CN110489518B - Self-service feedback method and system based on feature extraction - Google Patents
Self-service feedback method and system based on feature extraction Download PDFInfo
- Publication number
- CN110489518B CN110489518B CN201910579542.1A CN201910579542A CN110489518B CN 110489518 B CN110489518 B CN 110489518B CN 201910579542 A CN201910579542 A CN 201910579542A CN 110489518 B CN110489518 B CN 110489518B
- Authority
- CN
- China
- Prior art keywords
- user
- text
- knowledge base
- interest point
- voice
- 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 46
- 238000000605 extraction Methods 0.000 title claims abstract description 18
- 238000005516 engineering process Methods 0.000 claims abstract description 24
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 16
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 16
- 230000007246 mechanism Effects 0.000 claims description 28
- 238000001914 filtration Methods 0.000 claims description 19
- 238000012216 screening Methods 0.000 claims description 12
- 230000008520 organization Effects 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009849 deactivation Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive 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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3335—Syntactic pre-processing, e.g. stopword elimination, stemming
-
- 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
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a self-service feedback method and a self-service feedback system based on feature extraction, wherein the method and the self-service feedback system specifically identify voice sent by a user by using a voice identification technology when the user asks a question to obtain a text matched with the voice; positioning user problems from a pre-constructed interest point knowledge base according to the characteristic information of the text; and feeding back answers to the questions matched with the questions of the user to the user by utilizing a speech synthesis technology. The scheme is used for carrying out problem matching based on the extracted characteristic information, wherein the attention is always the essential content of the problem, so that deviation with the intention of the user is avoided, wrong answers are not fed back, and the problem that the user is frequently asked when the user is fed back by the conventional scheme is solved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a self-service feedback method and a self-service feedback system based on feature extraction.
Background
In some number self-service inquiry systems or automatic answering systems, automatic identification and answering are required according to questions provided by users. The conventional processing scheme is to perform sentence matching, that is, to judge whether a sentence provided by a user is matched with a sentence in a question bank, and if so, to feed back an answer corresponding to the sentence in the question bank to the user. However, this solution does not grasp the essence of the problem, and sometimes, based on the ambiguity of the natural language itself, the sentence with higher similarity to the sentence proposed by the user is not the sentence with the most similar meaning, so as to cause a question of no answer, and make the user experience worse.
Disclosure of Invention
In view of the above, the present invention provides a self-help feedback method and system based on feature extraction, so as to solve the problem that the existing solution frequently answers questions when feeding back to the user.
In order to solve the problems, the invention discloses a self-service feedback method based on feature extraction, which comprises the following steps:
when a user asks a question, recognizing the voice sent by the user by using a voice recognition technology to obtain a text matched with the voice;
positioning a user problem from a pre-constructed interest point knowledge base according to the characteristic information of the text;
and feeding back answers of the questions matched with the questions of the user to the user by utilizing a speech synthesis technology.
Optionally, the feature information includes an organization name feature and an address feature.
Optionally, the locating the user problem from the pre-constructed point of interest knowledge base according to the feature information of the text includes:
traversing the interest point knowledge base according to the mechanism name characteristics, and selecting a plurality of first candidate problems containing the mechanism name characteristics from standard problems;
selecting from the plurality of first candidate problems according to the address characteristics, and selecting a plurality of second candidate problems containing the address characteristics;
and selecting the second candidate problem with the longest continuous characters from the plurality of second candidate problems according to the principle of the longest continuous characters based on a matching algorithm of a continuous matching character weight compensation mechanism, and selecting the second candidate problem with the longest continuous characters as the user problem.
Optionally, before the step of traversing the point of interest knowledge base according to the organization name features, the method further includes the steps of:
and filtering useless information from the text by utilizing a pre-constructed stop word list.
Optionally, before the step of traversing the point of interest knowledge base according to the organization name features, the method further includes the steps of:
and normalizing the proper nouns in the text.
