CN108388553B - Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system - Google Patents
Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system Download PDFInfo
- Publication number
- CN108388553B CN108388553B CN201711466239.8A CN201711466239A CN108388553B CN 108388553 B CN108388553 B CN 108388553B CN 201711466239 A CN201711466239 A CN 201711466239A CN 108388553 B CN108388553 B CN 108388553B
- Authority
- CN
- China
- Prior art keywords
- sentence
- current
- kitchen
- preprocessed
- stored
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/16—Sound input; Sound output
- G06F3/167—Audio in a user interface, e.g. using voice commands for navigating, audio feedback
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a method for disambiguating dialogues, which comprises the following steps: acquiring a current input statement input by a user; performing sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence; calling a plurality of different pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the current preprocessed sentence category; and judging whether the current input sentence is ambiguous or not, if so, determining the type of the current input sentence according to the type of the input sentence in the previous round and the current preprocessed sentence, extracting corresponding first associated information from a pre-stored database according to the type of the current input sentence, and if not, extracting corresponding second associated information from the pre-stored database according to the type of the preprocessed sentence. The method for eliminating the ambiguity by the dialogue can more accurately identify the real intention of the inquiry of the user, thereby feeding back accurate information to the client.
Description
Technical Field
The present invention relates to the field of intelligent dialogues, and in particular, to a method for disambiguating dialogues, an electronic device, and a kitchen-oriented dialog system.
Background
Today, the AI has revolutionized the traditional industries, such as the kitchen home field, with the big outbreak of artificial intelligence. And the dialog system is used as an interface for interaction between a person and kitchen household equipment, and the performance of the dialog system directly determines the user experience of the whole intelligent kitchen household scheme. The machine can really understand human language, communicate with the user normally, judge and finish the requirements of the user, and is a great target for the evolution and development of a conversation system.
Many current dialog methods are simple question-and-answer patterns, which do not capture the semantics of the context well and thus do not understand the user's intent accurately. In addition, the judgment of the user intention is mostly performed by analyzing the text semantics of the query, in actual use, the query of many users is short, such as "watching video", and it cannot be judged whether the real intention of the user is to watch movies, art-integrated videos or videos demonstrating the step of making dishes only by the language model analysis of the sentence, so that the query is called as an ambiguous query. A truly intelligent dialog system should provide a mechanism to analyze the true intent of the ambiguous query in the above example. The conventional dialog method cannot accurately recognize the true intention of the user query.
Disclosure of Invention
In order to overcome the defects of the prior art, one of the objectives of the present invention is to provide a method for disambiguating a dialog, which can solve the problem that the conventional dialog method cannot accurately identify the true intention of a user query.
Another object of the present invention is to provide an electronic device, which can solve the problem that the conventional dialog method cannot accurately identify the true intention of the user query.
The invention also aims to provide a kitchen-oriented dialog system, which can solve the problem that the traditional dialog method cannot accurately identify the real intention of the user query.
One of the purposes of the invention is realized by adopting the following technical scheme:
a method for session disambiguation comprising the steps of:
s1: acquiring a current input statement input by a user;
s2: performing sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence;
s3: calling a plurality of different pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the current preprocessed sentence category;
s4: judging whether the current input statement is ambiguous or not, if so, determining the category of the current input statement according to the category of the input statement in the previous round and the current preprocessed statement, and extracting corresponding first associated information from a pre-stored database according to the category of the current input statement; if not, go to S5;
s5: and extracting corresponding second associated information from a pre-stored database according to the type of the current preprocessed statement.
Further, the S2 constructs the current input sentence in the kitchen according to the CRF model, the HMM model, and the N-gram model, and performs sentence segmentation, word segmentation, and part-of-speech tagging to obtain the current preprocessed sentence.
Further, the CRF model, the HMM model and the N-gram model are training models obtained by training the kitchen field training corpus through different classifiers.
Further, the step S3 is specifically to obtain word vectors of the current preprocessed sentence, respectively call different kitchen domain classification models to classify the word vectors, each kitchen domain classification model obtains a domain score after classifying the word vectors, and the domain category corresponding to the domain score with the highest score value and the score value reaching the threshold value is used as the current preprocessed sentence category.
Further, when the domain score with the highest score value does not reach a threshold value, the obtained current preprocessed statement category is none.
Further, in S4, it is determined whether the current preprocessed sentence is ambiguous, specifically, whether a pre-stored ambiguity database filters a vocabulary that is the same as the current preprocessed sentence, if so, the current preprocessed sentence is ambiguous, and if not, the current preprocessed sentence is unambiguous.
The second purpose of the invention is realized by adopting the following technical scheme:
an electronic device, comprising: a processor;
a memory; and a program, wherein the program is stored in the memory and configured to be executed by the processor, the program comprising instructions for performing the method of dialog disambiguation of the invention.
The third purpose of the invention is realized by adopting the following technical scheme:
an acquisition module: the system comprises a database, a database and a user interface, wherein the database is used for storing a current input statement input by a user;
a preprocessing module: the system is used for carrying out sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence;
a text classification module: the system comprises a plurality of pre-stored kitchen field classification models, a plurality of pre-stored kitchen field classification models and a plurality of pre-stored kitchen field classification models, wherein the pre-stored kitchen field classification models are used for calling the pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the classes of the current preprocessed sentences;
an ambiguity service module: the method is used for judging whether the current preprocessed statement is ambiguous or not;
an extraction module: the system comprises a database, a database and a database, wherein the database is used for storing the type of the current input statement, and extracting corresponding first associated information from the database according to the type of the current input statement; and the method is also used for extracting corresponding second associated information from a pre-stored database according to the type of the current preprocessed statement.
Furthermore, the kitchen field classification model comprises a cooking classification model, a music classification model, an equipment control classification model and an interface operation classification model, the cooking classification model is a classification model obtained after cooking information is put into the classifier and training is carried out on the cooking classification model, the music classification model is a classification model obtained after music information is put into the classifier and training is carried out on the music information, the equipment control classification model is a classification model obtained after equipment control information is put into the classifier and training is carried out on the equipment control classification model, and the interface operation classification model is a classification model obtained after interface control classification information is put into the classifier and training is carried out on the interface control classification model.
Compared with the prior art, the invention has the beneficial effects that: the method for eliminating the ambiguity through the dialogue obtains the current input sentence input by the user and carries out sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain the current preprocessed sentence. Calling a plurality of different pre-stored kitchen field classification models to classify the current preprocessed sentence to obtain a current preprocessed sentence type, judging whether the current input sentence is ambiguous, if yes, determining the type of the current input sentence according to the type of the input sentence in the previous round and the current preprocessed sentence, and extracting corresponding first associated information from a pre-stored database according to the type of the current input sentence; if not, extracting corresponding second correlation information from a pre-stored database according to the type of the current preprocessed statement; the real intention inquired by the user can be more accurately identified by preprocessing the current input sentence of the user and further extracting the associated information after judging whether the current input sentence is ambiguous, so that accurate information is fed back to the client.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood and to implement them in accordance with the contents of the description, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings. The detailed description of the present invention is given in detail by the following examples and the accompanying drawings.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of a method of session disambiguation according to the present invention;
fig. 2 is a block diagram of the architecture of the kitchen-oriented dialog system of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and the detailed description, and it should be noted that any combination of the embodiments or technical features described below can be used to form a new embodiment without conflict.
The method for disambiguating a dialog of the present invention as shown in fig. 1 comprises the steps of:
s1: acquiring a current input statement input by a user;
s2: performing sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence; specifically, training linguistic training materials in the kitchen field by using different classifiers to obtain a CRF model, an HMM model and an N-gram model, and performing sentence segmentation, word segmentation and part-of-speech tagging on a current input sentence according to the CRF model, the HMM model and the N-gram model to obtain a current preprocessed sentence; the kitchen field corpus is a corpus subjected to word annotation processing. The following are exemplified: for example, if the current input sentence is "i want to eat sweet and sour yellow river carp", the sentence is divided into "i/r, want/v, eat/v, sweet and sour yellow river carp/ndish" by using the above model, wherein r is represented as pronouns, v is represented as verbs, and ndish is represented as a dish name.
S3: calling a plurality of different pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the current preprocessed sentence category; the kitchen classification model comprises a cooking classification model, a music classification model, an equipment control classification model and an interface operation classification model; acquiring word vectors of a current preprocessed sentence, respectively calling different kitchen field classification models to classify the word vectors, obtaining a field score belonging to the kitchen field after each kitchen field classification model classifies the word vectors, and taking a field category of the kitchen field corresponding to the field score with the highest score value and the score value reaching a threshold value as the current preprocessed sentence category; and when the field score with the highest score value does not reach the threshold value, the obtained current preprocessed statement category is none.
S4: judging whether the current preprocessed statement is ambiguous or not, determining the category of the current input statement according to the category of the previous input statement and the current preprocessed statement, and extracting corresponding first associated information from a pre-stored database according to the category of the current input statement; if not, go to S5; judging whether the preprocessed sentence has ambiguity specifically by screening whether the words and phrases same as the preprocessed sentence exist in a pre-stored ambiguity database, if so, judging that the preprocessed sentence has ambiguity, otherwise, judging that the preprocessed sentence is unambiguous, and before S4, the method comprises a common pre-stored ambiguity database, collecting a plurality of words and phrases with ambiguity in the kitchen field, and storing the words and phrases with ambiguity in the pre-stored ambiguity database; for example: and when the user watches the video, the video watching is an ambiguous word, and the video watching can be a movie watching, a cooking video watching and the like.
S5: and extracting corresponding second associated information from a pre-stored database according to the type of the current preprocessed statement. The first and second association information in this embodiment are only for distinguishing the difference of the association information, and the first and second association information do not have any meaning, and the first association information and the second association information are both information having an association relationship with the category of the current sentence or the category of the preprocessed sentence in the pre-stored database. Before extracting corresponding first associated information from a pre-stored database according to the type of the current input sentence in S4 and corresponding second associated information from a pre-stored database according to the type of the pre-processed sentence in S5, creating the pre-stored database to collect information such as various vocabularies, videos, music, dish names and the like in the kitchen field, classifying the information to obtain different pre-stored types, and establishing a mapping relation between the pre-stored types and the data information of the kitchen field; storing the pre-stored type and data information in a pre-stored database; therefore, the specific extraction process is as follows: the method comprises the steps of firstly matching the category of a preprocessed sentence with the pre-stored category in a pre-stored database to obtain a corresponding pre-stored category, screening corresponding associated information from the pre-stored database according to the pre-stored category, then identifying the associated information based on a CRF model, an HMM model and a language model of an n-gram of word labeling, and converting the associated information into a clear logical expression.
The invention also includes an electronic device comprising: a processor;
a memory; and a program, wherein the program is stored in the memory and configured to be executed by the processor, the program comprising instructions for performing the method of dialog disambiguation of the invention.
The kitchen-oriented dialog system of the present invention, as shown in fig. 2, comprises: an acquisition module: the system comprises a database, a database and a user interface, wherein the database is used for storing a current input statement input by a user; a preprocessing module: the system is used for carrying out sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence; a text classification module: the system comprises a plurality of pre-stored kitchen field classification models, a plurality of pre-stored kitchen field classification models and a plurality of pre-stored kitchen field classification models, wherein the pre-stored kitchen field classification models are used for calling the pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the classes of the current preprocessed sentences; an ambiguity service module: the system comprises a database, a database and a database, wherein the database is used for storing the type of the current input statement, and extracting corresponding first associated information from the database according to the type of the current input statement; and the processor is also used for extracting corresponding second correlation information from a pre-stored database according to the current type of the preprocessed statement. The kitchen field classification model comprises a cooking classification model, a music classification model, an equipment control classification model and an interface operation classification model, wherein the cooking classification model is obtained by putting cooking information into a classifier and training, the music classification model is obtained by putting music information into the classifier and training, the equipment control classification model is obtained by putting equipment control information into the classifier and training, and the interface operation classification model is obtained by putting interface control classification information into the classifier and training.
The method for eliminating the ambiguity in the dialogue is characterized in that a current preprocessed sentence is obtained by obtaining a current input sentence input by a user and carrying out sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence. Calling a plurality of different pre-stored kitchen field classification models to classify the current preprocessed sentence to obtain a current preprocessed sentence type, judging whether the current input sentence is ambiguous, if yes, determining the type of the current input sentence according to the type of the input sentence in the previous round and the current preprocessed sentence, and extracting corresponding first associated information from a pre-stored database according to the type of the current input sentence; if not, extracting corresponding second correlation information from a pre-stored database according to the type of the current preprocessed statement; the real intention inquired by the user can be more accurately identified by preprocessing the current input sentence of the user and further extracting the associated information after judging whether the current input sentence is ambiguous, so that accurate information is fed back to the client. In the kitchen-oriented dialog system, the ambiguity service module is independent into a module with the same priority as the text classification module, and the ambiguity service module is mainly based on the following consideration: firstly, service decoupling and secondly, improving service reusability, enabling an ambiguity service module to better judge whether ambiguity exists in input sentences of a user, and further accurately judging according to specific ambiguity to obtain more accurate results.
The foregoing is merely a preferred embodiment of the invention and is not intended to limit the invention in any manner; those skilled in the art can readily practice the invention as shown and described in the drawings and detailed description herein; however, those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiments as a basis for designing or modifying other structures for carrying out the same purposes of the present invention without departing from the scope of the invention as defined by the appended claims; meanwhile, any changes, modifications, and evolutions of the equivalent changes of the above embodiments according to the actual techniques of the present invention are still within the protection scope of the technical solution of the present invention.
Claims (8)
1. A method for session disambiguation comprising the steps of:
s1: acquiring a current input statement input by a user;
s2: performing sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence;
s3: calling a plurality of different pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the current preprocessed sentence category;
s4: judging whether the current input statement is ambiguous or not, if so, determining the category of the current input statement according to the category of the input statement in the previous round and the current preprocessed statement, and extracting corresponding first associated information from a pre-stored database according to the category of the current input statement; if not, go to S5;
s5: extracting corresponding second associated information from a pre-stored database according to the type of the current preprocessed statement;
the S3 is specifically configured to obtain word vectors of the current preprocessed sentence, respectively call different kitchen domain classification models to classify the word vectors, where each kitchen domain classification model obtains a domain score after classifying the word vectors, and the domain category corresponding to the domain score with the highest score value and the score value reaching a threshold is used as the current preprocessed sentence category.
2. A method of dialog disambiguation as claimed in claim 1, characterized in that: and S2, specifically, constructing the current input sentence in the kitchen according to the CRF model, the HMM model and the N-gram model, and performing sentence segmentation, word segmentation and part-of-speech tagging to obtain the current preprocessed sentence.
3. A method of dialog disambiguation as claimed in claim 2, characterized in that: the CRF model, the HMM model and the N-gram model are training models obtained by training the kitchen field training corpus through different classifiers.
4. A method of dialog disambiguation as claimed in claim 1, characterized in that: and when the field score with the highest score value does not reach a threshold value, the obtained current preprocessed statement category is none.
5. A method of dialog disambiguation as claimed in claim 1, characterized in that: in S4, the step of determining whether the current preprocessed sentence is ambiguous is to filter a pre-stored ambiguity database to determine whether there is a word that is the same as the current preprocessed sentence, if so, the current preprocessed sentence is ambiguous, and if not, the current preprocessed sentence is unambiguous.
6. An electronic device, characterized by comprising: a processor;
a memory; and a program, wherein the program is stored in the memory and configured to be executed by the processor, the program comprising instructions for carrying out the method of any one of claims 1-5.
7. A kitchen-oriented dialog system, characterized by comprising:
an acquisition module: the system comprises a database, a database and a user interface, wherein the database is used for storing a current input statement input by a user;
a preprocessing module: the system is used for carrying out sentence segmentation, word segmentation and part-of-speech tagging on the current input sentence to obtain a current preprocessed sentence;
a text classification module: the system comprises a plurality of pre-stored kitchen field classification models, a plurality of pre-stored kitchen field classification models and a plurality of pre-stored kitchen field classification models, wherein the pre-stored kitchen field classification models are used for calling the pre-stored kitchen field classification models to classify the current preprocessed sentences to obtain the classes of the current preprocessed sentences;
an ambiguity service module: the method is used for judging whether the current preprocessed statement is ambiguous or not;
an extraction module: the system comprises a database, a database and a database, wherein the database is used for storing the type of the current input statement, and extracting corresponding first associated information from the database according to the type of the current input statement; the second correlation information is extracted from a pre-stored database according to the type of the current preprocessed statement;
the text classification module: the system is further used for obtaining word vectors of the current preprocessed sentence, different kitchen field classification models are respectively called to classify the word vectors, each kitchen field classification model can obtain a field score after classifying the word vectors, and the field category corresponding to the field score with the highest score value and the score value reaching a threshold value is used as the current preprocessed sentence category.
8. The kitchen-oriented dialog system of claim 7 wherein: the kitchen field classification model comprises a cooking classification model, a music classification model, an equipment control classification model and an interface operation classification model, the classification model obtained after cooking information is put into the classifier and training is the cooking classification model, the classification model obtained after music information is put into the classifier and training is the music classification model, the classification model obtained after equipment control information is put into the classifier and training is the equipment control classification model, and the classification model obtained after interface control classification information is put into the classifier and training is the interface operation classification model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711466239.8A CN108388553B (en) | 2017-12-28 | 2017-12-28 | Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711466239.8A CN108388553B (en) | 2017-12-28 | 2017-12-28 | Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108388553A CN108388553A (en) | 2018-08-10 |
| CN108388553B true CN108388553B (en) | 2021-10-15 |
Family
ID=63076568
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711466239.8A Active CN108388553B (en) | 2017-12-28 | 2017-12-28 | Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108388553B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109582965B (en) * | 2018-11-30 | 2022-03-01 | 四川长虹电器股份有限公司 | Distributed platform construction method and system of semantic analysis engine |
| CN109693244B (en) * | 2018-12-24 | 2020-06-12 | 零犀(北京)科技有限公司 | Method and device for optimizing conversation robot |
| CN110597958B (en) * | 2019-09-12 | 2022-03-25 | 思必驰科技股份有限公司 | Text classification model training and using method and device |
| CN110827802A (en) * | 2019-10-31 | 2020-02-21 | 苏州思必驰信息科技有限公司 | Speech recognition training and decoding method and device |
| CN111199149B (en) * | 2019-12-17 | 2023-10-20 | 航天信息股份有限公司 | Sentence intelligent clarification method and system for dialogue system |
| CN112100368B (en) * | 2020-07-21 | 2024-01-26 | 深思考人工智能科技(上海)有限公司 | Method and device for identifying dialogue interaction intention |
| CN112906379B (en) * | 2020-12-10 | 2023-12-22 | 苏州英特雷真智能科技有限公司 | Method for researching and developing natural language processing technology based on graph theory |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7031910B2 (en) * | 2001-10-16 | 2006-04-18 | Xerox Corporation | Method and system for encoding and accessing linguistic frequency data |
| CN104050157A (en) * | 2014-06-16 | 2014-09-17 | 海信集团有限公司 | Ambiguity elimination method and system |
| CN106469188A (en) * | 2016-08-30 | 2017-03-01 | 北京奇艺世纪科技有限公司 | A kind of entity disambiguation method and device |
| CN106528530A (en) * | 2016-10-24 | 2017-03-22 | 北京光年无限科技有限公司 | Method and device for determining sentence type |
| CN106598947A (en) * | 2016-12-15 | 2017-04-26 | 山西大学 | Bayesian word sense disambiguation method based on synonym expansion |
-
2017
- 2017-12-28 CN CN201711466239.8A patent/CN108388553B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7031910B2 (en) * | 2001-10-16 | 2006-04-18 | Xerox Corporation | Method and system for encoding and accessing linguistic frequency data |
| CN104050157A (en) * | 2014-06-16 | 2014-09-17 | 海信集团有限公司 | Ambiguity elimination method and system |
| CN106469188A (en) * | 2016-08-30 | 2017-03-01 | 北京奇艺世纪科技有限公司 | A kind of entity disambiguation method and device |
| CN106528530A (en) * | 2016-10-24 | 2017-03-22 | 北京光年无限科技有限公司 | Method and device for determining sentence type |
| CN106598947A (en) * | 2016-12-15 | 2017-04-26 | 山西大学 | Bayesian word sense disambiguation method based on synonym expansion |
Non-Patent Citations (2)
| Title |
|---|
| Word Sense Disambiguation Using the;Ho Lee et.al;《Computers and the Humanities》;20000430;全文 * |
| 基于多分类器决策的词义消歧方法;全昌勤 等;《计算机研究与发展》;20060530;第933-939页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108388553A (en) | 2018-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108388553B (en) | Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system | |
| CN108829893B (en) | Method and device for determining video label, storage medium and terminal equipment | |
| CN107818781B (en) | Intelligent interaction method, equipment and storage medium | |
| CN106571140B (en) | Intelligent electric appliance control method and system based on voice semantics | |
| CN107291783B (en) | Semantic matching method and intelligent equipment | |
| CN107480143B (en) | Method and system for segmenting conversation topics based on context correlation | |
| CN107729468B (en) | answer extraction method and system based on deep learning | |
| CN104503998B (en) | For the kind identification method and device of user query sentence | |
| CN103309846B (en) | A kind of processing method of natural language information and device | |
| CN106294774A (en) | User individual data processing method based on dialogue service and device | |
| CN111046656B (en) | Text processing method, text processing device, electronic equipment and readable storage medium | |
| CN110597952A (en) | Information processing method, server, and computer storage medium | |
| CN109637537B (en) | Method for automatically acquiring annotated data to optimize user-defined awakening model | |
| US20170221476A1 (en) | Method and system for constructing a language model | |
| CN111179935B (en) | Voice quality inspection method and device | |
| CN114757176B (en) | Method for acquiring target intention recognition model and intention recognition method | |
| CN109241332B (en) | Method and system for determining semantics through voice | |
| EP3620994A1 (en) | Methods, apparatuses, devices, and computer-readable storage media for determining category of entity | |
| CN110210036A (en) | A kind of intension recognizing method and device | |
| CN110019698A (en) | A kind of intelligent Service method and system of medicine question and answer | |
| CN112699686A (en) | Semantic understanding method, device, equipment and medium based on task type dialog system | |
| CN108710653B (en) | On-demand method, device and system for reading book | |
| CN111353026A (en) | Intelligent law attorney assistant customer service system | |
| CN112364622A (en) | Dialog text analysis method, dialog text analysis device, electronic device and storage medium | |
| CN109992651B (en) | Automatic identification and extraction method for problem target features |
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 |