CN106844646B - Artificial Intelligence Reply Method and Device Based on Emotion Prediction - Google Patents
Artificial Intelligence Reply Method and Device Based on Emotion Prediction Download PDFInfo
- Publication number
- CN106844646B CN106844646B CN201710047081.4A CN201710047081A CN106844646B CN 106844646 B CN106844646 B CN 106844646B CN 201710047081 A CN201710047081 A CN 201710047081A CN 106844646 B CN106844646 B CN 106844646B
- Authority
- CN
- China
- Prior art keywords
- information
- user
- emotion
- callback
- candidate
- 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
- 230000008451 emotion Effects 0.000 title claims abstract description 95
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 title claims abstract description 15
- 238000000605 extraction Methods 0.000 claims description 10
- 230000008909 emotion recognition Effects 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 8
- 230000002996 emotional effect Effects 0.000 abstract description 4
- 230000004044 response Effects 0.000 description 6
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010924 continuous production Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90332—Natural language query formulation or dialogue systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Machine Translation (AREA)
Abstract
本发明提供了一种基于情绪预测的人工智能回话方法及装置，能够根据用户的情绪信息、语意信息、情感变化趋势信息、个人状态信息，预测用户针对每条候选回话信息的触发情绪，并经由这预测情绪，在候选回话清单的众多候选回话信息中，找出一个最为合适的句子作为回答。本发明首先获取用户的各种当前信息，借由此提前预测用户对于候选回话清单中各候选回话信息的触发情绪，再根据这些候选回话信息的预测结果，选择最合适的候选回话信息作为回答，更人性化，提高了用户体验。
The present invention provides an artificial intelligence conversation method and device based on emotion prediction, which can predict the user's trigger emotion for each candidate conversation information according to the user's emotional information, semantic information, emotional change trend information, and personal status information, and use the This predicts the sentiment, and finds the most suitable sentence as the answer among the many candidate reply information in the candidate reply list. The present invention first obtains various current information of the user, thereby predicting the triggering emotion of the user for each candidate reply information in the candidate reply list in advance, and then selects the most suitable candidate reply information as the answer according to the prediction results of the candidate reply information, More user-friendly and improve user experience.
Description
Technical Field
The invention relates to an artificial intelligence dialogue system, in particular to an artificial intelligence callback method and device based on emotion prediction.
Background
In the current artificial intelligence dialog system considering the emotion of the user, only sentences which are responded to the current emotion of the user are often found directly, however, the emotion of human beings is a continuous process, different emotions are triggered due to different responses, and the dialog system without considering the emotion of the user cannot find out a proper response strategy effectively.
Disclosure of Invention
The invention aims to provide an artificial intelligence callback method and device based on emotion prediction, and aims to solve the problems that the existing man-machine conversation system cannot predict user emotion in advance and response strategy is not appropriate enough.
The technical scheme adopted by the invention for solving the technical problems is as follows:
an artificial intelligence callback method based on emotion prediction comprises the following steps:
receiving text information input by a user;
acquiring emotion information and semantic information of the user according to the text information;
acquiring emotion change trend information and personal state information of a user;
according to the text information, a candidate answer list responding to the text information is inquired from a corpus; the candidate answer list comprises one or more pieces of candidate answer information;
predicting the triggering emotion of the user aiming at each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information, and acquiring the prediction result of each candidate callback information;
and comparing the prediction results of all candidate callback information to determine the callback information.
On the basis of the above embodiment, further, the step of obtaining emotion information and semantic information of the user according to the text information includes:
acquiring emotion information of the user through a single sentence emotion recognition system according to the text information;
and acquiring the semantic information of the user through a text information extraction system according to the text information.
On the basis of any of the above embodiments, further, the step of acquiring emotion change tendency information and personal state information of the user specifically includes:
acquiring emotion change trend information of a user from historical information of the user;
personal status information of the user is obtained from the user representation.
On the basis of any of the above embodiments, further, the step of comparing the prediction results of all candidate callback information and determining the callback information includes:
carrying out satisfaction evaluation on the prediction result of each piece of answer information to obtain the prediction satisfaction of each piece of answer information;
and determining the callback information with the highest prediction satisfaction as the callback information.
