CN113515605B - Intelligent robot question-answering method based on artificial intelligence and intelligent robot - Google Patents
Intelligent robot question-answering method based on artificial intelligence and intelligent robot Download PDFInfo
- Publication number
- CN113515605B CN113515605B CN202110550212.7A CN202110550212A CN113515605B CN 113515605 B CN113515605 B CN 113515605B CN 202110550212 A CN202110550212 A CN 202110550212A CN 113515605 B CN113515605 B CN 113515605B
- Authority
- CN
- China
- Prior art keywords
- question
- actual
- initial
- feature
- initial standard
- 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
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 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
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to an intelligent robot question-answering method and an intelligent robot based on artificial intelligence, which are used for acquiring actual question sentences input by a user, acquiring at least two initial standard question sentences corresponding to the actual question sentences from a question sentence database according to the actual question sentences, combining a feature extraction algorithm and a preset sequence to acquire actual question sentence feature speech segments corresponding to the actual question sentences, acquiring target standard question sentences corresponding to the actual question sentences according to the actual question sentence feature speech segments and the initial standard question sentence feature speech segments corresponding to the initial standard question sentences, acquiring target answers corresponding to the target standard question sentences from a question-answering database according to the target standard question sentences, and accurately and reliably acquiring target standard question sentences corresponding to the actual question sentences input by the user.
Description
Technical Field
The invention relates to an intelligent robot question-answering method based on artificial intelligence and an intelligent robot.
Background
At present, intelligent robots capable of performing question-answering operations are increasingly widely applied, such as question-answering robots in public places such as markets, hospitals and the like, and intelligent customer service robots in various electronic commerce platforms. Because the language expression modes of different people are different, the following situations often occur: for a plurality of problems, the meanings of the problems are the same, but the expression modes are different. When the intelligent robot processes data, the intelligent robot analyzes according to the input questions, matches the questions to standard questions, obtains corresponding answers from a database according to the standard questions and outputs the answers. When a standard question is obtained according to an actual problem, a keyword mode is generally adopted at present, namely, keywords in the actual problem are obtained, then a corresponding standard question is obtained according to the keywords, the accuracy of the mode is low, the obtained standard question is easy to be mismatched with the actual problem, and therefore the accuracy of a question-answering result of the intelligent robot is low.
Disclosure of Invention
The invention provides an intelligent robot question-answering method based on artificial intelligence and an intelligent robot, which are used for solving the technical problem of accuracy of a question-answering mode of the existing intelligent robot.
An intelligent robot question-answering method based on artificial intelligence comprises the following steps:
acquiring an actual question input by a user;
acquiring at least two initial standard questions corresponding to the actual questions from a question database according to the actual questions;
analyzing the actual question according to a preset feature extraction algorithm to acquire feature information in the actual question;
sequencing the feature information according to the feature information in the actual question and a preset sequence to obtain an actual question feature sentence segment corresponding to the actual question;
acquiring a target standard question corresponding to the actual question according to the actual question characteristic speech segment and the initial standard question characteristic speech segment corresponding to each initial standard question;
and acquiring a target answer corresponding to the target standard question from a question-answer database according to the target standard question.
Preferably, the acquiring the actual question input by the user specifically includes:
acquiring an initial voice signal input by a user;
performing voice recognition on the initial voice signal to obtain at least two candidate initial questions;
and displaying each candidate initial question, and acquiring a candidate initial question selected by a user from each candidate initial question to obtain an actual question input by the user.
Preferably, the acquiring, according to the actual question, at least two initial standard questions corresponding to the actual question from a question database specifically includes:
acquiring keywords in the actual question according to the actual question;
obtaining keywords of each standard question in the question database;
comparing keywords in the actual question with keywords of each standard question in the question database, and obtaining the similarity between the actual question and each standard question in the question database according to the comparison result, wherein the more the same keywords are, the higher the similarity is;
and determining at least two initial standard questions corresponding to the actual questions according to the respective similarity.
Preferably, the process for obtaining the initial standard question feature speech segment corresponding to the initial standard question includes:
analyzing the initial standard question according to the feature extraction algorithm to acquire feature information in the initial standard question;
and ordering the feature information in the initial standard question according to the feature information in the initial standard question and the preset sequence to obtain an initial standard question feature sentence segment corresponding to the initial standard question.
Preferably, the obtaining, according to the actual question feature speech segment and the initial standard question feature speech segment corresponding to each initial standard question, a target standard question corresponding to the actual question specifically includes:
obtaining the similarity between the actual question and each initial standard question;
according to the similarity between the actual question and each initial standard question, determining the candidate sequence of each initial standard question;
according to the candidate sequence, determining whether the actual question feature speech segments are matched with the initial standard question feature speech segments corresponding to the initial standard questions in sequence; if so, taking the current initial standard question as the target standard question.
Preferably, the judging process of whether the actual question feature speech segment is matched with the initial standard question feature speech segment corresponding to each initial standard question includes:
for any one initial standard question, if the structure of the actual question feature sentence segment is the same as that of the initial standard question feature sentence segment corresponding to the initial standard question, judging that the actual question feature sentence segment is matched with the initial standard question feature sentence segment corresponding to the initial standard question.
Preferably, in the process of sequentially determining whether the actual question feature speech segment is matched with the initial standard question feature speech segment corresponding to each initial standard question, if not, determining whether the meaning of each feature information in the actual question feature speech segment is identical to the meaning of each feature information in the initial standard question feature speech segment of the current initial standard question, and if so, taking the current initial standard question as the target standard question.
Preferably, after the target answer corresponding to the target standard question is obtained from the question-answer database according to the target standard question, the intelligent robot question-answer method based on artificial intelligence further comprises the following steps:
detecting a mobile terminal with Bluetooth started in a preset range, and performing Bluetooth pairing;
and after the Bluetooth pairing is completed, sending the target answer to the mobile terminal.
An intelligent robot comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the intelligent robot question-answering method based on artificial intelligence when executing the computer program.
The technical effects of the invention include: at least two initial standard questions corresponding to the actual questions are determined according to the actual questions input by the user, then the actual questions are analyzed according to a preset feature extraction algorithm to obtain relevant feature information in the actual questions, the feature information is sequenced according to the feature information in the actual questions and a preset sequence to obtain actual question feature speech segments corresponding to the actual questions, then the initial standard question feature speech segments corresponding to the actual questions and the initial standard question feature speech segments corresponding to the initial standard questions are combined to obtain target standard questions corresponding to the actual questions, and finally target answers corresponding to the target standard questions are obtained from a question-answer database according to the obtained target standard questions. According to the intelligent robot question-answering method, the target standard question sentence corresponding to the actual question sentence input by the user can be accurately and reliably obtained, and compared with the traditional mode of determining the standard question sentence only according to the keywords, the accuracy is greatly improved, so that the obtained target standard question sentence has higher matching degree with the actual question, and the accuracy of the question-answering result of the intelligent robot is further improved.
Drawings
Fig. 1 is a flowchart of an intelligent robot question-answering method based on artificial intelligence.
Detailed Description
An intelligent robot question-answering method embodiment based on artificial intelligence:
the embodiment provides an intelligent robot question and answer method based on artificial intelligence, which can be applied to various intelligent robots, intelligent mobile terminals and the like. As shown in fig. 1, the intelligent robot question-answering method based on artificial intelligence comprises the following steps:
step 1: acquiring an actual question input by a user:
the intelligent robot acquires an actual question input by a user, the user can manually input text information corresponding to the actual question, and in general, the intelligent robot is provided with a microphone, and receives a voice signal uttered by the user through the microphone, and then the specific process of acquiring the actual question input by the user comprises the following steps:
(1) Acquiring an initial voice signal input by a user through a microphone;
(2) The method comprises the steps of carrying out voice recognition on an initial voice signal by adopting a built-in voice recognition algorithm, wherein the recognition result of the voice recognition algorithm is not absolutely correct, and the voice signal can be recognized as the same pronunciation or other text information with similar pitch, so that at least two candidate initial questions can be obtained by carrying out voice recognition on the initial voice signal by adopting the built-in voice recognition algorithm, and the pronunciation among the candidate initial questions is the same or highly similar.