Correspondingly, in order to ensure the implementation of the method, the invention also provides a self-service feedback system based on feature extraction, which comprises the following steps:
the voice recognition module is used for recognizing the voice sent by the user by using a voice recognition technology when the user asks a question to obtain a text matched with the voice;
the problem positioning module is used for positioning the user problem from a pre-constructed interest point knowledge base according to the characteristic information of the text;
and the voice synthesis module is used for feeding back the question answers matched with the user questions to the user by utilizing a voice synthesis technology.
Optionally, the feature information includes an organization name feature and an address feature.
Optionally, the problem location module includes:
the first screening unit is used for traversing the interest point knowledge base according to the mechanism name characteristics and selecting a plurality of first candidate problems containing the mechanism name characteristics from standard problems in the interest point knowledge base;
the second screening unit is used for selecting from the plurality of first candidate problems according to the address characteristics and selecting a plurality of second candidate problems containing the address characteristics;
and the third screening unit is used for selecting from the plurality of second candidate problems based on a matching algorithm of a continuous matching character weight compensation mechanism according to the principle that continuous characters are longest, and selecting the second candidate problem with the longest continuous characters as the user problem.
Optionally, before the step of traversing the point of interest knowledge base according to the organization name features, the method further includes:
and the information filtering unit is used for filtering useless information from the text by utilizing a pre-constructed stop word list before the first screening unit traverses the interest point knowledge base according to the mechanism name characteristics.
Optionally, the method further includes:
and the normalization processing unit is used for performing normalization processing on proper nouns in the text before the first screening unit traverses the interest point knowledge base according to the mechanism name characteristics.
According to the technical scheme, the invention provides a self-service feedback method and a self-service feedback system based on feature extraction, and the method and the system are specifically characterized in that when a user asks a question, the voice sent by the user is identified by using a voice identification technology to obtain a text matched with the voice; positioning user problems from a pre-constructed interest point knowledge base according to the characteristic information of the text; and feeding back answers to the questions matched with the questions of the user to the user by utilizing a speech synthesis technology. The scheme is used for carrying out problem matching based on the extracted characteristic information, wherein the attention is always the essential content of the problem, so that deviation with the intention of the user is avoided, wrong answers are not fed back, and the problem that the user is frequently asked when the user is fed back by the conventional scheme is solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart illustrating steps of a self-help feedback method based on feature extraction according to an embodiment of the present invention;
fig. 2 is a block diagram of a self-help feedback system based on feature extraction according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Fig. 1 is a flowchart illustrating steps of a self-help feedback method based on feature extraction according to an embodiment of the present invention.
Referring to fig. 1, the self-help feedback method is applied to a number self-help inquiry system or other automatic answering systems, and specifically includes the following steps:
and S1, recognizing the voice of the user by utilizing a voice recognition technology to obtain a text.
When a user sends a question by telephone or on site in a voice mode, the voice sent by the user is recognized by using a voice recognition technology, and the voice is converted into a text with the content matched with the actual meaning of the voice. The speech recognition method can be a hidden Markov model method or a neural network model recognition method based on machine learning.
And S2, positioning the user question according to the characteristic information of the text.
Specifically, feature information in a text is extracted, traversal is performed from a pre-constructed interest point knowledge base according to the extracted feature information, and a user problem matched with the feature information is located. The characteristic information comprises mechanism name characteristics and address characteristics, and the invention realizes the positioning of user problems by the following method.
The user question refers to a question which comprises a question matched with the content of a question asked by a user, and answers to the question corresponding to the content of the user exist in a pre-constructed interest point knowledge base. The answer to the question is the reply content required by the user to issue the question.
The interest point knowledge base is provided with a mechanism name word list, the mechanism name word list comprises proper nouns and interest point nouns, the proper nouns refer to field nouns such as restaurants and hospitals, and the interest point nouns refer to unique proper nouns such as restaurant names and hospital names.
In specific implementation, after determining domain nouns and point of interest nouns existing in a text, searching all standard questions preset in a point of interest knowledge base, and taking all standard answers containing the domain nouns and/or the point of interest nouns as first candidate questions, of course, the number of the first candidate questions is many, even tens of thousands, and of course, the range of the first candidate questions is much smaller than that of the standard questions.