An artificial intelligence answering device based on emotion prediction, comprising:
the receiving module is used for receiving text information input by a user;
the first acquisition module is used for acquiring emotion information and semantic information of the user according to the text information;
the second acquisition module is used for acquiring emotion change trend information and personal state information of the user;
the query module is used for querying a candidate answer list responding to the text information from the corpus according to the text information; the candidate answer list comprises one or more pieces of candidate answer information;
the prediction module is used for predicting the triggering emotion of the user aiming at each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information and acquiring the prediction result of each candidate callback information;
and the determining module is used for comparing the prediction results of all the candidate callback information to determine the callback information.
On the basis of the foregoing embodiment, further, the first obtaining module is configured to:
acquiring emotion information of the user through a single sentence emotion recognition system according to the text information;
and acquiring the semantic information of the user through a text information extraction system according to the text information.
On the basis of any of the foregoing embodiments, further, the second obtaining module is configured to:
acquiring emotion change trend information of a user from historical information of the user;
personal status information of the user is obtained from the user representation.
On the basis of any of the foregoing embodiments, further, the determining module is configured to:
carrying out satisfaction evaluation on the prediction result of each piece of answer information to obtain the prediction satisfaction of each piece of answer information;
and determining the callback information with the highest prediction satisfaction as the callback information.
The invention has the beneficial effects that:
the invention provides an artificial intelligence callback method and device based on emotion prediction, which can predict the triggering emotion of each candidate callback information of a user according to emotion information, semantic information, emotion change trend information and personal state information of the user, and find out a most appropriate sentence as an answer from a plurality of candidate callback information of a candidate callback list through the predicted emotion. The method and the device firstly acquire various current information of the user, predict the triggering emotion of the user on each candidate answer message in the candidate answer list in advance by means of the current information, and select the most appropriate candidate answer message as a response according to the prediction result of the candidate answer messages, so that the method and the device are more humanized and improve the user experience.
Drawings
The invention is further illustrated with reference to the following figures and examples.
Fig. 1 shows a flowchart of an artificial intelligence callback method based on emotion prediction according to an embodiment of the present invention;
FIG. 2 is a flow chart of an artificial intelligence answering method based on emotion prediction according to an embodiment of the present invention;
fig. 3 shows a schematic structural diagram of an artificial intelligence answering device based on emotion prediction according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
Detailed description of the preferred embodiment
As shown in fig. 1 and fig. 2, an embodiment of the present invention provides an artificial intelligence callback method based on emotion prediction, including:
step S101, receiving text information input by a user.
And step S102, acquiring emotion information and semantic information of the user according to the text information.
Step S103, obtaining the emotion change trend information and the personal state information of the user.
Step S104, according to the text information, inquiring a candidate answer list responding to the text information from a corpus; the candidate callback list comprises one or more pieces of candidate callback information. The embodiment of the invention does not limit the corpus, and the corpus can be a preset set of callback information, and the candidate callback list can be a set of callback information corresponding to the text information in the corpus.
And step S105, predicting the triggering emotion of the user aiming at each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information, and obtaining the prediction result of each candidate callback information.
And step S106, comparing the prediction results of all candidate callback information and determining the callback information.
The embodiment of the invention can predict the triggering emotion of each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information of the user, and find out a most appropriate sentence as an answer from a plurality of candidate callback information in the candidate callback list according to the predicted emotion. According to the embodiment of the invention, various current information of the user is firstly acquired, the triggering emotion of the user on each candidate callback information in the candidate callback list is predicted in advance by the aid of the current information, and the most appropriate candidate callback information is selected as a response according to the prediction results of the candidate callback information, so that the embodiment of the invention is more humanized and improves the user experience.
The embodiment of the present invention does not limit the manner of obtaining the emotion information and the text information of the user, and preferably, the step S102 may specifically be: acquiring emotion information of the user through a single sentence emotion recognition system according to the text information; and acquiring the semantic information of the user through a text information extraction system according to the text information. When a user inputs a text, the current emotional state of the user can be judged through a single sentence emotion recognition system; the semantic information of the user can be extracted through the text information extraction system, the semantic information can comprise one or more of a semantic meaning, a keyword, a proper noun and a verb, and meanwhile, the text information extraction system can also send out an original sentence and a sentence rewritten according to the context.
The embodiment of the present invention does not limit the manner of acquiring the emotion change trend information and the personal status information of the user, and preferably, the step S103 may specifically be: acquiring emotion change trend information of a user from historical information of the user; personal status information of the user is obtained from the user representation. The emotion change trend of the user in the past period of time can be acquired through the history of the user; the user image of the user can be used to obtain the current status record of the user, including psychological, physiological or other relevant information of the user.
The embodiment of the present invention does not limit the manner of determining the callback information, and preferably, the step S106 may specifically be: carrying out satisfaction evaluation on the prediction result of each piece of answer information to obtain the prediction satisfaction of each piece of answer information; and determining the callback information with the highest prediction satisfaction as the callback information. For example, three pieces of answer information are in the candidate answer list, the triggering emotion prediction results of the three pieces of answer information are respectively happy, feeble and angry, satisfaction evaluation is performed on the three prediction results, and if the prediction satisfaction degrees of the three pieces of answer information are respectively 80, 60 and 40 in percentage, the first piece of answer information with the prediction satisfaction degree of 80 is selected as the answer information.
Detailed description of the invention
As shown in fig. 3, an embodiment of the present invention provides an artificial intelligence answering device based on emotion prediction, including:
the receiving module 201 is configured to receive text information input by a user.
A first obtaining module 202, configured to obtain emotion information and semantic information of the user according to the text information.
And the second acquisition module 203 is used for acquiring the emotion change trend information and the personal state information of the user.
The query module 204 is configured to query, according to the text information, a candidate answer list responding to the text information from a corpus; the candidate callback list comprises one or more pieces of candidate callback information. The embodiment of the invention does not limit the corpus, and the corpus can be a preset set of callback information, and the candidate callback list can be a set of callback information corresponding to the text information in the corpus.
The prediction module 205 is configured to predict a trigger emotion of the user for each piece of candidate callback information according to the emotion information, the semantic information, the emotion change trend information, and the personal state information, and obtain a prediction result of each piece of candidate callback information.
And the determining module 206 is configured to compare the prediction results of all candidate callback information to determine the callback information.
The embodiment of the invention can predict the triggering emotion of each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information of the user, and find out a most appropriate sentence as an answer from a plurality of candidate callback information in the candidate callback list according to the predicted emotion. According to the embodiment of the invention, various current information of the user is firstly acquired, the triggering emotion of the user on each candidate callback information in the candidate callback list is predicted in advance by the aid of the current information, and the most appropriate candidate callback information is selected as a response according to the prediction results of the candidate callback information, so that the embodiment of the invention is more humanized and improves the user experience.
The embodiment of the present invention does not limit the manner of acquiring the emotion information and the text information of the user, and preferably, the first acquiring module 202 may be configured to: acquiring emotion information of the user through a single sentence emotion recognition system according to the text information; and acquiring the semantic information of the user through a text information extraction system according to the text information. When a user inputs a text, the current emotional state of the user can be judged through a single sentence emotion recognition system; the semantic information of the user can be extracted through the text information extraction system, the semantic information can comprise one or more of a semantic meaning, a keyword, a proper noun and a verb, and meanwhile, the text information extraction system can also send out an original sentence and a sentence rewritten according to the context.
The embodiment of the present invention does not limit the manner of acquiring the emotion change trend information and the personal status information of the user, and preferably, the second acquiring module 203 may be configured to: acquiring emotion change trend information of a user from historical information of the user; personal status information of the user is obtained from the user representation. The emotion change trend of the user in the past period of time can be acquired through the history of the user; the user image of the user can be used to obtain the current status record of the user, including psychological, physiological or other relevant information of the user.
The embodiment of the present invention does not limit the manner of determining the callback information, and preferably, the determining module 206 may be configured to: carrying out satisfaction evaluation on the prediction result of each piece of answer information to obtain the prediction satisfaction of each piece of answer information; and determining the callback information with the highest prediction satisfaction as the callback information. For example, three pieces of answer information are in the candidate answer list, the triggering emotion prediction results of the three pieces of answer information are respectively happy, feeble and angry, satisfaction evaluation is performed on the three prediction results, and if the prediction satisfaction degrees of the three pieces of answer information are respectively 80, 60 and 40 in percentage, the first piece of answer information with the prediction satisfaction degree of 80 is selected as the answer information.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. Although the present invention has been described to a certain extent, it is apparent that appropriate changes in the respective conditions may be made without departing from the spirit and scope of the present invention. It is to be understood that the invention is not limited to the described embodiments, but is to be accorded the scope consistent with the claims, including equivalents of each element described.
Claims (6)
1. An artificial intelligence callback method based on emotion prediction is characterized by comprising the following steps:
receiving text information input by a user;
acquiring emotion information and semantic information of the user according to the text information;
acquiring emotion information of a user through a single sentence emotion recognition system;
acquiring emotion change trend information and personal state information of a user;
according to the text information, a candidate answer list responding to the text information is inquired from a corpus; the candidate answer list comprises one or more pieces of candidate answer information;
predicting the triggering emotion of the user aiming at each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information, and obtaining the prediction result of each candidate callback information;
comparing the prediction results of all candidate callback information to determine the callback information;
the step of acquiring the emotion change trend information and the personal state information of the user specifically comprises the following steps: acquiring emotion change trend information of a user from historical information of the user; personal status information of the user is obtained from the user representation.
2. The artificial intelligence callback method based on emotion prediction according to claim 1, wherein the step of obtaining emotion information and semantic information of the user based on the text information specifically comprises:
and acquiring the semantic information of the user through a text information extraction system according to the text information.
3. The artificial intelligence callback method based on emotion prediction according to claim 1 or 2, wherein the step of comparing the prediction results of all candidate callback information to determine the callback information specifically comprises:
carrying out satisfaction evaluation on the prediction result of each candidate callback information to obtain the prediction satisfaction of each piece of callback information;
and determining the callback information with the highest prediction satisfaction as the callback information.
4. An artificial intelligence answering device based on emotion prediction, comprising:
the receiving module is used for receiving text information input by a user;
the first acquisition module is used for acquiring emotion information and semantic information of the user according to the text information;
acquiring emotion information of a user through a single sentence emotion recognition system;
the second acquisition module is used for acquiring emotion change trend information and personal state information of the user;
the query module is used for querying a candidate answer list responding to the text information from the corpus according to the text information; the candidate answer list comprises one or more pieces of candidate answer information;
the prediction module is used for predicting the triggering emotion of the user aiming at each candidate callback information according to the emotion information, the semantic information, the emotion change trend information and the personal state information and obtaining the prediction result of each candidate callback information;
the determining module is used for comparing the prediction results of all candidate callback information to determine the callback information;
the second obtaining module is specifically used for obtaining the emotion change trend information of the user from the historical information of the user; personal status information of the user is obtained from the user representation.
5. The artificial intelligence answering device based on emotion prediction according to claim 4, wherein the first obtaining module is configured to:
and acquiring the semantic information of the user through a text information extraction system according to the text information.
6. The artificial intelligence callback device according to claim 4 or 5, wherein said determining module is configured to:
carrying out satisfaction evaluation on the prediction result of each candidate callback information to obtain the prediction satisfaction of each piece of callback information;
and determining the callback information with the highest prediction satisfaction as the callback information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710047081.4A CN106844646B (en) | 2017-01-22 | 2017-01-22 | Artificial Intelligence Reply Method and Device Based on Emotion Prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710047081.4A CN106844646B (en) | 2017-01-22 | 2017-01-22 | Artificial Intelligence Reply Method and Device Based on Emotion Prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106844646A CN106844646A (en) | 2017-06-13 |
| CN106844646B true CN106844646B (en) | 2021-03-19 |
Family
ID=59120361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710047081.4A Active CN106844646B (en) | 2017-01-22 | 2017-01-22 | Artificial Intelligence Reply Method and Device Based on Emotion Prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106844646B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11188592B2 (en) * | 2018-02-05 | 2021-11-30 | International Business Machines Corporation | Quantum superposition and entanglement of social sentiment and natural language generation |
| CN113010784B (en) * | 2021-03-17 | 2024-02-06 | 北京十一贝科技有限公司 | Method, apparatus, electronic device and medium for generating prediction information |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593054B (en) * | 2013-11-25 | 2018-04-20 | 北京光年无限科技有限公司 | A kind of combination Emotion identification and the question answering system of output |
| CN105183848A (en) * | 2015-09-07 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Human-computer chatting method and device based on artificial intelligence |
| CN105929964A (en) * | 2016-05-10 | 2016-09-07 | 海信集团有限公司 | Method and device for human-computer interaction |
| CN106055662A (en) * | 2016-06-02 | 2016-10-26 | 竹间智能科技（上海）有限公司 | Emotion-based intelligent conversation method and system |
-
2017
- 2017-01-22 CN CN201710047081.4A patent/CN106844646B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106844646A (en) | 2017-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7581454B2 (en) | DETECTION AND/OR REGISTRATION OF HOT COMMANDS FOR TRIGGERING RESPONSE ACTIONS BY AN AUTOMATED ASSISTANT - Patent application | |
| CN106469212B (en) | Man-machine interaction method and device based on artificial intelligence | |
| CN106649704B (en) | Intelligent conversation control method and system | |
| US9380017B2 (en) | Human assisted chat information system | |
| CN105224278B (en) | Interactive voice service processing method and device | |
| TWI632469B (en) | An automatic reply method and device thereof | |
| EP3631793B1 (en) | Dynamic and/or context-specific hot words to invoke automated assistant | |
| EP3259754B1 (en) | Method and device for providing information | |
| WO2018152006A1 (en) | Intelligent assistant | |
| WO2019011824A1 (en) | Multi-modal dialogue agent | |
| JP7682950B2 (en) | Method and apparatus for managing reservations - Patents.com | |
| CN105991847A (en) | Call communication method and electronic device | |
| CN105930374B (en) | Emotional robot dialogue method, system and robot based on recent feedback | |
| CN105100518A (en) | Method and device for realizing voice mailbox | |
| CN106791107A (en) | Reminding method and device | |
| CN113342945A (en) | Voice session processing method and device | |
| CN109829037A (en) | Method, system, server and the storage medium of intelligent automatic question answering | |
| CN106844646B (en) | Artificial Intelligence Reply Method and Device Based on Emotion Prediction | |
| Strohmann et al. | Design guidelines for creating a convincing user experience with virtual in-vehicle assistants | |
| Prabhakaran et al. | Detecting institutional dialog acts in police traffic stops | |
| Dufour et al. | Access to talker-specific representations is dependent on word frequency | |
| CN117455430B (en) | Resume information processing method, device, equipment and storage medium based on AI | |
| CN110929014B (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN112085386A (en) | Model training method and device for risk behavior prediction | |
| US20190221295A1 (en) | Remote Medical Analysis Systems And Methods |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240520 Address after: Room 122, First Floor, No. 2429 Xingang East Road, Haizhu District, Guangzhou City, Guangdong Province, 510000 (for office only) Patentee after: Zhujian Intelligent Technology (Guangzhou) Co.,Ltd. Country or region after: China Address before: 200233 room 2075, 2 / F, building 1, 146 Fute East 1st Road, Pudong New Area Free Trade Zone, Shanghai Patentee before: ZHUJIAN INTELLIGENT TECHNOLOGY (SHANGHAI) Co.,Ltd. Country or region before: China |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240815 Address after: Room A228, 1st Floor, Building 3, No. 18 Keyuan Road, Economic Development Zone, Daxing District, Beijing 102600 Patentee after: Zhuzhi Technology (Beijing) Co.,Ltd. Country or region after: China Address before: Room 122, First Floor, No. 2429 Xingang East Road, Haizhu District, Guangzhou City, Guangdong Province, 510000 (for office only) Patentee before: Zhujian Intelligent Technology (Guangzhou) Co.,Ltd. Country or region before: China |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20250205 Address after: 701-12, 7th Floor, Building 6, No.5 Shengfang Road, Daxing District, Beijing 102600 Patentee after: Guangzhizhou Intelligent Technology (Beijing) Co.,Ltd. Country or region after: China Address before: Room A228, 1st Floor, Building 3, No. 18 Keyuan Road, Economic Development Zone, Daxing District, Beijing 102600 Patentee before: Zhuzhi Technology (Beijing) Co.,Ltd. Country or region before: China |
|
| TR01 | Transfer of patent right |