CN110795924A - Method and device for constructing intention template library and storage medium - Google Patents
Method and device for constructing intention template library and storage medium Download PDFInfo
- Publication number
- CN110795924A CN110795924A CN202010003689.9A CN202010003689A CN110795924A CN 110795924 A CN110795924 A CN 110795924A CN 202010003689 A CN202010003689 A CN 202010003689A CN 110795924 A CN110795924 A CN 110795924A
- Authority
- CN
- China
- Prior art keywords
- template
- result
- processing result
- processing
- data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012545 processing Methods 0.000 claims abstract description 101
- 238000000926 separation method Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000013135 deep learning Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 6
- 230000006698 induction Effects 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000002372 labelling Methods 0.000 claims description 4
- 238000012384 transportation and delivery Methods 0.000 description 5
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 210000001503 joint Anatomy 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and a device for constructing an intention template library and a computer storage medium, wherein firstly, marking data are obtained; then, the acquired marking data is subjected to data processing by using a named entity recognition model to obtain a processing result; then, generating a template for the processing result according to a specific generation rule; and finally, storing the generated template into an intention template library.
Description
Technical Field
The invention relates to the field of text information processing, in particular to a method and a device for constructing an intention template library and a computer storage medium.
Background
Currently, in intelligent short message processing, an intelligent short message paradigm system is widely applied. The cumulative coverage of the known model in the short message template is more than 70%.
However, the following obvious defects exist in the normal form generation process of the current intelligent short message normal form system: 1) the semantic template is manufactured manually, so that the efficiency is low and the generalization performance is poor; 2) the method belongs to an end-to-end solution completely depending on deep learning, cannot be well applied to a mobile phone end, and has high requirements on performance, RAM/ROM and the like.
Disclosure of Invention
The embodiment of the invention provides a method and a device for constructing an intention template library and a computer storage medium for solving the potential problems in the intelligent short message processing process.
According to a first aspect of the present invention, there is provided an intention template library construction method, the method comprising: acquiring label data; performing data processing on the acquired labeling data by using a named entity recognition model to obtain a processing result; generating a template for the processing result according to a specific generation rule; and storing the generated template into an intention template library.
According to an embodiment of the present invention, the performing data processing on the obtained tagged data by using the named entity recognition model to obtain a processing result includes: carrying out named entity recognition model recognition on the marked data to obtain a recognition result; and classifying the identification result to obtain a processing result.
According to an embodiment of the present invention, the classifying the recognition result to obtain a processing result includes: if the identification result is value, replacing the value with @ key @; if the recognition result is not value, replacing the recognition result with @ Ner _ type @; if the recognition result is that the Ner does not recognize the data, replacing the Ner does not recognize the data with @ key @; and if the identification result is an unprocessed number, performing regular normalization processing on the number.
According to an embodiment of the present invention, the generating a template for the processing result according to a specific generation rule includes: searching characteristics in specific threshold character lengths at two sides of each @ key @ in the processing result; and generating a template according to the searched features.
According to an embodiment of the present invention, the generating a template for the processing result according to a specific generation rule includes: clustering all texts in the processing result to generate a cluster ID and a corresponding text paradigm; performing separation processing on the head paradigm and other paradigms in all the generated text paradigms to obtain a separation processing result; searching features in specific threshold character lengths on two sides of each @ key @accordingto a paradigm corresponding to each cluster ID in the separation processing result; and generating a template according to the searched features.
According to an embodiment of the present invention, storing the generated template into an intent template library includes: performing uniqueness screening on all the generated templates, and reserving the quasi-template with the shortest character length corresponding to each key; and storing all quasi templates into an intention template library.
According to an embodiment of the invention, the method further comprises: classifying all keys in advance in a manual induction mode to obtain key classification results; and carrying out deep learning training on the obtained key classification result to obtain a named entity recognition model.
According to the second aspect of the present invention, there is also provided an intention template library constructing apparatus including: the acquisition module is used for acquiring the marking data; the data processing module is used for carrying out data processing on the acquired labeling data by utilizing the named entity recognition model to obtain a processing result; the template generation module is used for generating a template for the processing result according to a specific generation rule; and the storage module is used for storing the generated template into the intention template library.
According to an embodiment of the present invention, the data processing module includes: the named entity recognition model recognition unit is used for carrying out named entity recognition model recognition on the labeled data to obtain a recognition result; and the classification processing unit is used for performing classification processing on the identification result to obtain a processing result.
According to an embodiment of the present invention, the classification processing unit is specifically configured to, if the identification result is a value, replace the value with @ key @; if the recognition result is not value, replacing the recognition result with @ Ner _ type @; if the recognition result is that the Ner does not recognize the data, replacing the Ner does not recognize the data with @ key @; and if the identification result is an unprocessed number, performing regular normalization processing on the number.
According to an embodiment of the present invention, the template generating module is specifically configured to search for a feature in a specific threshold character length at two sides of each @ key @inthe processing result; and generating a template according to the searched features.
According to an embodiment of the present invention, the template generating module is specifically configured to cluster all texts in the processing result to generate a cluster ID and a corresponding text paradigm; performing separation processing on the head paradigm and other paradigms in all the generated text paradigms to obtain a separation processing result; searching features in specific threshold character lengths on two sides of each @ key @accordingto a paradigm corresponding to each cluster ID in the separation processing result; and generating a template according to the searched features.
According to an embodiment of the present invention, the storage module is specifically configured to perform uniqueness screening on all generated templates, and reserve a quasi-template with a shortest character length corresponding to each key; and storing all quasi templates into an intention template library.
According to an embodiment of the invention, the apparatus further comprises: the named entity recognition model training module is used for classifying all keys in advance in a manual induction mode to obtain key classification results; and carrying out deep learning training on the obtained key classification result to obtain a named entity recognition model.
According to a third aspect of the present invention, there is provided a computer storage medium having stored therein computer-executable instructions for performing the intention template library construction method of any one of the above when executed.
According to the method, the device and the computer storage medium for constructing the intention template library, disclosed by the embodiment of the invention, firstly, marking data are obtained; then, the acquired marking data is subjected to data processing by using a named entity recognition model to obtain a processing result; then, generating a template for the processing result according to a specific generation rule; and finally, storing the generated template into an intention template library. Therefore, the method can automatically generate a fine-grained generalization semantic template, combines a pre-trained named entity recognition deep learning model (namely, a named entity recognition model), can effectively control the performance and has lower requirements on RAM/ROM, and can ensure the high accuracy of the model and simultaneously give consideration to the high efficiency and the generalization requirements; and the model can be updated according to the requirement, seamless butt joint is realized, and the maintenance is easy.
It is to be understood that the teachings of the present invention need not achieve all of the above-described benefits, but rather that specific embodiments may achieve specific technical results, and that other embodiments of the present invention may achieve benefits not mentioned above.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a schematic diagram illustrating an implementation flow of a method for building an intent template library according to an embodiment of the present invention;
FIG. 2 is a first flowchart illustrating a specific implementation of a method for constructing an application instance intention template library according to the present invention;
FIG. 3 is a flowchart II illustrating a specific implementation of the method for constructing an intent template library according to another application example of the present invention;
fig. 4 is a schematic diagram illustrating a composition structure of an intention template library construction device according to an embodiment of the present invention.
Detailed Description
The principles and spirit of the present invention will be described with reference to a number of exemplary embodiments. It is understood that these embodiments are given only to enable those skilled in the art to better understand and to implement the present invention, and do not limit the scope of the present invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The technical solution of the present invention is further elaborated below with reference to the drawings and the specific embodiments.
FIG. 1 shows a schematic flow chart of an implementation of a method for building an intent template library according to an embodiment of the present invention. Referring to fig. 1, a method for constructing an intent template library according to an embodiment of the present invention includes: operation 101, acquiring annotation data; operation 102, performing data processing on the acquired tagged data by using the named entity identification model to obtain a processing result; operation 103, performing template generation on the processing result according to a specific generation rule; at operation 104, the generated template is stored in an intent template library.
Before the implementation of operations 101 to 104 in the embodiment of the present invention, the method further includes: classifying all keys in advance in a manual induction mode to obtain key classification results; and carrying out deep learning training on the obtained key classification result to obtain a named entity recognition model.
Specifically, the manual summary of all keys is in several broad categories, such as: departure time, arrival time, etc. - - - > time; departure place, receiving place, etc. - - - > address, etc. And further carrying out deep learning training on the obtained key classification result to obtain a named entity recognition model with a higher F1 value.
In operation 102, firstly, named entity recognition model recognition is performed on the tagged data to obtain a recognition result; and further classifying the identification result to obtain a processing result. The classifying the recognition result specifically comprises the following steps: if the identification result is value, replacing the value with @ key @; if the recognition result is not value, replacing the recognition result with @ Ner _ type @; if the recognition result is that the Ner does not recognize the data, replacing the Ner does not recognize the data by @ key @; and if the identification result is an unprocessed number, performing regular normalization processing on the number.
For example, the labeled data before processing is: your express delivery gives you a post cabinet to pay attention to check and receive, and the number of the express deliveries is 1, and every century express delivery is realized; fetching an address: post cabinet, express company: and (6) all-time express delivery. The processing result is as follows: your express gives you the @ pick-up address @ please pay attention to check and collect, quantity @ Ner Number @ @ express company @.
In operation 103, referring to the application example shown in fig. 2, a feature may be first searched for in the specific threshold character lengths on both sides of each @ key @ in the processing result, and then template generation may be performed according to the searched feature. Wherein, the specific threshold character length may take a value of 12.
For example, according to the processing result "your express delivers @ pickup address @ for you" please pay attention to check and accept, the Number @ Ner Number @, @ express company @ pickup address @ ", features are found at both sides of each @ key @ in the processing result, and a template is generated: put @ # # # fetch address; the quantity @ NER _ Number @, @ # delivery company.
In operation 103, referring to the application example shown in fig. 3, the present invention introduces a text clustering template in order to improve the generalization capability of the template; respectively generating template libraries according to each cluster, namely dividing the original total template library into a plurality of independent small template libraries to reduce the number of templates; in the application stage, similarity matching is firstly carried out on the template and the clustering template, and the template is determined to enter the small template library for matching. Specifically, as shown in fig. 3, in the implementation of operation 103, first, all texts in the processing result are clustered, and a cluster ID and a corresponding text paradigm are generated; further performing separation processing on the head paradigm and other paradigms in all the generated text paradigms to obtain a separation processing result; searching features in specific threshold character lengths on two sides of each @ key @accordingto a normal form corresponding to each cluster ID in the separation processing result; and further generating a template according to the searched features. Therefore, when the text classification is used for carrying out a scene recognition task, the generated fine-grained semantic template is used as the feature of the text classification, and the operation can improve the accuracy of the classification model, so that the volume of the whole model can be controlled after feature screening.
In operation 104, the uniqueness screening may be performed on all the generated templates, and the quasi-template with the shortest character length corresponding to each key is reserved; and storing all the quasi templates into an intention template library. Thus, the optimal template of each entity key can be ensured to be stored in the template library.
The method for constructing the intention template library comprises the steps of firstly, acquiring annotation data; then, the acquired marking data is subjected to data processing by using a named entity recognition model to obtain a processing result; then, generating a template for the processing result according to a specific generation rule; and finally, storing the generated template into an intention template library. Therefore, the method can automatically generate a fine-grained generalization semantic template, combines a pre-trained named entity recognition deep learning model (namely, a named entity recognition model), can effectively control the performance and has lower requirements on RAM/ROM, and can ensure the high accuracy of the model and simultaneously give consideration to the high efficiency and the generalization requirements; and the model can be updated according to the requirement, seamless butt joint is realized, and the maintenance is easy.
Also, based on the intention template library construction method as described above, an embodiment of the present invention further provides a computer-readable storage medium storing a program that, when executed by a processor, causes the processor to perform at least the operation steps of: operation 101, acquiring annotation data; operation 102, performing data processing on the acquired tagged data by using the named entity identification model to obtain a processing result; operation 103, performing template generation on the processing result according to a specific generation rule; at operation 104, the generated template is stored in an intent template library.
Further, based on the above-mentioned intention template library construction method, an embodiment of the present invention further provides an intention template library construction apparatus, as shown in fig. 4, where the apparatus 40 includes: an obtaining module 401, configured to obtain annotation data; a data processing module 402, configured to perform data processing on the obtained tagged data by using a named entity identification model to obtain a processing result; a template generating module 403, configured to perform template generation on the processing result according to a specific generating rule; a storage module 404, configured to store the generated template in the intent template library.
According to an embodiment of the present invention, the data processing module 402 includes: the named entity recognition model recognition unit is used for carrying out named entity recognition model recognition on the labeled data to obtain a recognition result; and the classification processing unit is used for performing classification processing on the identification result to obtain a processing result.
According to an embodiment of the present invention, the classification processing unit is specifically configured to, if the identification result is a value, replace the value with @ key @; if the recognition result is not value, replacing the recognition result with @ Ner _ type @; if the recognition result is that the Ner does not recognize the data, replacing the Ner does not recognize the data with @ key @; and if the identification result is an unprocessed number, performing regular normalization processing on the number.
According to an embodiment of the present invention, the template generating module 403 is specifically configured to search for features in specific threshold character lengths at two sides of each @ key @inthe processing result; and generating a template according to the searched features.
According to an embodiment of the present invention, the template generating module 403 is specifically configured to cluster all texts in the processing result, and generate a cluster ID and a corresponding text paradigm; performing separation processing on the head paradigm and other paradigms in all the generated text paradigms to obtain a separation processing result; searching features in specific threshold character lengths on two sides of each @ key @accordingto a paradigm corresponding to each cluster ID in the separation processing result; and generating a template according to the searched features.
According to an embodiment of the present invention, the storage module 404 is specifically configured to perform uniqueness screening on all generated templates, and reserve a quasi-template with a shortest character length corresponding to each key; and storing all quasi templates into an intention template library.
According to an embodiment of the present invention, the apparatus 40 further comprises: the named entity recognition model training module is used for classifying all keys in advance in a manual induction mode to obtain key classification results; and carrying out deep learning training on the obtained key classification result to obtain a named entity recognition model.
Here, it should be noted that: the above description of the embodiment of the apparatus is similar to the description of the embodiment of the method shown in fig. 1 to 3, and has similar beneficial effects to the embodiment of the method shown in fig. 1 to 3, and therefore, the description is omitted. For technical details that are not disclosed in the embodiments of the apparatus of the present invention, please refer to the description of the method embodiments shown in fig. 1 to 3 for understanding, and therefore, for brevity, will not be described again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. 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 apparatus that comprises the element.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read Only Memory (ROM), a magnetic disk, or an optical disk.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a magnetic or optical disk, or other various media that can store program code.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (9)
1. A method for constructing an intention template library, the method comprising:
acquiring label data;
performing data processing on the acquired labeling data by using a pre-trained named entity recognition model to obtain a processing result;
generating a template for the processing result according to a specific generation rule;
storing the generated template into an intention template library;
the template generation of the processing result according to a specific generation rule comprises the following steps: clustering all texts in the processing result to generate a cluster ID and a corresponding text paradigm; performing separation processing on the head paradigm and other paradigms in all the generated text paradigms to obtain a separation processing result; searching features in specific threshold character lengths on two sides of each @ key @accordingto a paradigm corresponding to each cluster ID in the separation processing result; and generating a template according to the searched features.
2. The method according to claim 1, wherein the performing data processing on the obtained tagged data using the named entity recognition model to obtain a processing result comprises:
carrying out named entity recognition model recognition on the marked data to obtain a recognition result;
and classifying the identification result to obtain a processing result.
3. The method according to claim 2, wherein the classifying the recognition result to obtain a processing result comprises:
if the identification result is value, replacing the value with @ key @;
if the recognition result is not value, replacing the recognition result with @ Ner _ type @;
if the recognition result is that the Ner does not recognize the data, replacing the Ner does not recognize the data with @ key @;
and if the identification result is an unprocessed number, performing regular normalization processing on the number.
4. The method according to claim 1, wherein the template generation of the processing result according to a specific generation rule comprises:
searching characteristics in specific threshold character lengths at two sides of each @ key @ in the processing result;
and generating a template according to the searched features.
5. The method of claim 4, wherein storing the generated template into an intent template library comprises:
performing uniqueness screening on all the generated templates, and reserving the quasi-template with the shortest character length corresponding to each key;
and storing all quasi templates into an intention template library.
6. The method according to any one of claims 1 to 5, further comprising:
classifying all keys in advance in a manual induction mode to obtain key classification results;
and carrying out deep learning training on the obtained key classification result to obtain a named entity recognition model.
7. An intent template library construction apparatus, the apparatus comprising:
the acquisition module is used for acquiring the marking data;
the data processing module is used for carrying out data processing on the acquired labeling data by utilizing the pre-trained named entity recognition model to obtain a processing result;
the template generation module is used for generating a template for the processing result according to a specific generation rule;
the storage module is used for storing the generated template into an intention template library;
the template generating module is specifically used for clustering all texts in the processing result to generate a cluster ID and a corresponding text paradigm; performing separation processing on the head paradigm and other paradigms in all the generated text paradigms to obtain a separation processing result; searching features in specific threshold character lengths on two sides of each @ key @accordingto a paradigm corresponding to each cluster ID in the separation processing result; and generating a template according to the searched features.
8. The apparatus of claim 7, wherein the data processing module comprises:
the named entity recognition model recognition unit is used for carrying out named entity recognition model recognition on the labeled data to obtain a recognition result;
and the classification processing unit is used for performing classification processing on the identification result to obtain a processing result.
9. A computer storage medium having computer-executable instructions stored thereon that, when executed, perform the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010003689.9A CN110795924B (en) | 2020-01-03 | 2020-01-03 | Method and device for constructing intention template library and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010003689.9A CN110795924B (en) | 2020-01-03 | 2020-01-03 | Method and device for constructing intention template library and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110795924A true CN110795924A (en) | 2020-02-14 |
| CN110795924B CN110795924B (en) | 2020-04-07 |
Family
ID=69448455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010003689.9A Active CN110795924B (en) | 2020-01-03 | 2020-01-03 | Method and device for constructing intention template library and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110795924B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170293871A1 (en) * | 2016-04-08 | 2017-10-12 | International Business Machines Corporation | Generation of an optimization model |
| CN110096594A (en) * | 2019-04-29 | 2019-08-06 | 北京泰迪熊移动科技有限公司 | A kind of short message normal form library generating method, device and computer memory device |
| CN110390006A (en) * | 2019-07-23 | 2019-10-29 | 腾讯科技（深圳）有限公司 | Question and answer corpus generation method, device and computer readable storage medium |
-
2020
- 2020-01-03 CN CN202010003689.9A patent/CN110795924B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170293871A1 (en) * | 2016-04-08 | 2017-10-12 | International Business Machines Corporation | Generation of an optimization model |
| CN110096594A (en) * | 2019-04-29 | 2019-08-06 | 北京泰迪熊移动科技有限公司 | A kind of short message normal form library generating method, device and computer memory device |
| CN110390006A (en) * | 2019-07-23 | 2019-10-29 | 腾讯科技（深圳）有限公司 | Question and answer corpus generation method, device and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110795924B (en) | 2020-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112685565B (en) | Text classification method based on multi-mode information fusion and related equipment thereof | |
| CN106649818B (en) | Application search intention identification method and device, application search method and server | |
| CN108345692B (en) | Automatic question answering method and system | |
| CN102779140B (en) | A kind of keyword acquisition methods and device | |
| CN108920675B (en) | Information processing method and device, computer storage medium and terminal | |
| CN107153641A (en) | Comment information determines method, device, server and storage medium | |
| US20200134398A1 (en) | Determining intent from multimodal content embedded in a common geometric space | |
| CN111291566A (en) | Event subject identification method and device and storage medium | |
| CN101561725B (en) | Method and system of fast handwriting input | |
| CN111563382A (en) | Text information acquisition method and device, storage medium and computer equipment | |
| CN101689198A (en) | Phonetic search using normalized string | |
| CN113051384B (en) | User portrait extraction method based on dialogue and related device | |
| CN110347696B (en) | Data conversion method, device, computer equipment and storage medium | |
| CN110795924B (en) | Method and device for constructing intention template library and storage medium | |
| CN112417996A (en) | Information processing method and device for industrial drawing, electronic equipment and storage medium | |
| CN114637831A (en) | Data query method based on semantic analysis and related equipment thereof | |
| CN113434631A (en) | Emotion analysis method and device based on event, computer equipment and storage medium | |
| KR20220079029A (en) | Method for providing automatic document-based multimedia content creation service | |
| CN111753548A (en) | Information acquisition method and device, computer storage medium and electronic equipment | |
| US11790045B2 (en) | Auto-tags with object detection and crops | |
| CN112347768A (en) | Entity identification method and device | |
| KR20220079060A (en) | Resource database device for document-based video resource matching and multimedia conversion content production | |
| KR20220079057A (en) | Method for building a resource database of a multimedia conversion content production service providing device | |
| CN114580420A (en) | Chinese entity identification method, device, equipment and medium | |
| CN117972117A (en) | Search method, search device, electronic 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: East of 1st floor, No.36 Haidian Street, Haidian District, Beijing, 100080 Patentee after: Beijing Teddy Future Technology Co.,Ltd. Address before: East of 1st floor, No.36 Haidian Street, Haidian District, Beijing, 100080 Patentee before: Beijing Teddy Bear Mobile Technology Co.,Ltd. |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: East of 1st floor, No.36 Haidian Street, Haidian District, Beijing, 100080 Patentee after: Beijing Teddy Bear Mobile Technology Co.,Ltd. Address before: 100085 07a36, block D, 7 / F, No.28, information road, Haidian District, Beijing Patentee before: BEIJING TEDDY BEAR MOBILE TECHNOLOGY Co.,Ltd. |