(3) Displaying each candidate initial question on a display screen or a touch screen of the intelligent robot, and selecting a correct candidate initial question from the candidate initial questions after the user sees the candidate initial questions, for example: and selecting by clicking the corresponding candidate initial question on the touch screen. And determining the obtained candidate initial question as an actual question input by the user.
Step 2: according to the actual question, acquiring at least two initial standard questions corresponding to the actual question from a question database:
the intelligent robot is preset with a question database, and the question database comprises a plurality of standard questions. Typically, intelligent robots have specific applications such as hospitals, malls, etc. Therefore, in order to improve the recognition accuracy, the question database comprises all currently known standard questions related to the application.
As a specific embodiment, a specific implementation procedure for acquiring at least two initial standard questions corresponding to an actual question from a question database according to the actual question is given as follows:
(1) And acquiring keywords in the actual question according to the actual question. The keyword extraction process can be performed by adopting a currently existing keyword extraction algorithm. In general, a plurality of keywords may be extracted from an actual question. Correspondingly, the same keyword extraction algorithm is adopted to extract keywords of each standard question in the question database, so that keywords of each standard question in the question database are obtained.
(2) And comparing the keywords in the actual question with the keywords of each standard question in the question database, and comparing the keywords in the actual question with the keywords of the standard question for any standard question to determine the number of the same keywords in the actual question and the standard question. It should be understood that when keywords are aligned, the keywords need not be aligned one by one in the order of the keywords in the question, i.e., the alignment need not be performed according to the following alignment rules: the first keyword in the actual question is compared with the first keyword in the standard question, the second keyword in the actual question is compared with the second keyword in the standard question, and so on. But simply determines the number of identical keywords in the actual question and in the standard question. And determining the similarity between the actual question and the standard question according to the number of the same keywords, wherein the more the same keywords are, the higher the similarity is. The specific numerical relation between the number of the same keywords and the similarity is set by the actual requirement.
(3) And determining at least two initial standard questions corresponding to the actual questions according to the respective similarity. Wherein, a similarity threshold value can be set, and standard questions with similarity larger than or equal to the similarity threshold value are obtained from the standard questions, and the obtained standard questions are the required initial standard questions.
As other embodiments, for any standard question, each text in the actual question may be compared with each text in the standard question one by one, specifically: comparing the first text in the actual question with the first text in the standard question, comparing the second text in the actual question with the second text in the standard question, comparing the third text in the actual question with the third text in the standard question, and so on until the last text in the actual question is compared with the last text in the standard question. The similarity is determined according to the number of the same characters, and the higher the similarity is according to the larger number of the same characters.
Step 3: analyzing the actual question according to a preset feature extraction algorithm to obtain feature information in the actual question:
the intelligent robot is preset with a feature extraction algorithm which is used for extracting feature information in the text. It should be appreciated that the feature extraction algorithm is an existing feature extraction algorithm, such as the TF-IDF algorithm. Analyzing the actual question through a feature extraction algorithm to obtain feature information in the actual question. The characteristic information may be related words, words or phrases. The types of the characteristic information may be: background, status, action, question, etc.
In this embodiment, the feature extraction algorithm may include a plurality of feature types, and each specific feature information included in each feature type. Then, the actual question is input into the feature extraction algorithm, so that each feature information in the actual question and the feature type of each feature information can be obtained.
Step 4: sequencing the feature information according to the feature information in the actual question and a preset sequence to obtain an actual question feature sentence segment corresponding to the actual question:
the preset sequence may be the sequence indicating each feature information in the actual question feature speech segment, that is, the sequence indicating the feature type corresponding to each feature information in the actual question feature speech segment, where the sequence of the feature type corresponding to each feature information in the preset sequence is specifically set by the actual requirement. It should be understood that, if the order of the feature information obtained after the analysis of the actual question is different from the preset order, the order of the feature information needs to be adjusted to the preset order. In addition, in order to meet the requirements of the actual questions input by different users, the preset sequence includes the sequence of the feature types corresponding to the feature information in the feature speech segments of the actual questions, and the preset sequence may also include the feature types corresponding to the feature information which is not included in the actual questions.
Correspondingly, the actual question feature sentence segments are sentences obtained by arranging the feature information according to a preset sequence, namely, sentences obtained by arranging feature types corresponding to the feature information in the actual question according to the preset sequence. It can be understood that the actual question feature speech segments are speech segments obtained by removing all the feature information remaining after other irrelevant information in the actual question is sequenced according to a preset sequence. For example: if the actual question is: a+a+b+b+c+c, wherein A, B and C are feature information, a, B and C are irrelevant information, and if the preset order is a feature type name corresponding to feature information B, a feature type name corresponding to feature information D, a feature type name corresponding to feature information C, a feature type name corresponding to feature information E, and a feature type name corresponding to feature information a, the actual question feature phrase is feature information B, feature information C, and feature information a.
Step 5: acquiring a target standard question corresponding to the actual question according to the actual question characteristic speech segment and the initial standard question characteristic speech segment corresponding to each initial standard question:
the initial standard question feature speech segments corresponding to the initial standard questions can also be obtained according to the actual question feature speech segment obtaining process, and the method specifically comprises the following steps: for any one initial standard sentence, analyzing the initial standard question according to a feature extraction algorithm to obtain feature information in the initial standard question; and then, sequencing the feature information in the initial standard question according to the feature information in the initial standard question and a preset sequence to obtain an initial standard question feature sentence segment corresponding to the initial standard question. And further obtaining the characteristic speech segments of the initial standard questions corresponding to the initial standard questions. It should be understood that the initial standard question feature speech segments corresponding to each initial standard question may be obtained by processing after each initial standard question is obtained, or may be obtained by processing after the actual question feature speech segments are obtained.
And acquiring a target standard question corresponding to the actual question according to the actual question characteristic speech segment and the initial standard question characteristic speech segment corresponding to each initial standard question. As a specific embodiment, a specific acquisition procedure of the target standard question is given below:
(1) And obtaining the similarity between the actual question and each initial standard question. It should be appreciated that the similarity may be obtained in accordance with the similarity obtaining procedure given above.
(2) And determining candidate sequences of all the initial standard questions according to the similarity between the actual questions and all the initial standard questions. The higher the similarity, the earlier in the candidate sequence, the candidate sequence is the ranking of the initial standard questions according to the height of the similarity.
(3) And sequentially determining whether the actual question feature speech segments are matched with the initial standard question feature speech segments corresponding to the initial standard questions according to the candidate sequence. Wherein, the meaning of matching can be: the structure of the actual question feature speech section is the same as that of the initial standard question feature speech section (namely, the feature types corresponding to the feature information in the actual question feature speech section are the same as those corresponding to the feature information in the initial standard question feature speech section, the sequence of the feature types is the same, and the specific content of the feature information in the two feature speech sections is the same), namely, the matched meaning is: the actual question feature speech segment is identical to the initial standard question feature speech segment. And when determining whether the actual question feature sentence segment is matched with the initial standard question feature sentence segment corresponding to a certain initial standard question, if so, taking the current initial standard question (namely the certain initial standard question) as a target standard question. Correspondingly, if the actual question feature speech segment is not matched with the initial standard question feature speech segment of the current initial standard question, whether the meaning of each feature information in the actual question feature speech segment is the same as the meaning of each feature information in the initial standard question feature speech segment of the current initial standard question is correspondingly, that is, if the actual question feature speech segment is not completely the same as the current initial standard question feature speech segment, the next step is to determine whether the meaning of each feature information in the actual question feature speech segment is the same as the meaning of each feature information in the initial standard question feature speech segment of the current initial standard question, for example: although the two words are not identical words, it is determined whether the two words are synonyms. If the two standard questions are the same, the current initial standard question is taken as the target standard question. And then determining whether the actual question feature speech segment is matched with the initial standard question feature speech segment corresponding to the next initial standard question.
As other embodiments, the ranking process of the candidate sequence may be not performed, but instead, whether the actual question feature sentence segments are matched with the initial standard question feature sentence segments corresponding to the initial standard question is directly obtained, and the matched initial standard question feature sentence segments are found, where the initial standard question corresponding to the initial standard question feature sentence segment is the required target standard question.
Step 6: according to the target standard question, acquiring a target answer corresponding to the target standard question from a question-answer database:
the intelligent robot is preset with a question-answer database, the question-answer database comprises the corresponding relation between at least two standard question sentences and corresponding answers, and it is understood that the question-answer database comprises the corresponding relation between enough standard question sentences and corresponding answers.
And acquiring target answers corresponding to the target standard question from a question-answer database according to the target standard question. Because the target answers corresponding to the target standard questions are obtained from the question-answer database according to the target standard questions, the method belongs to the conventional question-answer process of the intelligent robot and is not repeated.
After the intelligent robot obtains the target answer, the target answer can be displayed through a touch screen or a voice signal of the target answer can be output through a loudspeaker.
In this embodiment, the intelligent robot is further provided with a bluetooth module. If the user carries the mobile terminal with him, and the Bluetooth function of the mobile terminal is turned on. Then, in this embodiment, after obtaining the target answer, the intelligent robot detects the mobile terminal with bluetooth turned on within a preset range with the intelligent robot as the center, and then performs bluetooth pairing with the mobile terminal with bluetooth turned on.
After the Bluetooth pairing is completed, namely after the Bluetooth connection is established with the mobile terminal, the intelligent robot sends a target answer to the mobile terminal. Therefore, after the intelligent robot obtains the target answer, the target answer is also sent to the mobile terminal of the user, and the intelligent degree is improved.
Intelligent robot embodiment:
the embodiment provides an intelligent robot, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the intelligent robot question-answering method based on artificial intelligence. Because the intelligent robot question-answering method based on artificial intelligence is described in detail above, the detailed description is omitted.
Claims (6)
1. An intelligent robot question-answering method based on artificial intelligence is characterized by comprising the following steps:
acquiring an actual question input by a user;
acquiring at least two initial standard questions corresponding to the actual questions from a question database according to the actual questions;
analyzing the actual question according to a preset feature extraction algorithm to acquire feature information in the actual question;
sequencing the feature information according to the feature information in the actual question and a preset sequence to obtain an actual question feature sentence segment corresponding to the actual question;
acquiring a target standard question corresponding to the actual question according to the actual question characteristic speech segment and the initial standard question characteristic speech segment corresponding to each initial standard question;
acquiring a target answer corresponding to the target standard question from a question-answer database according to the target standard question;
the method comprises the steps of obtaining a target standard question corresponding to the actual question according to the actual question characteristic speech segment and the initial standard question characteristic speech segment corresponding to each initial standard question, wherein the target standard question is specifically:
obtaining the similarity between the actual question and each initial standard question;
according to the similarity between the actual question and each initial standard question, determining the candidate sequence of each initial standard question;
according to the candidate sequence, determining whether the actual question feature speech segments are matched with the initial standard question feature speech segments corresponding to the initial standard questions in sequence; if so, taking the current initial standard question as the target standard question;
the judging process of whether the actual question feature sentence segment is matched with the initial standard question feature sentence segment corresponding to each initial standard question comprises the following steps:
for any one initial standard question, if the structure of the actual question feature sentence segment is the same as that of the initial standard question feature sentence segment corresponding to the initial standard question, judging that the actual question feature sentence segment is matched with the initial standard question feature sentence segment corresponding to the initial standard question;
and in the process of sequentially determining whether the actual question feature sentence segment is matched with the initial standard question feature sentence segment corresponding to each initial standard question, if not, determining whether the meaning of each feature information in the actual question feature sentence segment is identical to the meaning of each feature information in the initial standard question feature sentence segment of the current initial standard question, and if so, taking the current initial standard question as the target standard question.
2. The intelligent robot question-answering method based on artificial intelligence according to claim 1, wherein the obtaining the actual question inputted by the user specifically comprises:
acquiring an initial voice signal input by a user;
performing voice recognition on the initial voice signal to obtain at least two candidate initial questions;
and displaying each candidate initial question, and acquiring a candidate initial question selected by a user from each candidate initial question to obtain an actual question input by the user.
3. The intelligent robot question-answering method based on artificial intelligence according to claim 1, wherein the obtaining at least two initial standard questions corresponding to the actual questions from a question database according to the actual questions comprises:
acquiring keywords in the actual question according to the actual question;
obtaining keywords of each standard question in the question database;
comparing keywords in the actual question with keywords of each standard question in the question database, and obtaining the similarity between the actual question and each standard question in the question database according to the comparison result, wherein the more the same keywords are, the higher the similarity is;
and determining at least two initial standard questions corresponding to the actual questions according to the respective similarity.
4. The intelligent robot question-answering method based on artificial intelligence according to claim 1, wherein the process of obtaining the initial standard question feature sentence segment corresponding to the initial standard question comprises:
analyzing the initial standard question according to the feature extraction algorithm to acquire feature information in the initial standard question;
and ordering the feature information in the initial standard question according to the feature information in the initial standard question and the preset sequence to obtain an initial standard question feature sentence segment corresponding to the initial standard question.
5. The intelligent robot question-answering method based on artificial intelligence according to claim 1, wherein after obtaining a target answer corresponding to the target standard question from a question-answer database according to the target standard question, the intelligent robot question-answering method based on artificial intelligence further comprises the steps of:
detecting a mobile terminal with Bluetooth started in a preset range, and performing Bluetooth pairing;
and after the Bluetooth pairing is completed, sending the target answer to the mobile terminal.
6. A smart robot comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the artificial intelligence based smart robot question-answering method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110550212.7A CN113515605B (en) | 2021-05-20 | 2021-05-20 | Intelligent robot question-answering method based on artificial intelligence and intelligent robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110550212.7A CN113515605B (en) | 2021-05-20 | 2021-05-20 | Intelligent robot question-answering method based on artificial intelligence and intelligent robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113515605A CN113515605A (en) | 2021-10-19 |
| CN113515605B true CN113515605B (en) | 2023-12-19 |
Family
ID=78064929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110550212.7A Active CN113515605B (en) | 2021-05-20 | 2021-05-20 | Intelligent robot question-answering method based on artificial intelligence and intelligent robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113515605B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315766A (en) * | 2017-05-16 | 2017-11-03 | 广东电网有限责任公司江门供电局 | A kind of voice response method and its device for gathering intelligence and artificial question and answer |
| CN108628811A (en) * | 2018-04-10 | 2018-10-09 | 北京京东尚科信息技术有限公司 | The matching process and device of address text |
| CN109035001A (en) * | 2018-06-28 | 2018-12-18 | 招联消费金融有限公司 | Intelligent voice response processing method and system |
| CN110647614A (en) * | 2019-08-01 | 2020-01-03 | 平安科技（深圳）有限公司 | Intelligent question and answer method, device, medium and electronic equipment |
| CN111444321A (en) * | 2019-01-16 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Question answering method, device, electronic equipment and storage medium |
| CN111597313A (en) * | 2020-04-07 | 2020-08-28 | 深圳追一科技有限公司 | Question answering method, device, computer equipment and storage medium |
| CN112052683A (en) * | 2020-09-03 | 2020-12-08 | 平安科技（深圳）有限公司 | Text matching method and device, computer equipment and storage medium |
| CN112052676A (en) * | 2020-08-31 | 2020-12-08 | 国网山东省电力公司荣成市供电公司 | Text content processing method, computer equipment and storage medium |
| CN112287069A (en) * | 2020-10-29 | 2021-01-29 | 平安科技（深圳）有限公司 | Information retrieval method and device based on voice semantics and computer equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101107760B1 (en) * | 2009-08-24 | 2012-01-20 | 송도규 | System and Method for Intelligent Searching and Question-Answering |
-
2021
- 2021-05-20 CN CN202110550212.7A patent/CN113515605B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315766A (en) * | 2017-05-16 | 2017-11-03 | 广东电网有限责任公司江门供电局 | A kind of voice response method and its device for gathering intelligence and artificial question and answer |
| CN108628811A (en) * | 2018-04-10 | 2018-10-09 | 北京京东尚科信息技术有限公司 | The matching process and device of address text |
| CN109035001A (en) * | 2018-06-28 | 2018-12-18 | 招联消费金融有限公司 | Intelligent voice response processing method and system |
| CN111444321A (en) * | 2019-01-16 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Question answering method, device, electronic equipment and storage medium |
| CN110647614A (en) * | 2019-08-01 | 2020-01-03 | 平安科技（深圳）有限公司 | Intelligent question and answer method, device, medium and electronic equipment |
| CN111597313A (en) * | 2020-04-07 | 2020-08-28 | 深圳追一科技有限公司 | Question answering method, device, computer equipment and storage medium |
| CN112052676A (en) * | 2020-08-31 | 2020-12-08 | 国网山东省电力公司荣成市供电公司 | Text content processing method, computer equipment and storage medium |
| CN112052683A (en) * | 2020-09-03 | 2020-12-08 | 平安科技（深圳）有限公司 | Text matching method and device, computer equipment and storage medium |
| CN112287069A (en) * | 2020-10-29 | 2021-01-29 | 平安科技（深圳）有限公司 | Information retrieval method and device based on voice semantics and computer equipment |
Non-Patent Citations (2)
| Title |
|---|
| 句子相似度计算新方法及在问答***中的应用;周法国;杨炳儒;;计算机工程与应用(第01期);全文 * |
| 多特征融合的中文问答***答案抽取算法;唐朝霞;;贵州大学学报(自然科学版)(第05期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113515605A (en) | 2021-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111046133B (en) | Question and answer method, equipment, storage medium and device based on mapping knowledge base | |
| US11062090B2 (en) | Method and apparatus for mining general text content, server, and storage medium | |
| CN108829822B (en) | Media content recommendation method and device, storage medium and electronic device | |
| CN106875941B (en) | Voice semantic recognition method of service robot | |
| US11494161B2 (en) | Coding system and coding method using voice recognition | |
| US7650283B2 (en) | Dialogue supporting apparatus | |
| CN110334179B (en) | Question-answer processing method, device, computer equipment and storage medium | |
| CN113168836A (en) | Computer system, speech recognition method and program | |
| EP4109295A1 (en) | Knowledge graph-based question answering method and apparatus, computer device, and medium | |
| CN109522397B (en) | Information processing method and device | |
| CN110827803A (en) | Method, device and equipment for constructing dialect pronunciation dictionary and readable storage medium | |
| CN111554276B (en) | Speech recognition method, device, equipment and computer readable storage medium | |
| CN113220862A (en) | Standard question recognition method and device, computer equipment and storage medium | |
| CN112632248A (en) | Question answering method, device, computer equipment and storage medium | |
| CN112256827A (en) | Sign language translation method and device, computer equipment and storage medium | |
| CN110110066B (en) | Interactive data processing method and device and computer readable storage medium | |
| CN113268981A (en) | Information processing method and device and electronic equipment | |
| CN113658690A (en) | Intelligent medical guide method and device, storage medium and electronic equipment | |
| CN113220854A (en) | Intelligent dialogue method and device for machine reading understanding | |
| CN113515605B (en) | Intelligent robot question-answering method based on artificial intelligence and intelligent robot | |
| CN112417210A (en) | Body-building video query method, device, terminal and storage medium | |
| CN117422067A (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN110263346B (en) | Semantic analysis method based on small sample learning, electronic equipment and storage medium | |
| CA3131060A1 (en) | Speech analysis system | |
| CN115881108A (en) | Voice recognition method, device, equipment and storage medium |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20231121 Address after: No. 1916, 19th Floor, Building 14, Binjiang Coast Phase III, No. 23 Haidian Island East Road, Meilan District, Haikou City, Hainan Province, 570100 Applicant after: Zhongchen Tianrun Industrial Co.,Ltd. Address before: 450000 No. 33, Zone D, floor 8, building 2, No. 85, Jinshui East Road, Zhengzhou area (Zhengdong), pilot Free Trade Zone, Zhengzhou City, Henan Province Applicant before: Henan GuangYue Network Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |