CN107729549B - Robot customer service method and system including element extraction - Google Patents
Robot customer service method and system including element extraction Download PDFInfo
- Publication number
- CN107729549B CN107729549B CN201711060950.3A CN201711060950A CN107729549B CN 107729549 B CN107729549 B CN 107729549B CN 201711060950 A CN201711060950 A CN 201711060950A CN 107729549 B CN107729549 B CN 107729549B
- Authority
- CN
- China
- Prior art keywords
- question
- user
- robot
- instance
- knowledge
- 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.)
- Active
Links
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
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/3325—Reformulation based on results of preceding query
- G06F16/3326—Reformulation based on results of preceding query using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention relates to a robot customer service method and a system containing element extraction, wherein the element extraction is carried out on the user question of a received task-type knowledge point according to a pre-classified FAQ knowledge point; and carrying out accurate answer of the robot according to the extracted elements. The invention utilizes natural language processing and deep learning technology to identify the user's consultation intention, collects the elements mentioned in the user's question, and synthesizes the element information to return more accurate answers to the user.
Description
Technical Field
The invention relates to a robot customer service technology, in particular to a robot customer service method and system including element extraction.
Background
The existing robot guest adopts a computer to judge the user questions in the customer service scene and provides corresponding answers so as to reduce the participation of human seats in the customer service scene, and the aim is to reduce the labor cost in the customer service. At present, the robot service mostly adopts a corresponding answer to each question of the user, and the content of the answer is generally a preset standard answer. These standard answers are designed for FAQ (Frequently Asked Questions).
The robot service is not limited to the channel of the instant communication form, and in the channel forms of telephone service, video service and the like, the robot service can be applied by adopting the technologies of audio and video coding, decoding, voice synthesis, animation synthesis and the like. The robot customer service generally adopts a question-and-answer mode.
1, firstly, an FAQ library is indexed, and standard questions and answer contents of the FAQ are set. For example, the query and Answer of the following FAQ libraries are subjected to word segmentation processing to build a text index, such as:
q: forget what do the password?
A: selecting '…' -payment management- > forgetting payment password- > selecting bank card- > filling related information according to prompt- > resetting password.
And 2, when a user asks, the robot understands the question of the user, then retrieves the FAQ and sorts the FAQ, finally returns the answer to the user, wherein the FAQ is matched with the FAQ closest to the question of the user.
The answers returned to the user can be divided into two categories:
the first category employs static text content, typically for general customer service consultation;
the second category needs to perform dynamic function call, and returns a result to a user after inquiring a service database, and is generally used for user information inquiry, self-service operation, shopping guide scenes and the like.
Because the existing customer service robot adopts a mode of matching the question of the user with the FAQ, only single-dimensional retrieval matching between the question semantics and the FAQ is supported, and the element conditions mentioned in the question by the user cannot be identified, the answer returned to the user in some scenes is not accurate enough.
Example 1:
in the query billing scenario, the user asks "how much money I spend last month in Shanghai", where the user provides two key conditions: time is equal to "last month", and place is equal to "shanghai"; however, the existing customer service robot can only reply the general current credit card bill and cannot accurately reply according to the key element conditions of the user.
Example 2:
in the air ticket booking scene, a user asks me an air ticket booking to Shenzhen, wherein the user provides an element condition: destination ═ Shenzhen "; however, the current service robot can only reply to the universal text answer, guide the user to go to the booking link, and the destination in the booking link is still the default address "shanghai" instead of the destination condition "Shenzhen" provided by the user.
There is a need for a customer service robot that returns answers to a user with precision in some scenarios.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a robot customer service method and system comprising element extraction.
The purpose of the invention is realized by adopting the following technical scheme:
the invention provides a robot customer service method comprising element extraction, and the improvement is as follows:
performing element extraction on the user questions of the received task-type knowledge points according to the pre-classified knowledge points;
and carrying out accurate answer of the robot according to the extracted elements.
Further: the pre-classified knowledge points comprise common FAQ question-answer knowledge points and task-type knowledge points, and the classified knowledge points are stored in a knowledge base; when the user question is a common FAQ question and answer, no element extraction is needed; when the user asks a task-type knowledge point, element extraction is needed; the elements are described in the form of word slots;
the word slot is an element information category which needs to be collected by the robot from question or answer of the user in the task; the element information category includes:
there are example keywords: enumeratable and standard names, and normalization processing is required when the elements are extracted;
no example keywords: the instances in the word slot are not enumerated, and can be considered to be without instances, and normalization processing is not needed during element extraction;
semantic type: the word slot has enumerable examples, but the examples cannot be directly normalized to one example through keywords.
Further: the element extraction of the user questions of the received task knowledge points according to the pre-classified knowledge points comprises the following steps:
after receiving a question of a user, scoring the matching degree of each knowledge point in the knowledge base according to the text content in the question of the user, and finally returning the knowledge point which is most matched with the user;
and if the knowledge points belong to task-type knowledge points, the robot extracts corresponding elements from the user questions.
Further: before the robot accurately answers according to the extracted elements, normalization processing is carried out on the extracted elements to form standard names;
the normalization process includes: creating and managing a word slot related in a task type scene in advance, an instance standard name and a plurality of aliases of the instance under the word slot; when the elements in the user question hit the alias of a certain instance or the similar alias with high confidence coefficient, the elements are normalized into the standard instance name, and then interface calling is carried out.
Further: the performing of the robot accurate answer according to the extracted elements includes: and the robot calls a service interface and accurately answers according to the element conditions.
6. A robot customer service system including element extraction, characterized in that:
the element extraction module is used for extracting elements according to the received user questions of the task knowledge points by the pre-classified knowledge points;
and the service answer module is used for carrying out accurate robot answer according to the extracted elements.
Further: the pre-classified knowledge points comprise common FAQ question-answer and task-type knowledge points, and the classified FAQ knowledge points are stored in a knowledge base; when the user question is a common FAQ question and answer, no element extraction is needed; when the user asks a task-type knowledge point, element extraction is needed; the elements are described in terms of word slots.
Further: the element extraction module comprises:
the matching unit is used for scoring the matching degree of each knowledge point in the knowledge base according to the text content in the question of the user and finally returning the knowledge point which is most matched with the user;
and the extracting unit is used for extracting corresponding elements from the user question by the robot if the knowledge points belong to the task-type knowledge points.
Further: the system also comprises a normalization module which is used for normalizing the extracted elements to form standard names before the robot accurately answers according to the extracted elements.
Further: and the root service answer module is also used for calling a service interface by the robot and carrying out accurate answer according to the element conditions.
Compared with the closest prior art, the technical scheme provided by the invention has the following beneficial effects:
(1) and the user experience is improved. And carrying out accurate answer according to the element conditions mentioned in the user question.
(2) And (6) analyzing the data. The robot can collect structured element information from the unstructured question of the user through normalization processing, and then analyze hot examples in each element word slot according to scenes.
(3) The scheme adopts the robot customer service in an instant communication channel, but the scheme is not limited to be applied to other channel forms such as telephone customer service, video customer service and the like, and has various forms and strong usability.
(4) The scheme adopts an account checking list scene as an example, but the scheme is not limited to be applied to other inquiry scenes, self-service operation scenes and shopping guide scenes, such as film seat number checking, self-service account opening, air ticket booking and the like, and the application range is wide.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow diagram of a method of robotic customer service including element extraction;
FIG. 2 is a diagram illustrating effects of the first embodiment;
FIG. 3 is a schematic illustration of an element management scheme;
FIG. 4 is a schematic diagram of an elemental educational program;
FIG. 5 is a schematic diagram of an intent tag;
fig. 6 is a schematic diagram of an example of selection.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
The first embodiment,
The invention provides a robot customer service method including element extraction, a flow chart of which is shown in figure 1, and the method comprises the following steps:
performing element extraction on the user questions of the received task-type knowledge points according to the pre-classified FAQ knowledge points;
and carrying out accurate answer of the robot according to the extracted elements. When a user asks questions, the robot firstly scores the matching degree of each knowledge point in the knowledge base according to the text content in the question of the user and matches the matching degree to one knowledge point in the knowledge base, and if the knowledge point belongs to a task-type knowledge point, the robot extracts corresponding elements from the question of the user and normalizes the elements into standard names so as to carry out structured query.
The pre-classified FAQ knowledge points are common FAQ knowledge points, and element extraction is not needed; for the consultation intention that the user may mention the elements, dividing the consultation intention into another type of knowledge points (hereinafter referred to as task-type knowledge points), extracting the elements, and storing the classified FAQ knowledge points into a knowledge base; the elements are described in terms of word slots.
The element extraction of the user questions of the received task knowledge points according to the pre-classified FAQ knowledge points comprises the following steps:
after receiving a question asked by a user, matching with an FAQ knowledge point in a knowledge base through intention identification;
and if the knowledge points belong to task-type knowledge points, the robot extracts corresponding elements from the user questions.
Before the accurate answer of the robot is carried out according to the extracted elements, normalization processing is carried out on the extracted elements to form standard names.
The performing of the robot accurate answer according to the extracted elements includes: and the robot calls a service interface and accurately answers according to the element conditions. The effect of the first embodiment is shown in fig. 2.
Example II,
Firstly, an element management scheme in element extraction: the overall structure is shown in fig. 3:
1. word groove:
the word slot is an element information category that the robot needs to collect from the user question (or answer) in the task, such as "bill time", "consumption place", "consumption type" in the audit scene; and "origin", "destination" and "departure time" in the air ticket booking scenario.
2. Example (c):
for a certain word slot, an instance is a value that may exist in the word slot, for example, a possible value in the word slot "origin" is a set of cities such as "Beijing", "Shenzhen", "Shanghai", etc., where "Beijing" is an instance of the word slot "origin".
3. Alias:
for one example, the alias is another name that may exist in the spoken language environment of the user, for example, the alias of "beijing" has "capital", "monarch", etc., where "capital" is an alias of the example "beijing".
4. Ordinary word groove classification:
1) having instance keywords
Instances in such word slots may be enumerated and have standard names, requiring normalization processing at element extraction. For example, in the word slot "movie", there is an instance "speed and passion 8", and "speed 8" is an alias of this instance, and after "speed 8" is extracted from the user question, it needs to be normalized to "speed and passion 8".
2) Instance-free keywords
The instances in such word slots are not enumeratable, and can be considered as no instances, and no normalization process is needed during element extraction. Instances may not be enumerated as in the word slot "person name" and correspondence cannot be normalized to instances.
3) Semantic type
The word slot has enumerable examples, but the examples cannot be directly normalized to one example through keywords. Such as word slot "consumption type", sometimes it cannot be judged by a certain word in the user question sentence which consumption type the user inquires, but it is identified which consumption type should be mapped by combining the semantics of the whole sentence.
5. Time word slot classification:
1) point in time
This type is only applicable to the system preset word slot "time point", and its example is a continuous value on the time axis, so it is not enumeratable, but needs to be normalized to a certain point on the time axis according to the user's text.
2) Time interval
This type is only applicable to the system preset word slot "time interval", and an example thereof is any interval continuous on the time axis, so it is not enumeratable, but needs to be normalized to a certain interval on the time axis according to the text of the user.
3) Floating time point
This type is applicable only to the system preset word slot "floating time point", examples of which are arbitrary time points not belonging to the time axis, such as "No. 1 per month", "early year", "late month". This type of word slot is not enumerable, but needs to be normalized to some floating point in time based on the user's text.
4) Floating time interval
This type is only applicable to the system preset word tank "floating time interval", examples of which are not ascribed to any period of time on the time axis, such as "months", "1-3 years", "days". This type of word slot is not enumerable, but needs to be normalized to some floating time interval according to the user's text.
6. The word slot inherits:
there is inheritance relationship between word slots: for example, the "origin" and the "destination" are inherited from the "city"
1) When a new word slot is created, an existing word slot can be inherited optionally
2) The inherited content includes: type, instance and alias, annotation data
3) Editing and education of sub-word slots does not affect parent word slots
Second, element education scheme
Element education is performed in a manner of manually marking the question sentence. The overall procedure is shown in figure 4.
1. Intention labeling
The first step requires intent labeling, i.e., matching the user question to an existing knowledge point in the knowledge base from the semantic dimension. The knowledge points are divided into task-type knowledge points and FAQ knowledge points, and if the knowledge points are marked as the task-type knowledge points, the word slot marking needs to be carried out in the second step. As shown in fig. 5.
2. Word groove label
The word slot labeling is completed through three steps of word extraction, word slot selection and instance selection.
1) Word-taking
The element key (except for the semantic word slot, see below) is selected from the user question, corresponding to the "alias" in element management.
2) Word selection groove
And selecting a word slot to which the keyword belongs.
3) Selection example
Select which instance the keyword should be normalized to under the selected word slot (except no instance type word slot and time type word slot, as shown in fig. 6.)
3. Different types of word slot labeling:
1) having instance keywords
When the example keywords are marked, words need to be fetched, word slots are selected, and example standard names are selected.
2) Instance-free keywords
The non-instance keywords need to be word-fetched and word slots are selected when being labeled, but the instance standard names are not selected.
3) Semantic type
The semantic word slot does not need to fetch words during labeling, but selects 'semantics'; it is necessary to select a word slot and select an instance standard name.
4) Time point, time interval, floating time point, floating time interval
When marking, it needs to fetch words and select word slot, but does not select example standard name.
Example III,
Based on the same inventive concept, the invention also provides a robot customer service system comprising the same, wherein the robot customer service system comprises
The element extraction module is used for extracting elements from the received user questions of the task knowledge points according to the pre-classified FAQ knowledge points;
and the service answer module is used for carrying out accurate robot answer according to the extracted elements.
Further: the pre-classified FAQ knowledge points comprise common FAQ question-answer and task type knowledge points, and the classified FAQ knowledge points are stored in a knowledge base; when the user question is a common FAQ question and answer, no element extraction is needed; when the user asks a task-type knowledge point, element extraction is needed.
Further: the element extraction module comprises:
the matching unit is used for scoring the matching degree of each knowledge point in the knowledge base according to the text content in the question of the user after receiving the question of the user and finally returning the knowledge point which is the best matched with the user;
and the extracting unit is used for extracting corresponding elements from the user question by the robot if the knowledge points belong to the task-type knowledge points.
Further: the system also comprises a normalization module which is used for normalizing the extracted elements to form standard names before the robot accurately answers according to the extracted elements.
Further: and the root service answer module is also used for calling a service interface by the robot and carrying out accurate answer according to the element conditions.
The invention improves the user experience. And carrying out accurate answer according to the element conditions mentioned in the user question. And the data analysis is carried out, so that the robot can collect structured element information from the unstructured question of the user through normalization processing, and then analyzes hot examples in each element word slot according to scenes, and can accurately answer the questions of the user.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention 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 invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (8)
1. A robot customer service method including element extraction is characterized in that:
receiving a question sentence asked by a user;
performing intention labeling on the question, wherein the intention labeling is to match the question to an existing knowledge point in a knowledge base from semantic dimensions, and the existing knowledge point comprises a task-type knowledge point and a common FAQ question-answer knowledge point;
storing the classified knowledge points to a knowledge base; when the question of the user belongs to the common FAQ question-answer knowledge point, no element extraction is needed; when a user question is a task-type knowledge point, element extraction is required to be carried out in a word slot labeling mode; the word slot is an element information category which needs to be collected by the robot from question or answer of the user in the task; the element information category includes: there are example keywords: enumeratable and standard names, and normalization processing is required when the elements are extracted; no example keywords: the instances in the word slot can not be enumerated, and are considered to be without instances, and normalization processing is not needed during element extraction; semantic type: the word slot has enumerable examples, but can not be directly normalized to a certain example through keywords; different types of word slot labeling methods include: when instance keywords are marked, words need to be fetched, word slots are selected, and instance standard names are selected; when the non-instance keywords are marked, words need to be fetched, word slots are selected, and instance standard names do not need to be selected; when the semantic keywords are labeled, words need to be taken, word slots are selected, and instance standard names do not need to be selected;
and carrying out accurate answer of the robot according to the extracted elements.
2. The robot customer service method of claim 1, wherein: the element extraction of the user questions of the received task knowledge points according to the pre-classified knowledge points comprises the following steps:
after receiving a question of a user, scoring the matching degree of each knowledge point in the knowledge base according to the text content in the question of the user, and finally returning the knowledge point which is most matched with the user;
and if the knowledge points belong to task-type knowledge points, the robot extracts corresponding elements from the user questions.
3. The robot customer service method of claim 1, wherein: before the robot accurately answers according to the extracted elements, normalization processing is carried out on the extracted elements to form standard names;
the normalization process includes: pre-creating and managing a word slot related in a task type scene, and an instance standard name and a plurality of aliases of an instance under the word slot; when the elements in the user question hit the alias of a certain instance or the similar alias with high confidence coefficient, the elements are normalized into the standard instance name, and then interface calling is carried out.
4. The robot customer service method of claim 3, wherein: the performing of the robot accurate answer according to the extracted elements includes: and the robot calls a service interface and accurately answers according to the element conditions.
5. A robot customer service system including element extraction, characterized in that:
the element extraction module is used for receiving question sentences asked by users; performing intention labeling on the question, wherein the intention labeling is to match the question to an existing knowledge point in a knowledge base from semantic dimensions, and the existing knowledge point comprises a task-type knowledge point and a common FAQ question-answer knowledge point; storing the classified knowledge points to a knowledge base; when the question of the user belongs to the common FAQ question-answer knowledge point, no element extraction is needed; when a user question is a task-type knowledge point, element extraction is required to be carried out in a word slot labeling mode; the word slot is an element information category which needs to be collected by the robot from question or answer of the user in the task; the element information category includes: there are example keywords: enumeratable and standard names, and normalization processing is required when the elements are extracted; no example keywords: the instances in the word slot can not be enumerated, and are considered to be without instances, and normalization processing is not needed during element extraction; semantic type: the word slot has enumerable examples, but can not be directly normalized to a certain example through keywords; different types of word slot labeling methods include: when instance keywords are marked, words need to be fetched, word slots are selected, and instance standard names are selected; when the non-instance keywords are marked, words need to be fetched, word slots are selected, and instance standard names do not need to be selected; when the semantic keywords are labeled, words need to be taken, word slots are selected, and instance standard names do not need to be selected;
and the service answer module is used for carrying out accurate robot answer according to the extracted elements.
6. The robotic customer service system according to claim 5, wherein: the element extraction module comprises:
the matching unit is used for scoring the matching degree of each knowledge point in the knowledge base according to the text content in the question of the user after receiving the question of the user and finally returning the most matched knowledge point to the user;
and the extracting unit is used for extracting corresponding elements from the user question by the robot if the knowledge points belong to the task-type knowledge points.
7. The robotic customer service system according to claim 5, wherein: the system also comprises a normalization module which is used for normalizing the extracted elements to form standard names before the robot accurately answers according to the extracted elements.
