text, so that the reply time of the query information is shortened, and the user experience is improved.
In an optional embodiment, the query information is analyzed according to a classification model, and the target category corresponding to the query information is determined, where the classification model is trained by deep learning using multiple sets of data, and each set of data in the multiple sets of data includes: the query information and a target category corresponding to the query information, wherein the target category comprises at least one of the following: question-answer type category, true-false type category.
In short, whether the target category corresponding to the query information is a question-answering type category or a true-false type category is determined according to a classification model, and the classification model is learned and trained by a machine for multiple groups of data. For example, when the query information input to the question-and-answer system by the target object is "how many yuan a sweet potato chip? ", which is obtained by analyzing according to the classification model, and is the query information of a question-answer type category; when the query information input to the question-answering system by the target object is "whether the production date of the cow milk was produced in the last three days? ", it is obtained by analysis according to the classification model, and this is the query information of the true or false type category. The above query information is only selected for understanding the technical solution of the embodiment of the present invention, and any possible query information may be used in the actual operation process.
It should be noted that the classification model may be updated in real time according to actual situations, and the classification model may be maintained by an engineer or may be automatically updated by a system, which is not limited in this embodiment of the present invention.
In one exemplary embodiment, performing component transformation on the structural data of the query information according to the target category to determine a query result of the query information includes: under the condition that the target category is the authenticity type category, acquiring entity data in the query information according to an extraction mode of the authenticity type category; carrying out component conversion on entity data in the query information to obtain a query condition of the query information; and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
That is, the target type of the query information is determined, if the target type of the query information is the authenticity type, the entity data in the query information is obtained according to the authenticity type extraction mode, the query result is obtained by querying in a preset knowledge base according to the query condition of the query information, and the query condition of the query information is obtained by performing component conversion on the entity data of the query information.
For example, the query information extraction method of the authenticity type category may be to extract a main component and an assumed component in the query information to determine entity data in the query information, or may be other extraction methods, which is not limited in the embodiment of the present invention.
In one exemplary embodiment, performing component transformation on the structural data of the query information according to the target category to determine a query result of the query information includes: under the condition that the target category is a question-answer type category, acquiring entity data in the query information according to an extraction mode of the question-answer type category; carrying out component conversion on entity data in the query information to obtain a query condition of the query information; and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
That is, the target type of the query information is determined, if the target type of the query information is question-answer type, the entity data in the query information is obtained according to the extraction mode of the question-answer type, the query result is obtained by querying in a preset knowledge base according to the query condition of the query information, and the query condition of the query information is obtained by performing component conversion on the entity data of the query information.
For example, the query information of the question-and-answer type may be extracted in a manner of analyzing query components to determine entity data in the query information, or in another manner of extracting the query information.
In one exemplary embodiment, performing component transformation on the structural data of the query information according to the target category to determine a query result of the query information includes: sending a query request to the target object under the condition that the components acquired in the component conversion process are missing; receiving the missing component of the target object feedback in response to the query request.
In short, in the process of converting the components of the query information, if the components of the query information are judged to be missing, the missing components needing to be supplemented are sent to the target object, and then the complete components are obtained.
It should be noted that, when the query information sent by the target object is analyzed and the target type of the query information cannot be analyzed, the query information is forwarded to the human customer service, or the confidence coefficient is insufficient when the components are converted, that is, the classification model determines that the query information is unreliable query information, and the query information is also forwarded to the human customer service.
In order to better understand the flow of the determination method of the query result, the following description is made with reference to an alternative embodiment, but is not intended to limit the technical solution of the embodiment of the present invention.
The method for determining the query result provided in the alternative embodiment of the present invention has the following main processes:
the method comprises the following steps: firstly, establishing a knowledge base in the following way, as shown in FIG. 3;
1. text data (corresponding to entity data in the above-described embodiment): text data generally represents unstructured data, and is data with structured knowledge extracted from the unstructured data.
2. Entity identification: entities in the vertical domain are identified from the text. Such as trade name, brand name, place of manufacture, price, etc., in the e-commerce field.
3. The relationship (equivalent to the relationship between the entity data in the above embodiment) extracts: and extracting the entity and the relation between the entities from the text. For example, "kat is the leading of the eagle team," including the relationship [ kat ] < occupational position > [ leading ], [ kat ] < affiliate team > [ eagle team ]. The entity identification and the relation extraction methods all adopt methods commonly used in the NLP field.
4. Structured/semi-structured data: the data can be well-organized and can be directly used, or the data can be well-organized and the data format is unavailable, and different data sets can be directly analyzed.
5. And (3) knowledge fusion: normalizing all the obtained entity and entity relation data, judging whether some entities are of one type or not according to attribute similarity, and adding expert knowledge for correction.
6. And (3) common QA set arrangement: the most common questions are listed and the corresponding answers are collated by experienced customer service.
The entities, entity relationships and common QA sets form the basic knowledge base (corresponding to the pre-established knowledge base in the above embodiments) of the intelligent customer service.
Step two: the question-answering process, as shown in fig. 4, includes the following steps:
step 1, the user asks a question (corresponding to the query information sent by the target object in the above embodiment): the user initiates a question.
Step 2, whether the chat is chatting: according to the method and the device for question recognition, a conventional text classification model is not used, entity recognition is carried out on the questions, whether the questions contain the entities mentioned in the knowledge base or not is carried out, if the questions do not contain the entities mentioned in the knowledge base, the questions initiated by the user are considered to be chatting, and the preset texts are directly replied for the questions judged to be chatting.
And 3, judging whether the QA is common QA: and judging whether the query is a common QA set mentioned in the knowledge base or not by using a task type of multi-classification of a text, and if the query corresponds to a certain answer, automatically replying based on a preset answer.
And 4, judging other intentions: it is determined whether the question is an "true or false" type question (xxx has xx effect, do).
Step 5, dependency syntax analysis-authenticity type: parse the question according to a grammatical structure, such as "sk-ii can preserve moisture? ", extract subject: "sk-ii" is "moisture retention" or not, and the trade name is obtained by ingredient conversion: sk-ii, query conditions: efficacy queries, hypothetical ingredients: efficacy-moisturizing, output result is whether.
The process of component conversion, as shown in fig. 5, can be implemented as follows: and converting part of contents of the user questions into a structural form with conditions and capable of inquiring the knowledge base, and returning corresponding inquiry results. Checking whether the components are complete: for example, "I want to know the price of the chip," the sentence is intended to be identified as "answer" type: "what is the price," but the presence of the body "chip" indicates the absence (without strict conditions to which chip) and is therefore not complete. When the components are incomplete, the obtained components are stored by maintaining the dialogue state information, the missing components are asked reversely, and the complete query information is monitored and supplemented. For example, the supplement of 'happy bubble potato chips' is converted into query conditions: entity: "happy bubble potato chip", entity attribute: and the price inquires corresponding knowledge from a knowledge base and returns a result.
Step 6, dependency syntax analysis-question answering: the problem is resolved according to the grammatical structure, for example, "what the most expensive potato chip is" extract the subject "is" what "the most expensive potato chip is", and the brand is obtained by component conversion: music, category: potato chips, query conditions and sorting conditions: the price is from big to small, and the result is finally output.
Step 7, manual operation: when the content analysis is carried out, the confidence coefficient is not enough when the components are converted, namely the classification model is judged to be unreliable query content by the model, and the part of the content is manually served.
For the judgment of the final effect, the optional embodiment of the present invention adopts the following indexes: 1. accuracy, i.e., the accuracy with which the classification is judged when a text classification task is used; 2. and the labor conversion rate is used for providing a labor conversion demand when the model cannot meet the customer demand, and when the labor conversion ratio meets a certain threshold value, the intelligent customer service is considered to meet the current demand.
The invention can construct a perfect knowledge base in the optional embodiment, so that the intelligent customer service has a knowledge range with high coverage and can quickly respond to most of the existing problems; the method of using a deep learning model to replace key word matching/template matching and the like improves the discrimination efficiency and the discriminable range; based on the requirements of the vertical field, whether the conventional problems are common QA problem sets or not is judged by multiple categories, and the quick response can obviously improve the operation efficiency; based on the vertical field requirement, the problem of a user is disassembled, the component analysis is realized based on the dependency syntax, the component conversion is introduced, the text problem is converted into an executable and queryable condition, so that the problem solution is implemented, and the intelligent customer service in the E-commerce vertical field is realized more efficiently and smoothly.
It should be noted that the execution order of the above steps may be exchanged or cyclically executed in some cases, which is not limited in the embodiment of the present invention.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
In this embodiment, a device for determining a query result is further provided, where the device is used to implement the foregoing embodiments and preferred embodiments, and details are not repeated for what has been described. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 6 is a block diagram of a structure of an apparatus for determining a query result according to an embodiment of the present invention, as shown in fig. 4, the apparatus including:
(1) the acquisition module 40 is used for acquiring query information input into the question-answering system by the target object;
(2) a first determining module 42, configured to determine a target category corresponding to the query information when the query information is not in a preset list, where the preset list includes: a plurality of query messages;
(3) and the second determining module 44 is configured to perform component conversion on the structural data of the query information according to the target category to determine a query result of the query information.
Acquiring query information input into a question-answering system by a target object through the module; determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages; the technical scheme is adopted, the problems that in the related technology, a common question-answering system can only strictly execute the content of matched templates, the question-answering efficiency is reduced and the like are solved, the query information is classified in a target mode, the query result of the query information is finally determined, and the question-answering efficiency is improved.
The query result of the query information is displayed on the computer terminal.
In an exemplary embodiment, the first determining module is further configured to determine entity data corresponding to the query information; under the condition that the entity data corresponding to the query information is not included in a pre-established knowledge base, taking a target fixed text as a query result of the query information; determining whether the query information is contained in the preset list or not under the condition that the entity data corresponding to the query information is contained in a pre-established knowledge base, wherein the knowledge base at least comprises one of the following items: a plurality of entity data, and relationships between the entity data.
That is, the first determining module determines whether the entity data corresponding to the query information is included in the pre-established knowledge base, if the entity data corresponding to the query information is not included in the pre-established knowledge base, the target fixed file is used as a query result of the query information, and if the entity data corresponding to the query information is included in the pre-established knowledge base, whether the query information is included in the preset list is determined again.
Optionally, the first determining module is further configured to analyze the query information according to a classification model, and determine a target category corresponding to the query information, where the classification model is trained through deep learning by using multiple groups of data, and each group of data in the multiple groups of data includes: the query information and a target category corresponding to the query information, wherein the target category comprises at least one of the following: question-answer type category, true-false type category.
In short, whether the target category corresponding to the query information is a question-answer type category or a true-false type category is obtained according to the classification model analysis, and the classification model is learned and trained by a machine for multiple groups of data. For example, when the query information input to the question-and-answer system by the target object is "how many yuan a sweet potato chip? ", which is obtained by analyzing according to the classification model, and is the query information of a question-answer type category; is the query information input into the question-answering system by the target object "is whether the production date of cow milk was the last three days? ", it is obtained by analysis according to the classification model, and this is the query information of the true or false type category. The above query information is only selected for understanding the technical solution of the embodiment of the present invention, and any possible query information may be used in the actual operation process.
It should be noted that the classification model may be updated in real time according to actual situations, and the classification model may be maintained by an engineer or may be automatically updated by a system, which is not limited in this embodiment of the present invention.
Optionally, the second determining module is further configured to, when the target category is an authenticity category, obtain entity data in the query information according to an extraction manner of the authenticity category; carrying out component conversion on entity data in the query information to obtain a query condition of the query information; and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
That is, the target type of the query information is determined, if the target type of the query information is the authenticity type, the entity data in the query information is obtained according to the authenticity type extraction mode, the query result is obtained by querying in a preset knowledge base according to the query condition of the query information, and the query condition of the query information is obtained by performing component conversion on the entity data of the query information.
For example, the query information extraction method of the authenticity type category may be to extract a main component and an assumed component in the query information to determine entity data in the query information, or may be other extraction methods, which is not limited in the embodiment of the present invention.
Optionally, the second determining module is further configured to, when the target category is a question-answer type category, obtain entity data in the query information according to an extraction manner of the question-answer type category; carrying out component conversion on entity data in the query information to obtain a query condition of the query information; and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
That is, the target type of the query information is determined, if the target type of the query information is question-answer type, the entity data in the query information is obtained according to the extraction mode of the question-answer type, the query result is obtained by querying in a preset knowledge base according to the query condition of the query information, and the query condition of the query information is obtained by performing component conversion on the entity data of the query information.
For example, the query information of the question-and-answer type may be extracted in a manner of analyzing query components to determine entity data in the query information, or in another manner of extracting the query information.
Optionally, the second determining module is further configured to perform component conversion on the structural data of the query information according to the target category to determine a query result of the query information, and includes: sending a query request to the target object under the condition that the components acquired in the component conversion process are missing; receiving the missing component of the target object feedback in response to the query request.
In short, in the process of converting the components of the query information, if the components of the query information are judged to be missing, the missing components needing to be supplemented are sent to the target object, and then the complete components are obtained.
It should be noted that, when the query information sent by the target object is analyzed, the confidence coefficient is not enough during component conversion, that is, the classification model determines that the query information is unreliable query information, and the query information is forwarded to manual customer service.
An embodiment of the present invention further provides a storage medium including a stored program, wherein the program executes any one of the methods described above.
Alternatively, in the present embodiment, the storage medium may be configured to store program codes for performing the following steps:
s1, acquiring query information input into the question-answering system by the target object;
s2, determining a target category corresponding to the query information when the query information is not in a preset list, where the preset list includes: a plurality of query messages;
s3, performing component conversion on the structural data of the query information according to the target category to determine the query result of the query information.
Embodiments of the present invention also provide an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.
Optionally, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:
s1, acquiring query information input into the question-answering system by the target object;
s2, determining a target category corresponding to the query information when the query information is not in a preset list, where the preset list includes: a plurality of query messages;
s3, performing component conversion on the structural data of the query information according to the target category to determine the query result of the query information.
Optionally, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.
Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation manners, and this embodiment is not described herein again.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for determining query results, comprising:
acquiring query information input into a question-answering system by a target object;
determining a target category corresponding to the query information under the condition that the query information is not in a preset list, wherein the preset list comprises: a plurality of query messages;
and performing component conversion on the structural data of the query information according to the target category to determine a query result of the query information.
2. The method of claim 1, wherein after obtaining query information input by the target object to the question-answering system, the method further comprises:
determining entity data corresponding to the query information;
under the condition that the entity data corresponding to the query information is not included in a pre-established knowledge base, taking a target fixed text as a query result of the query information;
determining whether the query information is contained in the preset list or not under the condition that the entity data corresponding to the query information is contained in a pre-established knowledge base, wherein the knowledge base at least comprises one of the following items: a plurality of entity data, and relationships between the entity data.
3. The method of claim 1, wherein determining the target category corresponding to the query information comprises:
analyzing the query information according to a classification model, and determining a target category corresponding to the query information, wherein the classification model is trained by deep learning by using multiple groups of data, and each group of data in the multiple groups of data comprises: the query information and a target category corresponding to the query information, wherein the target category comprises at least one of the following: question-answer type category, true-false type category.
4. The method of claim 3, wherein performing component transformation on the structural data of the query information according to the target category to determine the query result of the query information comprises:
under the condition that the target category is the authenticity type category, acquiring entity data in the query information according to an extraction mode of the authenticity type category;
carrying out component conversion on entity data in the query information to obtain a query condition of the query information;
and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
5. The method of claim 3, wherein performing component transformation on the structural data of the query information according to the target category to determine the query result of the query information comprises:
under the condition that the target category is a question-answer type category, acquiring entity data in the query information according to an extraction mode of the question-answer type category;
carrying out component conversion on entity data in the query information to obtain a query condition of the query information;
and querying in a preset knowledge base based on the query condition to determine a query result of the query information.
6. The method of claim 1, wherein performing component transformation on the structural data of the query information according to the target category to determine the query result of the query information comprises:
sending a query request to the target object under the condition that the components acquired in the component conversion process are missing;
receiving the missing component of the target object feedback in response to the query request.
7. An apparatus for determining a query result, comprising:
the acquisition module is used for acquiring query information input into the question-answering system by the target object;
a first determining module, configured to determine a target category corresponding to the query information when the query information is not in a preset list, where the preset list includes: a plurality of query messages;
and the second determination module is used for performing component conversion on the structural data of the query information according to the target category so as to determine the query result of the query information.
8. The apparatus of claim 7, wherein the first determining module is further configured to determine entity data corresponding to the query information; under the condition that the entity data corresponding to the query information is not included in a pre-established knowledge base, taking a target fixed text as a query result of the query information; determining whether the query information is contained in the preset list or not under the condition that the entity data corresponding to the query information is contained in a pre-established knowledge base, wherein the knowledge base at least comprises one of the following items: a plurality of entity data, and relationships between the entity data.
9. A computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to carry out the method of any one of claims 1 to 6 when executed.
10. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and wherein the processor is arranged to execute the computer program to perform the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010968765.XA CN112148853A (en) | 2020-09-15 | 2020-09-15 | Query result determination method and device, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010968765.XA CN112148853A (en) | 2020-09-15 | 2020-09-15 | Query result determination method and device, storage medium and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112148853A true CN112148853A (en) | 2020-12-29 |
Family
ID=73893362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010968765.XA Pending CN112148853A (en) | 2020-09-15 | 2020-09-15 | Query result determination method and device, storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112148853A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107885842A (en) * | 2017-11-10 | 2018-04-06 | 上海智臻智能网络科技股份有限公司 | Method, apparatus, server and the storage medium of intelligent answer |
| CN108197167A (en) * | 2017-12-18 | 2018-06-22 | 深圳前海微众银行股份有限公司 | Human-computer dialogue processing method, equipment and readable storage medium storing program for executing |
| CN111046133A (en) * | 2019-10-29 | 2020-04-21 | 平安科技（深圳）有限公司 | Question-answering method, question-answering equipment, storage medium and device based on atlas knowledge base |
| CN111274371A (en) * | 2020-01-14 | 2020-06-12 | 东莞证券股份有限公司 | Intelligent man-machine conversation method and equipment based on knowledge graph |
| US20200279001A1 (en) * | 2019-03-01 | 2020-09-03 | Cuddle Artificial Intelligence Private Limited | Systems and methods for adaptive question answering |
-
2020
- 2020-09-15 CN CN202010968765.XA patent/CN112148853A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107885842A (en) * | 2017-11-10 | 2018-04-06 | 上海智臻智能网络科技股份有限公司 | Method, apparatus, server and the storage medium of intelligent answer |
| CN108197167A (en) * | 2017-12-18 | 2018-06-22 | 深圳前海微众银行股份有限公司 | Human-computer dialogue processing method, equipment and readable storage medium storing program for executing |
| US20200279001A1 (en) * | 2019-03-01 | 2020-09-03 | Cuddle Artificial Intelligence Private Limited | Systems and methods for adaptive question answering |
| CN111046133A (en) * | 2019-10-29 | 2020-04-21 | 平安科技（深圳）有限公司 | Question-answering method, question-answering equipment, storage medium and device based on atlas knowledge base |
| CN111274371A (en) * | 2020-01-14 | 2020-06-12 | 东莞证券股份有限公司 | Intelligent man-machine conversation method and equipment based on knowledge graph |
Non-Patent Citations (3)
| Title |
|---|
| 娄红涛: "互联网下智能客服系统的建设与实践", 《信息与电脑》 * |
| 朱频频主编: "《智能客户服务技术与应用》", 中国铁道出版社 * |
| 杨燕: "面向电商领域的智能问答系统若干关键技术研究", 《中国博士学位论文全文数据库》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10055501B2 (en) | Web-based customer service interface | |
| CN110807085B (en) | Fault information query method and device, storage medium and electronic device | |
| CN107784051B (en) | Online customer service response system and method | |
| CN110019703B (en) | Data marking method and device and intelligent question-answering method and system | |
| CN104796434A (en) | Message pushing method and message server | |
| WO2012126259A1 (en) | System having information distributing and searching functions and information distribution method | |
| CN111797202B (en) | Customer service conversation method based on artificial intelligence | |
| CN101101605A (en) | Method, device and system for searching web page and device for establishing index database | |
| CN110287318B (en) | Service operation detection method and device, storage medium and electronic device | |
| CN112182186A (en) | Intelligent customer service operation method, device and system | |
| CN105701224B (en) | Security information customized service system based on big data | |
| CN109508367A (en) | Automatically extract the method, on-line intelligence customer service system and electronic equipment of question and answer corpus | |
| CN107832342B (en) | Robot chatting method and system | |
| CN101339560A (en) | Method and device for searching series data, device and search engine system | |
| CN110750626B (en) | Scene-based task-driven multi-turn dialogue method and system | |
| CN112860873B (en) | Intelligent response method, device and storage medium | |
| CN109145092B (en) | Database updating and intelligent question and answer management method, device and equipment | |
| CN111325562B (en) | Grain safety tracing system and method | |
| CN110929519B (en) | Entity attribute extraction method and device | |
| CN112148853A (en) | Query result determination method and device, storage medium and electronic device | |
| CN109145124B (en) | Information storage method and device, storage medium and electronic device | |
| CN111651554A (en) | Insurance question-answer method and device based on natural language understanding and processing | |
| CN111046632A (en) | Data extraction and conversion method, system, storage medium and electronic equipment | |
| CN106304026A (en) | The determination method and device of end message | |
| WO2016059505A1 (en) | A system and a method for recognition of aerospace parts in unstructured text |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201229 |
|
| RJ01 | Rejection of invention patent application after publication |