CN111259128A - Method and device for generating conversation target sequence and readable storage medium - Google Patents
Method and device for generating conversation target sequence and readable storage medium Download PDFInfo
- Publication number
- CN111259128A CN111259128A CN202010057755.0A CN202010057755A CN111259128A CN 111259128 A CN111259128 A CN 111259128A CN 202010057755 A CN202010057755 A CN 202010057755A CN 111259128 A CN111259128 A CN 111259128A
- Authority
- CN
- China
- Prior art keywords
- conversation
- target
- input information
- target sequence
- conversation target
- 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 33
- 238000013527 convolutional neural network Methods 0.000 claims description 18
- 238000003062 neural network model Methods 0.000 claims description 11
- 238000010586 diagram Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011176 pooling Methods 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3346—Query execution using probabilistic model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a method and a device for generating a conversation target sequence and a readable storage medium, wherein the method comprises the following steps: acquiring input information at the current moment and a conversation target at the previous moment; judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target; and determining a conversation target sequence according to the judgment result. Therefore, by introducing the conversation object sequence containing a plurality of conversation objects, the conversation system can simultaneously execute the plurality of conversation objects in the conversation object sequence, so that the limitation of single-object open-domain conversation is solved, and the experience of a user is improved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a device for generating a conversation target sequence and a readable storage medium.
Background
Although the single-target open domain dialog has diversity and consistency of dialog responses and satisfaction of users to a certain degree, the single-target open domain dialog is distinguished from real human-computer interaction, which generally involves multiple targets, and under the circumstance, the single-target open domain dialog is introduced, so that the user experience is reduced.
Disclosure of Invention
The embodiment of the invention provides a method and a device for generating a conversation target sequence and a readable storage medium, which are used for executing a plurality of conversation targets in the conversation target sequence, so that the limitation of single-target open domain conversation is solved, and the experience of a user is improved.
One aspect of the present invention provides a method for generating a dialog target sequence, where the method includes: acquiring input information at the current moment and a conversation target at the previous moment; judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target; and determining a conversation target sequence according to the judgment result.
In an implementation manner, the determining, according to the acquired input information and the dialog target, whether the dialog target corresponding to the input information is transferred includes: taking the input information and the dialogue target at the previous moment as input of a neural network model to obtain a probability value; and judging whether the conversation target is transferred or not according to the obtained probability value.
In one embodiment, the neural network model is a first convolutional neural network.
In an implementation manner, whether the conversation target is transferred or not is judged according to the obtained probability value; if the probability value is judged to exceed the probability threshold value, judging that the conversation target is transferred; and if the probability value is judged not to exceed the probability threshold value, judging that the conversation target is not transferred.
In an embodiment, the determining the dialog target sequence according to the determination result includes: if the conversation target is judged to be transferred, obtaining the conversation target at the current moment according to the historical input information and the historical reply information; acquiring a conversation target sequence at the previous moment; and adding the conversation target at the current moment in the conversation target sequence at the previous moment to generate the conversation target sequence at the current moment.
In an embodiment, the determining the dialog target sequence according to the determination result includes: if the conversation target is judged not to be transferred, acquiring a conversation target sequence at the previous moment; and taking the conversation target sequence at the previous moment as the conversation target sequence at the current moment.
In an implementation manner, the obtaining a dialog target at a current time according to the history input information and the history reply information includes: and taking the historical input information and the historical reply information as the input of a second convolutional neural network to obtain the conversation target at the current moment.
Another aspect of the present invention provides an apparatus for generating a dialog target sequence, the apparatus comprising: the target transfer acquisition module is used for acquiring input information at the current moment and a conversation target at the previous moment; the target transfer judging module is used for judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target; and the target sequence generation module is used for determining the conversation target sequence according to the judgment result.
In an implementation manner, the target sequence generation module is specifically configured to: if the conversation target is judged to be transferred, obtaining the conversation target at the current moment according to the historical input information and the historical reply information; acquiring a conversation target sequence at the previous moment; and adding the conversation target at the current moment in the conversation target sequence at the previous moment to generate the conversation target sequence at the current moment.
Another aspect of the invention provides a computer-readable storage medium comprising a set of computer-executable instructions that, when executed, perform a method of generating a dialog target sequence.
In the embodiment of the invention, the input information at the current moment and the conversation target at the previous moment are obtained; judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target; and determining a conversation target sequence according to the judgment result.
Therefore, by introducing the conversation object sequence containing a plurality of conversation objects, the conversation system can simultaneously execute the plurality of conversation objects in the conversation object sequence, so that the limitation of single-object open-domain conversation is solved, and the experience of a user is improved.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
Fig. 1 is a schematic flow chart illustrating an implementation of a method for generating a dialog target sequence according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a specific implementation of a method for generating a dialog target sequence according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a dialog target sequence generation apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart illustrating an implementation of a method for generating a dialog target sequence according to an embodiment of the present invention.
As shown in fig. 1, an aspect of the present invention provides a method for generating a dialog target sequence, where the method includes:
and 103, determining a conversation target sequence according to the judgment result.
In this embodiment, in the process of chatting with the robot, the information input by the user may or may not include one or more conversation objects. In dialog systems, dialog state tracking is often utilized to capture and record the intent of each input statement or dialog target of a user so that the dialog system generates reply information based on the resulting dialog target.
The dialog target sequence is generated as follows:
firstly, acquiring input information at the current moment and a dialogue target at the previous moment, wherein the input information is text information and can be input by a user or the voice of the user can be converted into the text information; the dialog target at the previous moment is obtained by the dialog system according to the input information at the previous moment.
And then judging whether the dialogue target corresponding to the input information is transferred or not according to the input information at the current moment and the dialogue target at the previous moment.
And finally, determining a conversation target sequence according to the judgment result. The dialog target sequence is an ordered sequence formed by a plurality of dialog targets, and the dialog system executes corresponding operation or service according to the dialog target sequence.
Therefore, by introducing the conversation object sequence containing a plurality of conversation objects, the conversation system can simultaneously execute the plurality of conversation objects in the conversation object sequence, so that the limitation of single-object open-domain conversation is solved, and the experience of a user is improved.
In an implementation manner, the determining whether the dialog target corresponding to the input information is shifted according to the acquired input information and the dialog target includes:
taking the input information and the dialogue target at the previous moment as the input of a neural network model to obtain a probability value;
and judging whether the conversation target is transferred or not according to the obtained probability value.
In this embodiment, the specific process of step 102 is: the input information of the current moment and the dialogue target of the previous moment are used as the input of the neural network model together, the probability value is obtained through the neural network model, and finally whether the dialogue target corresponding to the input information is transferred or not is judged according to the obtained probability value.
The neural network model needs to be trained to accurately judge whether the transfer occurs, and the training process is approximate: and training a large number of sentences and corresponding dialogue targets as linguistic data, adjusting the weight in the neural network model by using a loss function, and repeatedly iterating for a specified number of times until the loss is within a specified range to represent that the training is finished.
In one possible embodiment, the neural network model is a first convolutional neural network.
In this embodiment, the neural network model is a convolutional neural network, and before inputting the input information and the dialogue target into the convolutional neural network, each word of the input information and the dialogue target needs to be converted into a corresponding word vector by a word vector tool, and then the converted word vector is input into the convolutional neural network.
And expressing the word vector as a matrix of n rows and k columns, sequentially passing through a convolutional layer, a pooling layer and a full-link layer in the convolutional neural network, and finally outputting a probability value.
In an implementation mode, whether the conversation target is transferred or not is judged according to the obtained probability value;
if the probability value exceeds the probability threshold value, judging that the conversation target is transferred;
and if the probability value is judged not to exceed the probability threshold value, judging that the conversation target is not transferred.
In this embodiment, the probability threshold is manually specified in advance, and the specific process of "determining whether the dialog target is transferred according to the obtained probability value" includes:
and if the obtained probability value exceeds the probability threshold value, the fact that the conversation target corresponding to the current input information is inconsistent with the conversation target at the previous moment is indicated, and the conversation target is judged to be transferred.
Otherwise, if the obtained probability does not exceed the probability threshold, the conversation target corresponding to the current input information is consistent with the conversation target at the previous moment, and the conversation target is judged not to be transferred.