8. The robotic customer service system according to claim 5, wherein: and the service answer module is also used for calling a service interface by the robot and accurately answering according to the element conditions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711060950.3A CN107729549B (en) | 2017-10-31 | 2017-10-31 | Robot customer service method and system including element extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711060950.3A CN107729549B (en) | 2017-10-31 | 2017-10-31 | Robot customer service method and system including element extraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107729549A CN107729549A (en) | 2018-02-23 |
| CN107729549B true CN107729549B (en) | 2021-05-11 |
Family
ID=61221969
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711060950.3A Active CN107729549B (en) | 2017-10-31 | 2017-10-31 | Robot customer service method and system including element extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107729549B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110232573A (en) * | 2018-03-06 | 2019-09-13 | 广州供电局有限公司 | Based on interactive intelligent response system |
| CN109241269B (en) * | 2018-07-27 | 2020-07-17 | 深圳追一科技有限公司 | Task-based robot word slot filling method |
| CN109241251B (en) * | 2018-07-27 | 2022-05-27 | 众安信息技术服务有限公司 | Conversation interaction method |
| CN109992655B (en) * | 2019-03-29 | 2021-04-13 | 深圳追一科技有限公司 | Intelligent customer service method, device, equipment and storage medium |
| CN110378485B (en) * | 2019-06-03 | 2021-05-11 | 广东幽澜机器人科技有限公司 | Robot self-learning new business knowledge method and device |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104598445B (en) * | 2013-11-01 | 2019-05-10 | 腾讯科技（深圳）有限公司 | Automatically request-answering system and method |
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| CN104679910A (en) * | 2015-03-25 | 2015-06-03 | 北京智齿博创科技有限公司 | Intelligent answering method and system |
| CN105072173A (en) * | 2015-08-03 | 2015-11-18 | 谌志群 | Customer service method and system for automatically switching between automatic customer service and artificial customer service |
| CN106844335A (en) * | 2016-12-21 | 2017-06-13 | 海航生态科技集团有限公司 | Natural language processing method and device |
| CN106888259A (en) * | 2017-02-22 | 2017-06-23 | 镇江康恒信息科技有限公司 | A kind of method of online interaction question and answer |
| CN107239538B (en) * | 2017-06-01 | 2021-01-19 | 深圳追一科技有限公司 | Parallel customer service robot system with self-learning function and self-learning method thereof |
-
2017
- 2017-10-31 CN CN201711060950.3A patent/CN107729549B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107729549A (en) | 2018-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107729549B (en) | Robot customer service method and system including element extraction | |
| CN111984779B (en) | Dialogue text analysis method, device, equipment and readable medium | |
| CN107481720B (en) | Explicit voiceprint recognition method and device | |
| CN109727041A (en) | Intelligent customer service takes turns answering method, equipment, storage medium and device more | |
| CN109299320B (en) | Information interaction method and device, computer equipment and storage medium | |
| CN110597952A (en) | Information processing method, server, and computer storage medium | |
| CN109767318A (en) | Loan product recommended method, device, equipment and storage medium | |
| CN109508458B (en) | Legal entity identification method and device | |
| CN108305050B (en) | Method, device, equipment and medium for extracting report information and service demand information | |
| CN116737915B (en) | Semantic retrieval method, device, equipment and storage medium based on knowledge graph | |
| CN110399473B (en) | Method and device for determining answers to user questions | |
| US20220129491A1 (en) | Method for supporting agent and computing device thereof | |
| CN102402717A (en) | Data analysis facility and method | |
| CN115238688B (en) | Method, device, equipment and storage medium for analyzing association relation of electronic information data | |
| CN110880142A (en) | Risk entity acquisition method and device | |
| CN111222837A (en) | Intelligent interviewing method, system, equipment and computer storage medium | |
| CN110389941A (en) | Database method of calibration, device, equipment and storage medium | |
| CN112506981A (en) | Online training service pushing method and device | |
| CN109241075A (en) | Processing method, equipment and the computer readable storage medium of index basic data | |
| CN114186040A (en) | Operation method of intelligent robot customer service | |
| CN114186041A (en) | Answer output method | |
| CN109660621A (en) | Content pushing method and service equipment | |
| CN111309882B (en) | Method and device for realizing intelligent customer service question and answer | |
| CN117034949A (en) | Consultation work order classification method and device, electronic equipment and storage medium | |
| CN111488737B (en) | Text recognition method, device and equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |