CN107025283A - The answer method and system of candidate answers sequence are carried out based on subscriber data - Google Patents
The answer method and system of candidate answers sequence are carried out based on subscriber data Download PDFInfo
- Publication number
- CN107025283A CN107025283A CN201710218177.2A CN201710218177A CN107025283A CN 107025283 A CN107025283 A CN 107025283A CN 201710218177 A CN201710218177 A CN 201710218177A CN 107025283 A CN107025283 A CN 107025283A
- Authority
- CN
- China
- Prior art keywords
- user
- answer
- information
- inventory
- sequence
- 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 63
- 238000013135 deep learning Methods 0.000 claims description 28
- 238000010801 machine learning Methods 0.000 claims description 26
- 230000004044 response Effects 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 101100075747 Drosophila melanogaster Lztr1 gene Proteins 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000013456 study Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The answer method and system that candidate answers sequence is carried out based on subscriber data that the present invention is provided, method is：Obtain the text message of user's input；The text message inputted according to user, alternative answer inventory is obtained by the corpus pre-established；The personal information of user is obtained, alternative answer inventory is ranked up, the answer inventory after being sorted；According to the answer inventory after sequence, the answer that user inputs text message is obtained.The answer method and system that candidate answers sequence is carried out based on subscriber data of the present invention, employ the mode being ranked up based on userspersonal information to candidate answers, the answer closer to user itself is given, so as to reduce the error rate of answer, Consumer's Experience is improved.
Description
Technical field
The present invention relates to artificial intelligence field, more particularly to interactive system field.
Background technology
In existing artificial intelligence conversational system, using corpus generation generation candidate answers, then to these candidates
Answer is ranked up to select suitable reply.For the sort method of candidate answers, in the prior art by calculating user
The degree of correlation between the text message of input and answer, so as to select suitable candidate answers as the text of answer user's input
The answer of information.But, this sort method lacks the understanding for subscriber data, so erroneous answers are easily caused, or
It is not close to the users.
Therefore, defect of the prior art is that in artificial intelligence dialog procedure, robot is unable to base to the answer of user
Intelligent response is made in the feature of individual subscriber, the erroneous answers that robot is provided is easily caused, makes user experience low.
The content of the invention
For above-mentioned technical problem, the present invention provides a kind of answer method that candidate answers sequence is carried out based on subscriber data
And system, the mode being ranked up based on userspersonal information to candidate answers is employed, is given closer to user's itself
Answer, so as to reduce the error rate of answer, improve Consumer's Experience.
In order to solve the above technical problems, the technical scheme that the present invention is provided is：
In a first aspect, the present invention provides a kind of answer method that candidate answers sequence is carried out based on subscriber data, including：
Step S1, obtains the text message of user's input；
Step S2, the text message inputted according to the user obtains alternative answer by the corpus pre-established
Inventory；
Step S3, obtains the personal information of the user, the alternative answer inventory is ranked up, after being sorted
Answer inventory；
Step S4, according to the answer inventory after the sequence, obtains the answer that the user inputs text message.
The answer method that candidate answers sequence is carried out based on subscriber data that the present invention is provided, its technical scheme is：Obtain
The text message of user's input；The text message inputted according to the user, obtains alternative by the corpus pre-established
Answer inventory；The personal information of the user is obtained, the alternative answer inventory is ranked up, the answer after being sorted is clear
It is single；According to the answer inventory after the sequence, the answer that the user inputs text message is obtained.
The answer method that candidate answers sequence is carried out based on subscriber data of the present invention, is employed based on userspersonal information
The mode being ranked up to candidate answers, gives the answer closer to user itself, so as to reduce the error rate of answer, carries
High Consumer's Experience.
Further, the personal information of the user includes user preference information, user's geographical location information, user's birthday
Age information, user gender information, user's education background information, user job information and user's habits and customs information.
Further, the step S3, be specially：
Obtain the personal information of the user；
The text message that is inputted by the method for statistical learning, machine learning and deep learning to the user and user's
Personal information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is ranked up by rule-based method, arranged
Answer inventory after sequence.
Further, the step S3, be specially：
Obtain the personal information of the user；
The text message that is inputted by the method for statistical learning, machine learning, deep learning to the user and user's
Personal information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is arranged by the method for machine learning and deep learning
Sequence, the answer inventory after being sorted.
Second aspect, the invention provides a kind of answer system that candidate answers sequence is carried out based on subscriber data, including：
Text message acquisition module, the text message for obtaining user's input；
Alternative answer checklist module, for the text message inputted according to the user, passes through the language material pre-established
Storehouse obtains alternative answer inventory；
Alternative answer inventory order module, the personal information for obtaining the user is entered to the alternative answer inventory
Row sequence, the answer inventory after being sorted；
Response means, for according to the answer inventory after the sequence, obtaining the answer that the user inputs text message.
A kind of answer system that candidate answers sequence is carried out based on subscriber data that the present invention is provided, its technical scheme is：
By text message acquisition module, the text message of user's input is obtained；By alternative answer checklist module, according to the user
The text message of input, alternative answer inventory is obtained by the corpus pre-established；Pass through alternative answer inventory sequence mould
Block, obtains the personal information of the user, the alternative answer inventory is ranked up, the answer inventory after being sorted；It is logical
Response means are crossed, according to the answer inventory after the sequence, the answer that the user inputs text message are obtained.
The answer system that candidate answers sequence is carried out based on subscriber data of the present invention, is employed based on userspersonal information
The mode being ranked up to candidate answers, gives the answer closer to user itself, so as to reduce the error rate of answer, carries
High Consumer's Experience.
Further, the personal information of the user includes user preference information, user's geographical location information, user's birthday
Age information, user gender information, user's education background information, user job information and user's habits and customs information.
Further, the alternative answer inventory order module, specifically for：
Obtain the personal information of the user；
The text message that is inputted by the method for statistical learning, machine learning and deep learning to the user and user's
Personal information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is ranked up by rule-based method, arranged
Answer inventory after sequence.
Further, the alternative answer inventory order module, specifically for：
Obtain the personal information of the user；
The text message that is inputted by the method for statistical learning, machine learning, deep learning to the user and user's
Personal information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is arranged by the method for machine learning and deep learning
Sequence, the answer inventory after being sorted.
Brief description of the drawings
, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical scheme of the prior art
The accompanying drawing used required in embodiment or description of the prior art is briefly described.
Fig. 1 shows a kind of answer side that candidate answers sequence is carried out based on subscriber data that the embodiment of the present invention is provided
The flow chart of method；
Fig. 2 shows a kind of answer system that candidate answers sequence is carried out based on subscriber data that the embodiment of the present invention is provided
The schematic diagram of system.
Embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Clearly illustrate technical scheme, therefore be intended only as example, and the protection of the present invention can not be limited with this
Scope.
Embodiment one
Fig. 1 shows a kind of answer side that candidate answers sequence is carried out based on subscriber data that the embodiment of the present invention is provided
The flow chart of method；As shown in figure 1, a kind of answer side that candidate answers sequence is carried out based on subscriber data that the present embodiment is provided
Method, including：
Step S1, obtains the text message of user's input；
Step S2, the text message inputted according to user obtains alternative answer inventory by the corpus pre-established；
Step S3, obtains the personal information of user, alternative answer inventory is ranked up, the answer after being sorted is clear
It is single；
Step S4, according to the answer inventory after sequence, obtains the answer that user inputs text message.
The answer method that candidate answers sequence is carried out based on subscriber data that the present invention is provided, its technical scheme is：Obtain
The text message of user's input；The text message inputted according to user, alternative answer is obtained by the corpus pre-established
Inventory；The personal information of user is obtained, alternative answer inventory is ranked up, the answer inventory after being sorted；According to sequence
Answer inventory afterwards, obtains the answer that user inputs text message.
The answer method that candidate answers sequence is carried out based on subscriber data of the present invention, is employed based on userspersonal information
The mode being ranked up to candidate answers, gives the answer closer to user itself, so as to reduce the error rate of answer, carries
High Consumer's Experience.
Wherein, the personal information of user is believed including user preference information, user's geographical location information, user's age birthday
Breath, user gender information, user's education background information, user job information and user's habits and customs information.
The invention is not restricted to these above-mentioned personal information on user, include the constellation information of user, often purchase
Clothes information etc., these information are created as user database, when user inputs text message, user's money are directly invoked
Expect the personal information of the user in storehouse, the alternative answer in alternative answer inventory is entered using the one or more in these information
Row sequence, therefrom selects suitable answer.It is ranked up with the personal information of user, can not only considers that answer is inputted with user
The degree of correlation, and the data of user in itself can be considered, so the life that the answer of generation can more be close to the users, so as to deepen
Interaction and emotion with user are connected.
Wherein, the text message of user's input includes the dialog information that user inputs, and dialog history information.
Wherein, user database can be set up by the input of user or the personal information of inquiry user.
Specifically, step S3, be specially：
Obtain the personal information of user；
The personal information of user is modeled by the method for statistical learning, machine learning and deep learning, spy is obtained
Reference ceases；
According to characteristic information, alternative answer inventory is ranked up by rule-based method, answering after being sorted
Case inventory.
Wherein, rule is formulated according to the preferred answer type of different user types.
The personal information of user is modeled by the method for statistical learning, machine learning and deep learning, passes through system
The model that the method for meter study, machine learning and deep learning is set up, the information of user database for making foundation more comprehensively, passes through
Rule-based method is ranked up to alternative answer inventory, can effectively exclude with the incongruent alternative answer of subscriber data,
Make the answer that robot is provided closer to the life of user.
Specifically, step S3, be specially：
Obtain the personal information of user；
The personal information of user is modeled by the method for statistical learning, machine learning, deep learning, feature is obtained
Information；
According to characteristic information, alternative answer inventory is ranked up by the method for machine learning and deep learning, obtained
Answer inventory after sequence.
Alternative answer inventory can also be ranked up by the method for machine learning and deep learning, or with based on rule
Method then can be excluded effectively and subscriber data is incongruent alternatively answers with reference to alternative answer inventory is ranked up
Case, makes the answer that robot is provided closer to the life of user.
Wherein, characteristic information represents user characteristics, can there is two kinds of expression-forms, and one kind is word/category feature, is used for
Rule-based sort method, another is vector characteristics, for the sort method based on deep learning.
For example, character features are a kind of descriptions carried out using word for user, illustrate " night owl ", " geek " " learns
Despot ".
Wherein, obtained by the method for statistical learning, machine learning, deep learning after characteristic information (user characteristics), will
User characteristics vector, and alternative answer input a deep neural network, include but is not limited to following several forms：
1), CNN (Convolutional Neural Networks, convolutional neural networks)；
2), LSTM (Long Short Term, length Memory Neural Networks)；
3) LSTM, based on notice mechanism, and the scoring to this answer is exported, in this, as sort by answer
It is ranked up.
Fig. 2 shows a kind of answer system that candidate answers sequence is carried out based on subscriber data that the embodiment of the present invention is provided
The schematic diagram of system.As shown in Fig. 2 present embodiments providing a kind of answer system that candidate answers sequence is carried out based on subscriber data
10, including：
Text message acquisition module 101, the text message for obtaining user's input；
Alternative answer checklist module 102, for the text message inputted according to user, passes through the corpus pre-established
Obtain alternative answer inventory；
Alternative answer inventory order module 103, the personal information for obtaining user is arranged alternative answer inventory
Sequence, the answer inventory after being sorted；
Response means 104, for according to the answer inventory after sequence, obtaining the answer that user inputs text message.
A kind of answer system 10 that candidate answers sequence is carried out based on subscriber data that the present invention is provided, its technical scheme
For：By text message acquisition module 101, the text message of user's input is obtained；Pass through alternative answer checklist module 102, root
The text message inputted according to user, alternative answer inventory is obtained by the corpus pre-established；Pass through alternative answer inventory
Order module 103, obtains the personal information of user, alternative answer inventory is ranked up, the answer inventory after being sorted；It is logical
Response means 104 are crossed, according to the answer inventory after sequence, the answer that user inputs text message are obtained.
The answer system 10 that candidate answers sequence is carried out based on subscriber data of the present invention, is employed based on individual subscriber letter
The mode being ranked up to candidate answers is ceased, the answer closer to user itself is given, so that the error rate of answer is reduced,
Improve Consumer's Experience.
Wherein, the personal information of user is believed including user preference information, user's geographical location information, user's age birthday
Breath, user gender information, user's education background information, user job information and user's habits and customs information.
The invention is not restricted to these above-mentioned personal information on user, include the constellation information of user, often purchase
Clothes information etc., these information are created as user database, when user inputs text message, user's money are directly invoked
Expect the personal information of the user in storehouse, the alternative answer in alternative answer inventory is entered using the one or more in these information
Row sequence, therefrom selects suitable answer.It is ranked up with the personal information of user, can not only considers that answer is inputted with user
The degree of correlation, and the data of user in itself can be considered, so the life that the answer of generation can more be close to the users, so as to deepen
Interaction and emotion with user are connected.
Wherein, the text message of user's input includes the dialog information that user inputs, and dialog history information.
Wherein, user database can be set up by the input of user or the personal information of inquiry user.
Specifically, alternative answer inventory order module 103, specifically for：
Obtain the personal information of user；
The personal information of user is modeled by the method for statistical learning, machine learning and deep learning, spy is obtained
Reference ceases；
According to characteristic information, alternative answer inventory is ranked up by rule-based method, answering after being sorted
Case inventory.
Wherein, rule is formulated according to the preferred answer type of different user types.
The personal information of user is modeled by the method for statistical learning, machine learning and deep learning, passes through system
The model that the method for meter study, machine learning and deep learning is set up, the information of user database for making foundation more comprehensively, passes through
Rule-based method is ranked up to alternative answer inventory, can effectively exclude with the incongruent alternative answer of subscriber data,
Make the answer that robot is provided closer to the life of user.
Specifically, alternative answer inventory order module 103, specifically for：
Obtain the personal information of user；
The personal information of user is modeled by the method for statistical learning, machine learning, deep learning, feature is obtained
Information；
According to characteristic information, alternative answer inventory is ranked up by the method for machine learning and deep learning, obtained
Answer inventory after sequence.
Alternative answer inventory can also be ranked up by the method for machine learning and deep learning, or with based on rule
Method then can be excluded effectively and subscriber data is incongruent alternatively answers with reference to alternative answer inventory is ranked up
Case, makes the answer that robot is provided closer to the life of user.
Wherein, characteristic information represents user characteristics, can there is two kinds of expression-forms, and one kind is word/category feature, is used for
Rule-based sort method, another is vector characteristics, for the sort method based on deep learning.
For example, character features are a kind of descriptions carried out using word for user, illustrate " night owl ", " geek " " learns
Despot ".
Wherein, obtained by the method for statistical learning, machine learning, deep learning after characteristic information (user characteristics), will
User characteristics vector, and alternative answer input a deep neural network, include but is not limited to following several forms：
1), CNN (Convolutional Neural Networks, convolutional neural networks)；
2), LSTM (Long Short Term, length Memory Neural Networks)；
3) LSTM, based on notice mechanism, and the scoring to this answer is exported, in this, as sort by answer
It is ranked up.
Embodiment two
A kind of answer method that candidate answers sequence is carried out based on subscriber data provided based on embodiment one, and based on use
Family data carries out the answer system 10 of candidate answers sequence, is illustrated this gives specific embodiment.
Alternative answer is ranked up by the personal information of user, can effectively be excluded incongruent standby with subscriber data
Select answer,
For example：In the personal information of user, age of user is 10 years old, then the text message that is inputted for user " on
Learn well tired ".
If not combining the personal information of user, the answer that may be provided is：" yes, university's school work is very heavy " this is obvious
The reply of mistake.
Also include the habits and customs of user in the personal information of user, therefore, the understanding to user's habits and customs, Ke Yijia
The deep care to user, life of being more close to the users.
For example：The habits and customs of some user are not late sleep, then the text message inputted for user：" evening now
12 points ".
With reference to the personal habits of user, system provides reply：" you are generally not at 10 points and just sleptWhether what is had
Worry is so can't fall asleep",
The reply provided compared to the personal habits for not combining user：" want to eat food", by contrast, for simultaneously
There is no evening to sleep the user of custom, be more intimate answer with reference to the reply that provides of personal habits of user.
Therefore, a kind of subscriber data that is based on that the present invention is provided carries out the answer method of candidate answers sequence and is
System, is employed the mode being ranked up based on userspersonal information to candidate answers, gives the answer closer to user itself,
So as to reduce the error rate of answer, Consumer's Experience is improved.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent
The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that：Its according to
The technical scheme described in foregoing embodiments can so be modified, or which part or all technical characteristic are entered
Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme, it all should cover among the claim of the present invention and the scope of specification.
Claims (8)
1. the answer method of candidate answers sequence is carried out based on subscriber data, it is characterised in that including：
Step S1, obtains the text message of user's input；
Step S2, the text message inputted according to the user obtains alternative answer inventory by the corpus pre-established；
Step S3, obtains the personal information of the user, the alternative answer inventory is ranked up, the answer after being sorted
Inventory；
Step S4, according to the answer inventory after the sequence, obtains the answer that the user inputs text message.
2. the answer method according to claim 1 that candidate answers sequence is carried out based on subscriber data, it is characterised in that
The personal information of the user includes user preference information, user's geographical location information, user's birthday age information, user
Gender information, user's education background information, user job information and user's habits and customs information.
3. the answer method according to claim 1 that candidate answers sequence is carried out based on subscriber data, it is characterised in that
The step S3, be specially：
Obtain the personal information of the user；
The text message inputted by the method for statistical learning, machine learning and deep learning to the user and the individual of user
Information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is ranked up by rule-based method, obtained after sequence
Answer inventory.
4. the answer method according to claim 1 that candidate answers sequence is carried out based on subscriber data, it is characterised in that
The step S3, be specially：
Obtain the personal information of the user；
The text message inputted by the method for statistical learning, machine learning, deep learning to the user and the individual of user
Information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is ranked up by the method for machine learning and deep learning,
Answer inventory after being sorted.
5. the answer system of candidate answers sequence is carried out based on subscriber data, it is characterised in that
Text message acquisition module, the text message for obtaining user's input；
Alternative answer checklist module, for the text message inputted according to the user, is obtained by the corpus pre-established
Obtain alternative answer inventory；
Alternative answer inventory order module, the personal information for obtaining the user is arranged the alternative answer inventory
Sequence, the answer inventory after being sorted；
Response means, for according to the answer inventory after the sequence, obtaining the answer that the user inputs text message.
6. the answer system according to claim 5 that candidate answers sequence is carried out based on subscriber data, it is characterised in that
The personal information of the user includes user preference information, user's geographical location information, user's birthday age information, user
Gender information, user's education background information, user job information and user's habits and customs information.
7. the answer system according to claim 5 that candidate answers sequence is carried out based on subscriber data, it is characterised in that
The alternative answer inventory order module, specifically for：
Obtain the personal information of the user；
The text message inputted by the method for statistical learning, machine learning and deep learning to the user and the individual of user
Information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is ranked up by rule-based method, obtained after sequence
Answer inventory.
8. the answer system according to claim 5 that candidate answers sequence is carried out based on subscriber data, it is characterised in that
The alternative answer inventory order module, specifically for：
Obtain the personal information of the user；
The text message inputted by the method for statistical learning, machine learning, deep learning to the user and the individual of user
Information is modeled, and obtains characteristic information；
According to the characteristic information, the alternative answer inventory is ranked up by the method for machine learning and deep learning,
Answer inventory after being sorted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710218177.2A CN107025283A (en) | 2017-04-05 | 2017-04-05 | The answer method and system of candidate answers sequence are carried out based on subscriber data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710218177.2A CN107025283A (en) | 2017-04-05 | 2017-04-05 | The answer method and system of candidate answers sequence are carried out based on subscriber data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107025283A true CN107025283A (en) | 2017-08-08 |
Family
ID=59526644
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710218177.2A Pending CN107025283A (en) | 2017-04-05 | 2017-04-05 | The answer method and system of candidate answers sequence are carried out based on subscriber data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107025283A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107329986A (en) * | 2017-06-01 | 2017-11-07 | 竹间智能科技（上海）有限公司 | The interactive method and device recognized based on language performance |
| TWI629606B (en) * | 2017-09-26 | 2018-07-11 | 大仁科技大學 | Dialog method of dialog system |
| CN108388944A (en) * | 2017-11-30 | 2018-08-10 | 中国科学院计算技术研究所 | LSTM neural network chips and its application method |
| CN108595498A (en) * | 2018-03-16 | 2018-09-28 | 阿里巴巴集团控股有限公司 | Problem feedback method and device |
| CN109327657A (en) * | 2018-07-16 | 2019-02-12 | 广东小天才科技有限公司 | A kind of taking pictures based on camera searches topic method and private tutor's equipment |
| CN109739970A (en) * | 2018-12-29 | 2019-05-10 | 联想(北京)有限公司 | Information processing method and device and electronic equipment |
| CN110297544A (en) * | 2019-06-28 | 2019-10-01 | 联想(北京)有限公司 | Input information response's method and device, computer system and readable storage medium storing program for executing |
| CN110674276A (en) * | 2019-09-23 | 2020-01-10 | 深圳前海微众银行股份有限公司 | Robot self-learning method, robot terminal, device and readable storage medium |
| CN111741104A (en) * | 2020-06-18 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Method for determining response message, method for configuring response message, device, equipment and storage medium |
| CN111767923A (en) * | 2020-07-28 | 2020-10-13 | 腾讯科技（深圳）有限公司 | Image data detection method and device and computer readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090287678A1 (en) * | 2008-05-14 | 2009-11-19 | International Business Machines Corporation | System and method for providing answers to questions |
| CN104598445A (en) * | 2013-11-01 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Automatic question-answering system and method |
| CN106448670A (en) * | 2016-10-21 | 2017-02-22 | 竹间智能科技（上海）有限公司 | Dialogue automatic reply system based on deep learning and reinforcement learning |
-
2017
- 2017-04-05 CN CN201710218177.2A patent/CN107025283A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090287678A1 (en) * | 2008-05-14 | 2009-11-19 | International Business Machines Corporation | System and method for providing answers to questions |
| CN104598445A (en) * | 2013-11-01 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Automatic question-answering system and method |
| CN106448670A (en) * | 2016-10-21 | 2017-02-22 | 竹间智能科技（上海）有限公司 | Dialogue automatic reply system based on deep learning and reinforcement learning |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107329986A (en) * | 2017-06-01 | 2017-11-07 | 竹间智能科技（上海）有限公司 | The interactive method and device recognized based on language performance |
| TWI629606B (en) * | 2017-09-26 | 2018-07-11 | 大仁科技大學 | Dialog method of dialog system |
| CN108388944A (en) * | 2017-11-30 | 2018-08-10 | 中国科学院计算技术研究所 | LSTM neural network chips and its application method |
| CN108388944B (en) * | 2017-11-30 | 2019-10-18 | 中国科学院计算技术研究所 | A kind of automatic chatting method and robot based on deep neural network |
| CN108595498B (en) * | 2018-03-16 | 2021-09-03 | 创新先进技术有限公司 | Question feedback method and device |
| CN108595498A (en) * | 2018-03-16 | 2018-09-28 | 阿里巴巴集团控股有限公司 | Problem feedback method and device |
| CN109327657A (en) * | 2018-07-16 | 2019-02-12 | 广东小天才科技有限公司 | A kind of taking pictures based on camera searches topic method and private tutor's equipment |
| CN109739970A (en) * | 2018-12-29 | 2019-05-10 | 联想(北京)有限公司 | Information processing method and device and electronic equipment |
| CN110297544A (en) * | 2019-06-28 | 2019-10-01 | 联想(北京)有限公司 | Input information response's method and device, computer system and readable storage medium storing program for executing |
| CN110674276A (en) * | 2019-09-23 | 2020-01-10 | 深圳前海微众银行股份有限公司 | Robot self-learning method, robot terminal, device and readable storage medium |
| CN111741104A (en) * | 2020-06-18 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Method for determining response message, method for configuring response message, device, equipment and storage medium |
| CN111767923A (en) * | 2020-07-28 | 2020-10-13 | 腾讯科技（深圳）有限公司 | Image data detection method and device and computer readable storage medium |
| CN111767923B (en) * | 2020-07-28 | 2024-02-20 | 腾讯科技（深圳）有限公司 | Image data detection method, device and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107025283A (en) | The answer method and system of candidate answers sequence are carried out based on subscriber data | |
| Vitezić et al. | Artificial intelligence acceptance in services: connecting with Generation Z | |
| Shrivastava et al. | Failure prediction of Indian Banks using SMOTE, Lasso regression, bagging and boosting | |
| US9171262B2 (en) | Directed expertise level-based discovery system, method, and device | |
| Lu | A personalized e-learning material recommender system | |
| Cortez et al. | Using data mining to predict secondary school student performance | |
| Weber et al. | Coding the News: The role of computer code in filtering and distributing news | |
| US20200126540A1 (en) | Machine Learning Tool for Navigating a Dialogue Flow | |
| Rahmatullah Imon | Identifying multiple influential observations in linear regression | |
| Rane | Enhancing customer loyalty through Artificial Intelligence (AI), Internet of Things (IoT), and Big Data technologies: improving customer satisfaction, engagement, relationship, and experience | |
| WO2017081715A1 (en) | Reasoning system, reasoning method, and recording medium | |
| Rowe et al. | Mobility patterns of overseas human capital in Australia: the role of a ‘new’graduate visa scheme and rural development policy | |
| Marrone et al. | Automatic assessment of mathematical creativity using natural language processing | |
| Almagrabi et al. | A survey of quality prediction of product reviews | |
| Calvo-Poyo et al. | Citizens’ opinion about investment in public transport projects in cities | |
| JP6449378B2 (en) | Generating device, generating method, and generating program | |
| Eyupoglu et al. | The identification of job satisfaction under Z-information | |
| Ramadhanti et al. | Determining customer relationship management strategy with customer personality analysis using ontology model approach | |
| US20230252418A1 (en) | Apparatus for classifying candidates to postings and a method for its use | |
| US20220138620A1 (en) | Method and system for selecting an alimentary provider | |
| Ushada et al. | Multiple Affective Attributes for the Customization of Post-Pandemic Food Services | |
| WO2020148689A1 (en) | System and method for recommending stickers in a social networking application | |
| JP6165306B1 (en) | Generating device, generating method, and generating program | |
| Hodge et al. | Partial effects in ordered response models with factor variables | |
| Shivashankar et al. | Deep Learning based Campus Assistive Chatbot |
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: 20170808 |