In an embodiment, determining the dialog target sequence according to the determination result includes:
if the conversation target is judged to be transferred, obtaining the conversation target at the current moment according to the historical input information and the historical reply information;
acquiring a conversation target sequence at the previous moment;
and adding the conversation target at the current moment in the conversation target sequence at the previous moment to generate the conversation target sequence at the current moment.
In this embodiment, the specific process of step 103 is: and if the conversation target corresponding to the current input information and the conversation target at the previous moment are judged to be transferred through the first convolutional neural network, acquiring the conversation target corresponding to the current input information according to the historical input information input by the user and the historical reply information replied by the conversation system.
And then acquiring the conversation target sequence at the previous moment, and newly adding the conversation target at the current moment in the original conversation target sequence to obtain the conversation target sequence at the current moment.
In an embodiment, determining the dialog target sequence according to the determination result and according to the determination result includes:
if the conversation target is judged not to be transferred, acquiring a conversation target sequence at the previous moment;
and taking the conversation target sequence at the previous moment as the conversation target sequence at the current moment.
In this embodiment, when it is determined that the current session target is not transferred through the first convolutional neural network, a session target sequence at a previous time is obtained, and the session target sequence is used as the session target sequence at the current time.
In an implementation manner, obtaining the dialog target at the current moment according to the historical input information and the historical reply information includes:
and taking the historical input information and the historical reply information as the input of the second convolutional neural network to obtain the conversation target at the current moment.
In this embodiment, the specific process of "obtaining the dialog target at the current time according to the history input information and the history reply information" in the above steps is as follows: and taking all historical input information and reply information as the input of the second convolutional neural network to obtain the conversation target at the current moment.
Wherein, the second neural network is specially used for receiving the word vector corresponding to the input sentence and outputting the dialogue target, and the training process is consistent with the steps of training the first neural network, and is not explained again here.
Fig. 2 is a schematic flowchart of a specific implementation of a method for generating a dialog target sequence according to an embodiment of the present invention, as shown in fig. 2.
In the figure, X represents a word vector matrix corresponding to an input sentence, CNN represents a convolutional neural network, gc represents a dialogue target, gt-1 represents a dialogue target at a previous time, and gt represents a dialogue target at a current time.
Firstly, the word vector matrix and the dialogue target are used as the input of a first convolutional neural network to carry out dialogue target estimation, and Pgc (1| X, gt-1), namely the dialogue transfer probability between the dialogue target at the previous moment and the current input statement, is obtained.
When Pgc <0.5, it means that the dialogue object has not been transferred, gc is the dialogue object at the previous time, otherwise, the dialogue information of the user and the system is used as the input of the second convolutional neural network to transfer the dialogue object, and a new dialogue object is obtained, at this time, gc is the dialogue object at the current time.
Each time a dialog object is obtained, it is added to the dialog object sequence to obtain a dialog object sequence (not shown in the process diagram).
Fig. 3 is a schematic structural diagram of a dialog target sequence generation apparatus according to an embodiment of the present invention.
Another aspect of the present invention provides an apparatus for generating a dialog target sequence, the apparatus comprising:
a target transfer obtaining module 201, configured to obtain input information at a current time and a session target at a previous time;
the target transfer judging module 202 is configured to judge whether a conversation target corresponding to the input information is transferred according to the acquired input information and the conversation target;
and the target sequence generation module 203 is used for determining a conversation target sequence according to the judgment result.
In this embodiment, first, the target transfer obtaining module 201 obtains input information at the current time and a dialog target at the previous time, where the input information is text information and can be input by a user, or voice of the user can be converted into text information; the dialog target at the previous moment is obtained by the dialog system according to the input information at the previous moment.
Next, the target transfer determining module 202 determines whether the dialog target corresponding to the input information is transferred according to the input information at the current time and the dialog target at the previous time.
And finally, determining a conversation target sequence through the target sequence generation module 203 according to the judgment result. The dialog target sequence is an ordered sequence formed by a plurality of dialog targets, and the dialog system executes corresponding operation or service according to the dialog target sequence.
Therefore, by introducing the conversation object sequence containing a plurality of conversation objects, the conversation system can simultaneously execute the plurality of conversation objects in the conversation object sequence, so that the limitation of single-object open-domain conversation is solved, and the experience of a user is improved.
In an implementation manner, the target sequence generating module 203 is specifically configured to:
if the conversation target is judged to be transferred, obtaining the conversation target at the current moment according to the historical input information and the historical reply information;
acquiring a conversation target sequence at the previous moment;
and adding the conversation target at the current moment in the conversation target sequence at the previous moment to generate the conversation target sequence at the current moment.
In this embodiment, the target sequence generating module 203 is specifically configured to: and if the conversation target corresponding to the current input information and the conversation target at the previous moment are judged to be transferred through the first convolutional neural network, acquiring the conversation target corresponding to the current input information according to the historical input information input by the user and the historical reply information replied by the conversation system.
And then acquiring the conversation target sequence at the previous moment, and newly adding the conversation target at the current moment in the original conversation target sequence to obtain the conversation target sequence at the current moment.
Another aspect of the invention provides a computer-readable storage medium comprising a set of computer-executable instructions which, when executed, perform any one of the above-described methods of generating a dialog object sequence.
In an embodiment of the present invention, a computer-readable storage medium includes a set of computer-executable instructions, which when executed, are configured to obtain input information at a current time and a dialog target at a previous time; judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target; and determining a conversation target sequence according to the judgment result.
Therefore, by introducing the conversation object sequence containing a plurality of conversation objects, the conversation system can simultaneously execute the plurality of conversation objects in the conversation object sequence, so that the limitation of single-object open-domain conversation is solved, and the experience of a user is improved.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
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 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 claims.
Claims (10)
1. A method for generating a dialog target sequence, the method comprising:
acquiring input information at the current moment and a conversation target at the previous moment;
judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target;
and determining a conversation target sequence according to the judgment result.
2. The method according to claim 1, wherein the determining whether the dialog target corresponding to the input information is shifted according to the acquired input information and the dialog target comprises:
taking the input information and the dialogue target at the previous moment as input of a neural network model to obtain a probability value;
and judging whether the conversation target is transferred or not according to the obtained probability value.
3. The method of claim 2, wherein the neural network model is a first convolutional neural network.
4. The method of claim 2, wherein the determining whether the dialog target is transferred is performed according to the obtained probability value;
if the probability value is judged to exceed the probability threshold value, judging that the conversation target is transferred;
and if the probability value is judged not to exceed the probability threshold value, judging that the conversation target is not transferred.
5. The method according to claim 1 or 4, wherein the determining a dialog target sequence according to the determination result comprises:
if the conversation target is judged to be transferred, obtaining the conversation target at the current moment according to the historical input information and the historical reply information;
acquiring a conversation target sequence at the previous moment;
and adding the conversation target at the current moment in the conversation target sequence at the previous moment to generate the conversation target sequence at the current moment.
6. The method of claim 5, wherein determining the dialog target sequence based on the determination comprises:
if the conversation target is judged not to be transferred, acquiring a conversation target sequence at the previous moment;
and taking the conversation target sequence at the previous moment as the conversation target sequence at the current moment.
7. The method of claim 5, wherein obtaining the dialog target at the current time according to the historical input information and the historical reply information comprises:
and taking the historical input information and the historical reply information as the input of a second convolutional neural network to obtain the conversation target at the current moment.
8. An apparatus for generating a dialog target sequence, the apparatus comprising:
the target transfer acquisition module is used for acquiring input information at the current moment and a conversation target at the previous moment;
the target transfer judging module is used for judging whether the conversation target corresponding to the input information is transferred or not according to the acquired input information and the conversation target;
and the target sequence generation module is used for determining the conversation target sequence according to the judgment result.
9. The apparatus of claim 8, wherein the target sequence generation module is specifically configured to:
if the conversation target is judged to be transferred, obtaining the conversation target at the current moment according to the historical input information and the historical reply information;
acquiring a conversation target sequence at the previous moment;
and adding the conversation target at the current moment in the conversation target sequence at the previous moment to generate the conversation target sequence at the current moment.
10. A computer-readable storage medium comprising a set of computer-executable instructions which, when executed, perform a method of generating a dialog object sequence according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010057755.0A CN111259128A (en) | 2020-01-19 | 2020-01-19 | Method and device for generating conversation target sequence and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010057755.0A CN111259128A (en) | 2020-01-19 | 2020-01-19 | Method and device for generating conversation target sequence and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111259128A true CN111259128A (en) | 2020-06-09 |
Family
ID=70954203
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010057755.0A Pending CN111259128A (en) | 2020-01-19 | 2020-01-19 | Method and device for generating conversation target sequence and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111259128A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105704013A (en) * | 2016-03-18 | 2016-06-22 | 北京光年无限科技有限公司 | Context-based topic updating data processing method and apparatus |
| CN106294854A (en) * | 2016-08-22 | 2017-01-04 | 北京光年无限科技有限公司 | A kind of man-machine interaction method for intelligent robot and device |
| CN108268443A (en) * | 2017-12-21 | 2018-07-10 | 北京百度网讯科技有限公司 | It determines the transfer of topic point and obtains the method, apparatus for replying text |
| CN108268616A (en) * | 2018-01-04 | 2018-07-10 | 中国科学院自动化研究所 | The controllability dialogue management extended method of fusion rule information |
| CN109815319A (en) * | 2018-12-24 | 2019-05-28 | 联想(北京)有限公司 | Information processing method and information processing unit |
| CN110309170A (en) * | 2019-07-02 | 2019-10-08 | 北京大学 | A kind of Task takes turns the complicated intension recognizing method in dialogue more |
-
2020
- 2020-01-19 CN CN202010057755.0A patent/CN111259128A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105704013A (en) * | 2016-03-18 | 2016-06-22 | 北京光年无限科技有限公司 | Context-based topic updating data processing method and apparatus |
| CN106294854A (en) * | 2016-08-22 | 2017-01-04 | 北京光年无限科技有限公司 | A kind of man-machine interaction method for intelligent robot and device |
| CN108268443A (en) * | 2017-12-21 | 2018-07-10 | 北京百度网讯科技有限公司 | It determines the transfer of topic point and obtains the method, apparatus for replying text |
| CN108268616A (en) * | 2018-01-04 | 2018-07-10 | 中国科学院自动化研究所 | The controllability dialogue management extended method of fusion rule information |
| CN109815319A (en) * | 2018-12-24 | 2019-05-28 | 联想(北京)有限公司 | Information processing method and information processing unit |
| CN110309170A (en) * | 2019-07-02 | 2019-10-08 | 北京大学 | A kind of Task takes turns the complicated intension recognizing method in dialogue more |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111309889B (en) | Method and device for text processing | |
| CN109460463B (en) | Model training method, device, terminal and storage medium based on data processing | |
| CN106503236B (en) | Artificial intelligence based problem classification method and device | |
| CN108446322A (en) | A kind of implementation method and device of intelligent Answer System | |
| CN108304679A (en) | A kind of adaptive reliability analysis method | |
| CN111275780B (en) | Character image generation method and device | |
| EP3443450A1 (en) | Two-stage training of a spoken dialogue system | |
| CN111125328B (en) | Text processing method and related equipment | |
| CN111191450A (en) | Corpus cleaning method, corpus entry device and computer-readable storage medium | |
| CN105046366A (en) | Model training method and device | |
| CN111666393A (en) | Verification method and device of intelligent question-answering system, computer equipment and storage medium | |
| CN111400466A (en) | Intelligent dialogue method and device based on reinforcement learning | |
| CN111124898B (en) | Question-answering system testing method and device, computer equipment and storage medium | |
| CN113726545A (en) | Network traffic generation method and device for generating countermeasure network based on knowledge enhancement | |
| CN113642652A (en) | Method, device and equipment for generating fusion model | |
| CN114882307A (en) | Classification model training and image feature extraction method and device | |
| CN110689359A (en) | Method and device for dynamically updating model | |
| CN111259128A (en) | Method and device for generating conversation target sequence and readable storage medium | |
| CN111274374B (en) | Data processing method and device, computer storage medium and electronic equipment | |
| CN116017528A (en) | Traffic prediction method, traffic prediction device and server | |
| CN116795971A (en) | Man-machine dialogue scene construction system based on generated language model | |
| KR20240034804A (en) | Evaluating output sequences using an autoregressive language model neural network | |
| Yoshino et al. | Statistical dialogue management using intention dependency graph | |
| CN110222161B (en) | Intelligent response method and device for conversation robot | |
| CN108536811B (en) | Voice interaction path determining method and device based on machine learning, storage medium and terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200609 |