Then, after determining the plurality of first candidate questions, screening the plurality of first candidate questions according to the address features, and selecting all the first candidate questions containing the address features as second candidate questions, wherein the number of the second candidate questions is already greatly reduced although the number of the second candidate questions is still more.
Because the addresses have a logical inclusion relationship, for example, three-layer addresses exist in the "sunny district in beijing of china", analysis shows that sometimes a user has a problem that the reported addresses are too detailed, and sometimes the too fine addresses filter correct answers, so that the problem is avoided by extracting the highest two-layer addresses.
Finally, problem similarity matching is carried out, and the inventor of the application finds that continuous characters which are often longer and longer are also matched with the real problem, so that the continuous characters in all second candidate problems are calculated by the matching algorithm based on the continuous matching character weight compensation mechanism, and the longest second candidate problem is selected as the finally selected user problem.
And S3, feeding back answers to the questions to the user by utilizing a voice synthesis technology.
After determining the user question matching the original text, selecting a question answer corresponding to the user question from the interest point knowledge base to feed back to the user. Specifically, the answer to the question in the text state is synthesized into a voice signal by using a voice synthesis technology, and then the voice signal is played to the user, so that the user is provided with question feedback.
According to the technical scheme, the embodiment provides the self-service feedback method based on the feature extraction, and the method specifically comprises the steps of identifying the voice sent by the user by using a voice identification technology when the user asks a question to obtain a text matched with the voice; positioning user problems from a pre-constructed interest point knowledge base according to the characteristic information of the text; and feeding back answers to the questions matched with the questions of the user to the user by utilizing a speech synthesis technology. The scheme is used for carrying out problem matching based on the extracted characteristic information, wherein the attention is always the essential content of the problem, so that deviation with the intention of the user is avoided, wrong answers are not fed back, and the problem that the user is frequently asked when the user is fed back by the conventional scheme is solved.
In addition, a user has a plurality of useless information, the information can be interfered during calculation, the matching rate is reduced, for example, the word "can provide me with a Kendeji telephone of the front door of Beijing", because the user fixes that the using scene is telephone inquiry, the word "can provide me with the telephone" and the "telephone" are useless information, and the useless information is deleted firstly by stopping the word list, so that the difficulty of extracting the characteristics is simplified.
Therefore, in a specific implementation manner of this embodiment, before matching from the point of interest knowledge base according to the organization name feature, the text is filtered by using a pre-constructed deactivation vocabulary, so as to filter out useless information therein, thereby simplifying the difficulty of feature extraction.
In daily life, different people say that many proper nouns are different, such as "passenger transport" and "automobile transport", "taxi" and "taxi" are all expressed with the same meaning, and only the difference is generated because different people speak differently.
In view of this, before filtering out the useless information or separately, a normalization method can be adopted to normalize different utterances into a unified utterance, so that the situation that the matching error problem is caused by different utterances can be avoided.
It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.
Example two
Fig. 2 is a block diagram of a self-help feedback system based on feature extraction according to an embodiment of the present invention.
Referring to fig. 2, the system is applied to a number self-service inquiry system or other automatic answering systems, and specifically comprises a speech recognition module 10, a question positioning module 20 and a speech synthesis module 30.
The voice recognition module is used for recognizing the voice of the user by utilizing a voice recognition technology to obtain a text.
Specifically, when a user sends a question through a telephone or on site in a voice mode, the voice sent by the user is recognized by using a voice recognition technology, and the voice is converted into a text with the content matched with the actual meaning of the voice. The speech recognition method can be a hidden Markov model method or a neural network model recognition method based on machine learning.
And the problem positioning module is used for positioning the user problem according to the characteristic information of the text.
Specifically, feature information in a text is extracted, traversal is performed from a pre-constructed interest point knowledge base according to the extracted feature information, and a user problem matched with the feature information is located. The characteristic information comprises mechanism name characteristics and address characteristics, and the invention realizes the positioning of user problems by the following method.
The user question refers to a question which comprises a question matched with the content of a question asked by a user, and answers to the question corresponding to the content of the user exist in a pre-constructed interest point knowledge base. The answer to the question is the reply content required by the user to issue the question.
The interest point knowledge base is provided with a mechanism name word list, the mechanism name word list comprises proper nouns and interest point nouns, the proper nouns refer to field nouns such as restaurants and hospitals, and the interest point nouns refer to unique proper nouns such as restaurant names and hospital names.
The module specifically comprises a first filtering unit, a second filtering unit and a third filtering unit. After determining the domain nouns and the point of interest nouns existing in the text, the first filtering unit searches all standard questions preset in the point of interest knowledge base, and takes all standard answers containing the domain nouns and/or the point of interest nouns as first candidate questions, wherein the first candidate questions are many, even tens of thousands, and of course, the range of the first candidate questions is much smaller than that of the standard questions.
The second filtering unit is used for screening the plurality of first candidate questions according to the address characteristics after the first filtering unit determines the plurality of first candidate questions, and selecting all the first candidate questions containing the address characteristics as second candidate questions, wherein the number of the second candidate questions is greatly reduced although the number of the second candidate questions is still more.
Because the addresses have a logical inclusion relationship, for example, three-layer addresses exist in the "sunny district in beijing of china", analysis shows that sometimes a user has a problem that the reported addresses are too detailed, and sometimes the too fine addresses filter correct answers, so that the problem is avoided by extracting the highest two-layer addresses.
The third filtering unit is used for performing problem similarity matching, and the inventor of the present application finds that continuous characters which are often longer and longer are also matched with real problems, so that the continuous characters in all the second candidate problems are calculated by a matching algorithm based on a continuous matching character weight compensation mechanism, and the longest second candidate problem is selected as the finally selected user problem.
The voice synthesis module is used for feeding back answers to the questions to the user by utilizing a voice synthesis technology.
After determining the user question matching the original text, selecting a question answer corresponding to the user question from the interest point knowledge base to feed back to the user. Specifically, the answer to the question in the text state is synthesized into a voice signal by using a voice synthesis technology, and then the voice signal is played to the user, so that the user is provided with question feedback.
It can be seen from the above technical solutions that, the present embodiment provides a self-help feedback system based on feature extraction, and the system specifically identifies a voice sent by a user by using a voice identification technology when the user asks a question, so as to obtain a text matched with the voice; positioning user problems from a pre-constructed interest point knowledge base according to the characteristic information of the text; and feeding back answers to the questions matched with the questions of the user to the user by utilizing a speech synthesis technology. The scheme is used for carrying out problem matching based on the extracted characteristic information, wherein the attention is always the essential content of the problem, so that deviation with the intention of the user is avoided, wrong answers are not fed back, and the problem that the user is frequently asked when the user is fed back by the conventional scheme is solved.
In addition, a user has a plurality of useless information, the information can be interfered during calculation, the matching rate is reduced, for example, the word "can provide me with a Kendeji telephone of the front door of Beijing", because the user fixes that the using scene is telephone inquiry, the word "can provide me with the telephone" and the "telephone" are useless information, and the useless information is deleted firstly by stopping the word list, so that the difficulty of extracting the characteristics is simplified.
Therefore, the problem location module can further comprise an information filtering unit, wherein the information filtering unit is used for filtering the text by utilizing the pre-constructed stop word list before matching from the interest point knowledge base according to the organization name characteristics, and filtering useless information in the text to simplify the difficulty of characteristic extraction.
In daily life, different people say that many proper nouns are different, such as "passenger transport" and "automobile transport", "taxi" and "taxi" are all expressed with the same meaning, and only the difference is generated because different people speak differently.
In view of this, the problem location module may also adopt a normalization method to normalize different utterances into a unified utterance before filtering out the useless information or separately, so as to avoid the situation of matching error problems due to different utterances.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The technical solutions provided by the present invention are described in detail above, and the principle and the implementation of the present invention are explained in this document by applying specific examples, and the descriptions of the above examples are only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (6)
1. A self-help feedback method based on feature extraction is characterized in that the method is applied to telephone inquiry; the method comprises the following steps:
when a user asks a question, recognizing the voice sent by the user by using a voice recognition technology to obtain a text matched with the voice;
positioning a user problem from a pre-constructed interest point knowledge base according to the characteristic information of the text; the characteristic information comprises an organization name characteristic and an address characteristic; the interest point knowledge base is constructed with a mechanism name word table, and the mechanism name word table comprises field names and interest point names;
feeding back a question answer matched with the user question to a user by utilizing a voice synthesis technology; the question answers are stored in the interest point knowledge base and correspond to the content of the questions asked by the user;
according to the characteristic information of the text, positioning a user problem from a pre-constructed interest point knowledge base, wherein the positioning comprises the following steps:
traversing the interest point knowledge base according to the mechanism name characteristics, and selecting a plurality of first candidate problems containing the mechanism name characteristics from standard problems;
selecting from the plurality of first candidate problems according to the address characteristics, and selecting a plurality of second candidate problems containing the address characteristics; the address features are the highest two-layer addresses in the address information in the text;
and selecting the second candidate problem with the longest continuous characters from the plurality of second candidate problems according to the principle of the longest continuous characters based on a matching algorithm of a continuous matching character weight compensation mechanism, and selecting the second candidate problem with the longest continuous characters as the user problem.
2. The self-help feedback method of claim 1, further comprising, before the step of traversing the point of interest repository according to the agency name characteristics, the steps of:
and filtering useless information from the text by utilizing a pre-constructed stop word list.
3. The self-help feedback method of claim 1, further comprising, before the step of traversing the point of interest repository according to the agency name characteristics, the steps of:
and normalizing the proper nouns in the text.
4. A self-service feedback system based on feature extraction is characterized by being applied to telephone inquiry; the method comprises the following steps:
the voice recognition module is used for recognizing the voice sent by the user by using a voice recognition technology when the user asks a question to obtain a text matched with the voice;
the problem positioning module is used for positioning the user problem from a pre-constructed interest point knowledge base according to the characteristic information of the text; the characteristic information comprises an organization name characteristic and an address characteristic; the interest point knowledge base is constructed with a mechanism name word table, and the mechanism name word table comprises field names and interest point names;
the voice synthesis module is used for feeding back a question answer matched with the user question to the user by utilizing a voice synthesis technology; the question answers are stored in the interest point knowledge base and correspond to the content of the questions asked by the user;
the problem location module includes:
the first screening unit is used for traversing the interest point knowledge base according to the mechanism name characteristics and selecting a plurality of first candidate problems containing the mechanism name characteristics from standard problems in the interest point knowledge base;
the second screening unit is used for selecting from the plurality of first candidate problems according to the address characteristics and selecting a plurality of second candidate problems containing the address characteristics; the address features are the highest two-layer addresses in the address information in the text;
and the third screening unit is used for selecting from the plurality of second candidate problems based on a matching algorithm of a continuous matching character weight compensation mechanism according to the principle that continuous characters are longest, and selecting the second candidate problem with the longest continuous characters as the user problem.
5. The self-service feedback system of claim 4, further comprising, prior to said step of traversing said point of interest repository according to said agency name characteristics:
and the information filtering unit is used for filtering useless information from the text by utilizing a pre-constructed stop word list before the first screening unit traverses the interest point knowledge base according to the mechanism name characteristics.
6. The self-service feedback system of claim 4, further comprising:
and the normalization processing unit is used for performing normalization processing on proper nouns in the text before the first screening unit traverses the interest point knowledge base according to the mechanism name characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910579542.1A CN110489518B (en) | 2019-06-28 | 2019-06-28 | Self-service feedback method and system based on feature extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910579542.1A CN110489518B (en) | 2019-06-28 | 2019-06-28 | Self-service feedback method and system based on feature extraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110489518A CN110489518A (en) | 2019-11-22 |
| CN110489518B true CN110489518B (en) | 2021-09-17 |
Family
ID=68546572
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910579542.1A Active CN110489518B (en) | 2019-06-28 | 2019-06-28 | Self-service feedback method and system based on feature extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110489518B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101018259A (en) * | 2006-02-08 | 2007-08-15 | 中国电信股份有限公司 | Telecom integrated information system and method |
| CN101030994A (en) * | 2007-04-11 | 2007-09-05 | 华为技术有限公司 | Speech discriminating method system and server |
| CN102708863A (en) * | 2011-03-28 | 2012-10-03 | 德信互动科技（北京）有限公司 | Voice dialogue equipment, system and voice dialogue implementation method |
| CN104598445A (en) * | 2013-11-01 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Automatic question-answering system and method |
| CN108256009A (en) * | 2018-01-03 | 2018-07-06 | 国网江苏省电力有限公司电力科学研究院 | A kind of method for improving electric intelligent response robot and answering accuracy rate |
| CN109858007A (en) * | 2017-11-30 | 2019-06-07 | 上海智臻智能网络科技股份有限公司 | Semantic analysis answering method and device, computer equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103425640A (en) * | 2012-05-14 | 2013-12-04 | 华为技术有限公司 | Multimedia questioning-answering system and method |
-
2019
- 2019-06-28 CN CN201910579542.1A patent/CN110489518B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101018259A (en) * | 2006-02-08 | 2007-08-15 | 中国电信股份有限公司 | Telecom integrated information system and method |
| CN101030994A (en) * | 2007-04-11 | 2007-09-05 | 华为技术有限公司 | Speech discriminating method system and server |
| CN102708863A (en) * | 2011-03-28 | 2012-10-03 | 德信互动科技（北京）有限公司 | Voice dialogue equipment, system and voice dialogue implementation method |
| CN104598445A (en) * | 2013-11-01 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Automatic question-answering system and method |
| CN109858007A (en) * | 2017-11-30 | 2019-06-07 | 上海智臻智能网络科技股份有限公司 | Semantic analysis answering method and device, computer equipment and storage medium |
| CN108256009A (en) * | 2018-01-03 | 2018-07-06 | 国网江苏省电力有限公司电力科学研究院 | A kind of method for improving electric intelligent response robot and answering accuracy rate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110489518A (en) | 2019-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10672391B2 (en) | Improving automatic speech recognition of multilingual named entities | |
| CN107945790B (en) | Emotion recognition method and emotion recognition system | |
| US20190005961A1 (en) | Method and device for processing voice message, terminal and storage medium | |
| EP3405912A1 (en) | Analyzing textual data | |
| JP2019053126A (en) | Growth type interactive device | |
| CN108682420A (en) | A kind of voice and video telephone accent recognition method and terminal device | |
| CN107943786B (en) | Chinese named entity recognition method and system | |
| Howell et al. | Development of a two-stage procedure for the automatic recognition of dysfluencies in the speech of children who stutter: I. Psychometric procedures appropriate for selection of training material for lexical dysfluency classifiers | |
| CN108536807B (en) | Information processing method and device | |
| CN112687291B (en) | Pronunciation defect recognition model training method and pronunciation defect recognition method | |
| JP3920431B2 (en) | Automatic telephone number support system using a heuristic model to predict the most appropriate requested number, method for speech recognition in automatic telephone number support system, and computer-readable recording medium | |
| CN116686045A (en) | End-to-port language understanding without complete transcripts | |
| CN111768789A (en) | Electronic equipment and method, device and medium for determining identity of voice sender thereof | |
| CN115509485A (en) | Filling-in method and device of business form, electronic equipment and storage medium | |
| US20180012602A1 (en) | System and methods for pronunciation analysis-based speaker verification | |
| US20060129398A1 (en) | Method and system for obtaining personal aliases through voice recognition | |
| CN110489518B (en) | Self-service feedback method and system based on feature extraction | |
| Cole et al. | Experiments with a spoken dialogue system for taking the US census | |
| CN115063155B (en) | Data labeling method, device, computer equipment and storage medium | |
| Das et al. | Multi-style speaker recognition database in practical conditions | |
| CN110853674A (en) | Text collation method, apparatus, and computer-readable storage medium | |
| CN113593580B (en) | Voiceprint recognition method and device | |
| US20220012420A1 (en) | Process, system, and method for collecting, predicting, and instructing the pronunciaiton of words | |
| CN114969295A (en) | Dialog interaction data processing method, device and equipment based on artificial intelligence | |
| CN111782779B (en) | Voice question-answering method, system, mobile terminal 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |