CN111428506B - Entity classification method, entity classification device and electronic equipment - Google Patents
Entity classification method, entity classification device and electronic equipment Download PDFInfo
- Publication number
- CN111428506B CN111428506B CN202010247706.3A CN202010247706A CN111428506B CN 111428506 B CN111428506 B CN 111428506B CN 202010247706 A CN202010247706 A CN 202010247706A CN 111428506 B CN111428506 B CN 111428506B
- Authority
- CN
- China
- Prior art keywords
- information
- entity
- category
- processed
- search result
- 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
- 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
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present disclosure provides an entity classification method, including: acquiring information to be processed, and extracting entity characteristics in the information to be processed; determining the category of the entity features based on the data set to obtain an initial category corresponding to the entity features; under the condition that the initial category does not meet the preset condition, calling a search engine, searching based on the entity characteristics, and obtaining search result information; determining the category of the entity characteristics at least based on the search result information to obtain a target category corresponding to the entity characteristics; and storing the entity characteristics and the target category into a data set. The disclosure also provides an entity classification device and an electronic device.
Description
Technical Field
The present disclosure relates to an entity classification method, an entity classification apparatus, and an electronic device.
Background
With the rapid development of artificial intelligence, entity identification is required in more and more scenes, and the entity identification includes identifying categories of named entities, for example, in an application scene of a smart client system, a request statement of a user can be obtained, and a user intention can be analyzed according to the request statement, wherein the named entities in the request statement are extracted and classified. At present, classification and identification of named entities are mainly carried out on the basis of a knowledge base or rules, however, the knowledge base and the rules are often not updated timely, and related fields cannot be tracked in time for updating, so that the named entities cannot be classified accurately.
Disclosure of Invention
One aspect of the present disclosure provides an entity classification method, including: acquiring information to be processed, and extracting entity features in the information to be processed; determining the category of the entity feature based on a data set to obtain an initial category corresponding to the entity feature; under the condition that the initial category does not meet the preset condition, calling a search engine, searching based on the entity characteristics, and obtaining search result information; determining the category of the entity characteristics at least based on the search result information to obtain a target category corresponding to the entity characteristics; and storing the entity characteristics and the target categories into the data set.
Optionally, the method further comprises: determining whether recorded information matched with the entity characteristics exists in the data set; determining the category of the entity feature based on the data set under the condition that the record information matched with the entity feature exists in the data set, and obtaining the initial category; and under the condition that the record information matched with the entity features does not exist in the data set, calling a search engine and searching based on the entity features to obtain a target category corresponding to the entity features based on search result information.
Optionally, the obtaining the information to be processed includes: obtaining the information to be processed based on request information input by a user; or network information in a preset time period is obtained, and the information to be processed is obtained from the network information.
Optionally, the method further comprises: determining whether the initial category satisfies a predetermined condition; and in the case that the initial category meets a predetermined condition, taking the initial category as a target category corresponding to the entity feature, wherein the determining whether the initial category meets the predetermined condition comprises: inputting the entity features and the initial categories into a particular neural network model; obtaining score information about the initial category based on a particular neural network model; and determining whether the score information satisfies a predetermined score range, in which case the initial category satisfies the predetermined condition.
Optionally, the performing a search based on the entity characteristics and obtaining search result information includes: searching a webpage by taking the entity characteristics as a search target to obtain search result information about the entity characteristics; or performing web page search by using the entity characteristics and the information to be processed as search targets to obtain search result information about the entity characteristics and the information to be processed.
Optionally, the search result information includes a plurality of entry information; determining the category of the entity feature based on at least the search result information, and obtaining a target category corresponding to the entity feature comprises: extracting at least one item information satisfying a specific condition from the search result information; and inputting the item information into a specific classification model to obtain a classification result, and taking the classification result as a target class corresponding to the entity characteristics.
Optionally, the determining, based on at least the search result information, a category of the entity feature, and obtaining a target category corresponding to the entity feature includes: inputting at least part of entry information in the search result information into a specific classification model to obtain a plurality of classification results; and determining one classification result from the plurality of classification results as a target class based on a predetermined selection condition.
Optionally, the method further comprises: generating response information corresponding to the request information based on the target category of the entity feature; and outputting the response information to the user.
Another aspect of the present disclosure provides an entity classification apparatus, including: the acquisition module is used for acquiring information to be processed; the extraction module is used for extracting entity characteristics in the information to be processed; the searching module is used for calling a searching engine under the condition that the initial category does not meet the preset condition, searching based on the entity characteristics and obtaining searching result information; a classification module for determining the category of the entity feature based on at least the search result information to obtain a target category corresponding to the entity feature, and a storage module for storing the entity feature and the target category in the data set.
Another aspect of the present disclosure provides an electronic device including:
one or more processors;
a memory for storing one or more programs,
wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method as described above.
Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions for implementing the method as described above when executed.
Another aspect of the disclosure provides a computer program comprising computer executable instructions for implementing the method as described above when executed.
Drawings
For a more complete understanding of the present disclosure and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
FIG. 1 schematically illustrates a flow chart of an entity classification method according to an embodiment of the present disclosure;
FIG. 2 schematically shows a schematic diagram of obtaining search result information according to an embodiment of the disclosure;
FIG. 3 schematically illustrates a flow diagram for determining a category of an entity feature based on search result information, in accordance with an embodiment of the present disclosure;
FIG. 4 schematically illustrates a block diagram of an entity classification apparatus according to an embodiment of the present disclosure; and
fig. 5 schematically shows a block diagram of an electronic device adapted to implement the entity classification method according to an embodiment of the present disclosure.
Detailed Description
Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that these descriptions are illustrative only and are not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "at least one of A, B, and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B, and C" would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.). Where a convention analogous to "at least one of A, B, or C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B, or C" would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.).
Some block diagrams and/or flow diagrams are shown in the figures. It will be understood that some blocks of the block diagrams and/or flowchart illustrations, or combinations thereof, 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, or other programmable data processing apparatus, such that the instructions, which execute via the processor, create means for implementing the functions/acts specified in the block diagrams and/or flowchart block or blocks. The techniques of this disclosure may be implemented in hardware and/or software (including firmware, microcode, etc.). In addition, the techniques of this disclosure may take the form of a computer program product on a computer-readable storage medium having instructions stored thereon for use by or in connection with an instruction execution system.
An embodiment of the present disclosure provides an entity classification method, including: and acquiring information to be processed, and extracting entity characteristics in the information to be processed. And determining the category of the entity feature based on the data set to obtain an initial category corresponding to the entity feature. And under the condition that the initial category does not meet the preset condition, calling a search engine, searching based on the entity characteristics, and obtaining search result information. And determining the category of the entity characteristics at least based on the search result information to obtain a target category corresponding to the entity characteristics. And storing the entity characteristics and the target category into a data set.
The entity classification method of the embodiment of the disclosure can be applied to an intelligent customer service system, for example, to obtain a request statement input by a user, and the intelligent customer service system can obtain the request statement input by the user in various ways, for example, a text request statement input by the user can be obtained through a system interface or a voice request statement input by the user can be obtained through a voice input device.
After the request statement input by the user is obtained, the request statement may be analyzed, for example, the named entity in the request statement is extracted, the category to which the named entity belongs is obtained, semantic recognition and other processing are performed according to the category of the named entity, so as to obtain the intention of the user, and then corresponding operation may be performed according to the intention of the user. For example, a user inputs a request statement "i want to download an absolute place for survival", a named entity "absolute place for survival" in the request statement may be extracted, and then a category corresponding to the named entity "absolute place for survival" may be determined, and an entity classification method according to an embodiment of the present disclosure may determine that the category corresponding to "absolute place for survival" is, for example, a network game, and may analyze that the user wants to download a network game, and then may send a website for downloading the network game to the user or directly open a web page or an application for downloading the network game for the user.
The intelligent customer service system is only a scenario example of an entity classification method to which the embodiments of the present disclosure may be applied to help those skilled in the art understand the technical content of the present disclosure, but does not mean that the embodiments of the present disclosure may not be applied to other devices, systems, environments or scenarios.
Fig. 1 schematically shows a flow chart of an entity classification method according to an embodiment of the present disclosure.
As shown in fig. 1, the entity classification method includes operations S110 to S150.
In operation S110, information to be processed is obtained, and entity features in the information to be processed are extracted.
For example, an entity feature may refer to a named entity, which may include a person's name, an organization's name, a place name, a product name, and all other entities identified by names, such as "Mingming," "Beijing university," "Motorola," "PUBG (absolute survival)," and the like.
For example, the information to be processed may include a request statement input by a user, and the request statement input by the user may be obtained in various ways, for example, through an input interface or through a voice input device.
According to the embodiment of the disclosure, after the to-be-processed information is obtained, the named entity in the to-be-processed information is extracted. For example, if the information to be processed is "i want to download PUBG", the named entity "PUBG" in the request sentence may be extracted, and if the information to be processed is "i want to purchase motorola", the named entity "motorola" in the request sentence may be extracted.
In operation S120, a category of the entity feature is determined based on the data set, resulting in an initial category corresponding to the entity feature. The method includes determining whether record information matched with the entity features exists in a data set, and determining the category of the entity features based on the data set under the condition that the record information matched with the entity features exists in the data set to obtain an initial category.
For example, a data set, such as a knowledge base, may be pre-stored. The data set may record a plurality of named entities and categories corresponding to the named entities. The category corresponding to the name entity may include, for example: computers, mobile phones, games, life, food, travel, social contacts, information, and the like.
After the named entity in the information to be processed is obtained, whether the named entity or an entity close to the named entity is recorded or not can be searched from the data set, if so, the category of the named entity can be determined according to the recorded information in the data set, and the category determined according to the data set can be used as an initial category.
In operation S130, in case that the initial category does not satisfy the predetermined condition, a search engine is invoked to perform a search based on the entity characteristics, and search result information is obtained.
For example, based on the accuracy of determining the initial category, when the accuracy of the initial category is lower than a certain value, the initial category may be considered not to satisfy the predetermined condition. Wherein the search result information may include a plurality of entry information.
Fig. 2 schematically shows a schematic diagram of obtaining search result information according to an embodiment of the present disclosure.
As shown in fig. 2, for example, if the entity feature is "PUBG", a search engine may be invoked to search for "PUBG", and multiple search result entries about "PUBG" are obtained.
In operation S140, a category of the entity feature is determined based on at least the search result information, resulting in a target category corresponding to the entity feature.
For example, specific information in the search result information, such as a title and a text, may be extracted, the specific information may be subjected to processing such as feature extraction, the processed specific information may be input into a classification model obtained through pre-training, and an output result of the classification model may be used as a target category corresponding to the entity feature.
In operation S150, the entity characteristics and the object categories are stored into a data set.
And storing the entity characteristics and the corresponding target types into the data set, and updating records in the data set or adding new records to the data set.
According to the embodiment of the disclosure, the entity features can be used as search input, a search engine is called to obtain the search result about the entity features, and then the entity features are classified according to the search result, so that the category of the entity features can be accurately judged. And after the accurate classification result is obtained, the named entity and the accurate classification result thereof can be stored in a knowledge base so as to facilitate subsequent classification work.
According to the embodiment of the present disclosure, in operation S120, it may be determined whether the data set includes record information that matches the entity feature, and if the data set does not include record information that matches the entity feature, a search engine may be directly invoked and a search may be performed based on the entity feature, and a target category corresponding to the entity feature is obtained based on search result information.
According to an embodiment of the present disclosure, obtaining the information to be processed may include: obtaining information to be processed based on request information input by a user; or network information in a preset time period is obtained, and information to be processed is obtained from the network information.
The information to be processed may be obtained from network information in addition to being obtained based on user input. The network information may include, for example, news article information, forum information, and the like. The predetermined time period may refer to a time period from a current time to a previous time, for example, a last week, a last day, a last ten hours, and the like. After the network information in the predetermined time period is acquired, the network information may be directly used as the information to be processed, or specific information in the network information may be extracted as the information to be processed, for example, news headlines and news texts in news articles may be extracted as the information to be processed. The above-described operations S110 to S150 are then performed on the acquired information to be processed.
According to the embodiment of the disclosure, network information in a preset time period is obtained, information to be processed is obtained from the network information, named entities in the information to be processed are extracted, and categories corresponding to the named entities are determined based on a data set and/or a webpage search result.
According to an embodiment of the present disclosure, in operation S130, the method may further include: it is determined whether the initial category satisfies a predetermined condition.
Determining whether the initial category meets the predetermined condition may include: (1) Inputting the entity characteristics and the initial category into a specific neural network model; (2) Obtaining score information about the initial category based on the specific neural network model; and (3) determining whether the score information satisfies a predetermined score range, in which case the initial category satisfies a predetermined condition.
For example, a neural network model for accuracy scoring of classes may be trained in advance using multiple named entities and multiple classes. For example, where the named entity is "motorola" and the category is "cell phone," the accuracy score for the category may be 95; in the case where the named entity is "motorola" and the category is "vehicle", the accuracy score for the category may be 60; in the case where the named entity is "motorola" and the category is "fruit", the accuracy score for the category may be 5. And training by utilizing the pre-acquired named entities, categories and scores to obtain the specific neural network model.
In operation S120, an initial class corresponding to the entity feature is obtained, and the entity feature and the initial class corresponding to the entity feature can be input into the specific neural network model to obtain a score of the initial class. The predetermined score range may be, for example, 80 to 100, and if the score corresponding to the initial category does not belong to the predetermined score range, the initial category may be considered to not satisfy the predetermined condition, and if the score of the initial category belongs to the predetermined score range, the initial category may be considered to satisfy the predetermined condition.
Under the condition that the initial category does not meet the preset condition, a search engine can be called, and searching is carried out based on the entity characteristics to obtain search result information; and then determining the category of the entity characteristics based on the search result information to obtain a target category corresponding to the entity characteristics.
In the case where the initial category satisfies a predetermined condition, the initial category may be taken as a target category corresponding to the entity feature.
According to an embodiment of the present disclosure, in operation S130, performing a search based on the entity characteristics, and obtaining the search result information may include: searching a webpage by taking the entity characteristics as a search target to obtain search result information about the entity characteristics; or the entity characteristics and the information to be processed are used as search targets to search for webpages, and search result information about the entity characteristics and the information to be processed is obtained.
For example, if the information to be processed is "i want to download PUBG", the entity feature of the information to be processed is "PUBG", and during the search, the web page search may be performed only by using the entity feature "PUBG" as a search statement, or the entity feature "PUBG" + the information to be processed "i want to download PUBG" may be used together as a search statement to perform the web page search.
FIG. 3 schematically shows a flow diagram for determining a category of an entity feature based on search result information according to an embodiment of the disclosure.
As shown in fig. 3, determining a category of the entity feature based on at least the search result information in operation S140, and obtaining a target category corresponding to the entity feature may include operations S141 to S142 according to an embodiment of the present disclosure.
At least one item information satisfying a specific condition is extracted from the search result information in operation S141.
The at least one item information satisfying the specific condition may refer to, for example, top N item information located in the search result, where N is a positive integer.
After at least one item information meeting the specific condition is obtained, operations such as feature extraction and the like can be performed on the item information to obtain features corresponding to the item information.
In operation S142, at least one item information is input into the specific classification model, a classification result is obtained, and the classification result is used as a target class corresponding to the entity feature.
For example, a classification model may be established in advance, and the classification model may be trained based on search results corresponding to a plurality of named entities and categories corresponding to the named entities.
The feature corresponding to the item information obtained in operation S141 may be input into the classification model, and the classification model may output a classification result, which may be a target class corresponding to the entity feature.
According to an embodiment of the present disclosure, determining a category of the entity feature based on at least the search result information in operation S140, and obtaining the target category corresponding to the entity feature may further include: inputting at least part of entry information in the search result information into a specific classification model to obtain a plurality of classification results; and determining one classification result from the plurality of classification results as a target class based on a predetermined selection condition.
According to the embodiments of the present disclosure, one named entity may correspond to multiple categories, for example, "apple" may refer to a mobile phone and may also refer to a fruit, and therefore, an accurate classification of the named entity needs to be obtained in combination with an application scenario to know the true intention of a user.
The classification model obtained by the pre-training is, for example, a model capable of outputting a plurality of classification results, for example, if the named entity to be classified is "apple", the classification model may output a plurality of classes with higher probability, for example, may output two classes with the highest probability: the mobile phone and the fruit. One classification result may then be determined as the target class from the plurality of classification results based on a predetermined selection condition. In which the predetermined selection condition may be determined according to actual situations, for example, if the entity classification method is applied to an intelligent customer service system in the field of electronic products, the classification result related to the electronic product may be selected as the target class, in which case, following the above example, the "mobile phone" may be used as the target class for naming the entity "apple".
According to an embodiment of the present disclosure, the entity classification method may further include: generating response information corresponding to the request information based on the target category of the entity characteristics; and outputting the response information to the user.
The information to be processed being a request input by the userWhen a statement is found, a response needs to be given to the request statement. After the category of each named entity in the request statement is obtained, semantic recognition can be performed on the request statement according to the category of each named entity to obtain the intention of the user, then a corresponding response statement is generated, and the response statement is output to the user. For example, the request statement is "i want to download absolutely necessary survival", the named entity "absolutely necessary survival" in the request statement may be extracted, the category corresponding to "absolutely necessary survival" is determined to be the online game according to the entity classification method of the embodiment of the present disclosure, and it may be analyzed that the user wants to download an online game. The user may then be presented with a website to download the online game, for example, an answer sentence "download dead game if necessary, please go tohttp：//XXXXX.comDownload "is performed and the answer sentence is presented to the user. Alternatively, a web page or application for downloading the network game may be directly opened for the user, for example, a response sentence "good, a web page for downloading the network game is now opened for you", the response sentence is output to the user, and the corresponding web page is opened.
According to the embodiment of the disclosure, under the condition that one named entity corresponds to multiple categories, the classification model can output several categories with higher probability, and then one category meeting the tendency condition can be selected as a target category, so that the correct classification of the named entity can be obtained by combining application scenes, and the real intention of a user can be obtained.
Another aspect of the embodiments of the present disclosure further provides an entity classification apparatus.
Fig. 4 schematically shows a block diagram of an entity classification apparatus according to an embodiment of the present disclosure.
As shown in fig. 4, the entity classification apparatus 200 includes an acquisition module 210, an extraction module 220, a search module 230, a classification module 240, and a storage module 250.
The obtaining module 210 is used for obtaining information to be processed.
The extracting module 220 is used for extracting entity features in the information to be processed.
The search module 230 is configured to, when the initial category does not satisfy the predetermined condition, invoke a search engine to perform a search based on the entity characteristics, and obtain search result information.
The classification module 240 is configured to determine a category of the entity feature based on at least the search result information, and obtain a target category corresponding to the entity feature.
The storage module 250 is used to store the entity characteristics and the object categories into a data set.
According to an embodiment of the present disclosure, the entity classification apparatus may further include a first determining module, where the first determining module is configured to: determining whether recorded information matched with the entity characteristics exists in the data set; determining the category of the entity feature based on the data set under the condition that the record information matched with the entity feature exists in the data set, and obtaining the initial category; and under the condition that the record information matched with the entity features does not exist in the data set, calling a search engine and searching based on the entity features to obtain a target category corresponding to the entity features based on search result information.
According to an embodiment of the present disclosure, the obtaining module 210 is configured to: obtaining the information to be processed based on request information input by a user; or network information in a preset time period is obtained, and the information to be processed is obtained from the network information.
According to an embodiment of the present disclosure, the entity classification apparatus may further include a second determining module, where the second determining module is configured to: determining whether the initial category satisfies a predetermined condition; and in the case that the initial category satisfies a predetermined condition, regarding the initial category as a target category corresponding to the entity feature,
wherein the determining whether the initial category meets a predetermined condition comprises: inputting the entity features and the initial categories into a particular neural network model; obtaining score information about the initial category based on a particular neural network model; and determining whether the score information satisfies a predetermined score range, in which case the initial category satisfies the predetermined condition.
According to an embodiment of the present disclosure, the search module 230 is further configured to: performing webpage search by taking the entity characteristics as a search target to obtain search result information about the entity characteristics; or performing web page search by using the entity characteristics and the information to be processed as search targets to obtain search result information about the entity characteristics and the information to be processed.
According to an embodiment of the present disclosure, the search result information includes a plurality of entry information. The classification module 240 is further configured to: extracting at least one item information satisfying a specific condition from the search result information; and inputting the at least one item of information into a specific classification model to obtain a classification result, and taking the classification result as a target class corresponding to the entity characteristics.
According to the embodiment of the disclosure, at least part of entry information in the search result information is input into a specific classification model to obtain a plurality of classification results; and determining one classification result from the plurality of classification results as a target class based on a predetermined selection condition.
Any number of modules, sub-modules, units, sub-units, or at least part of the functionality of any number thereof according to embodiments of the present disclosure may be implemented in one module. Any one or more of the modules, sub-modules, units, and sub-units according to the embodiments of the present disclosure may be implemented by being split into a plurality of modules. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented at least in part as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in any other reasonable manner of hardware or firmware by integrating or packaging a circuit, or in any one of or a suitable combination of software, hardware, and firmware implementations. Alternatively, one or more of the modules, sub-modules, units, sub-units according to embodiments of the disclosure may be at least partially implemented as a computer program module, which when executed may perform the corresponding functions.
For example, any number of the obtaining module 210, the extracting module 220, the searching module 230, the classifying module 240, the storing module 250, the first determining module, and the second determining module may be combined and implemented in one module, or any one of the modules may be split into a plurality of modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the obtaining module 210, the extracting module 220, the searching module 230, the classifying module 240, the storing module 250, the first determining module, and the second determining module may be at least partially implemented as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented by hardware or firmware in any other reasonable manner of integrating or packaging a circuit, or implemented by any one of three implementations of software, hardware, and firmware, or by a suitable combination of any of them. Alternatively, at least one of the obtaining module 210, the extracting module 220, the searching module 230, the classifying module 240, the storing module 250, the first determining module, and the second determining module may be at least partially implemented as a computer program module, which when executed, may perform a corresponding function.
Fig. 5 schematically shows a block diagram of an electronic device adapted to implement the above described method according to an embodiment of the present disclosure. The electronic device shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 5, computer system 500 includes a processor 510, a computer-readable storage medium 520. The electronic device 500 may perform a method according to an embodiment of the present disclosure.
In particular, processor 510 may include, for example, a general purpose microprocessor, an instruction set processor and/or related chip set and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), and/or the like. The processor 510 may also include on-board memory for caching purposes. Processor 510 may be a single processing unit or a plurality of processing units for performing different actions of a method flow according to an embodiment of the present disclosure.
Computer-readable storage media 520, for example, may be non-volatile computer-readable storage media, specific examples including, but not limited to: magnetic storage devices, such as magnetic tape or Hard Disk Drives (HDDs); optical storage devices, such as compact disks (CD-ROMs); a memory, such as a Random Access Memory (RAM) or a flash memory; and so on.
The computer-readable storage medium 520 may include a computer program 521, which computer program 521 may include code/computer-executable instructions that, when executed by the processor 510, cause the processor 510 to perform a method according to an embodiment of the disclosure, or any variation thereof.
The computer program 521 may be configured with, for example, computer program code comprising computer program modules. For example, in an example embodiment, code in computer program 521 may include one or more program modules, including 521A, modules 521B, \8230;, for example. It should be noted that the division and number of modules are not fixed, and those skilled in the art may use suitable program modules or program module combinations according to actual situations, and when these program modules are executed by the processor 510, the processor 510 may execute the method according to the embodiment of the present disclosure or any variation thereof.
According to an embodiment of the present invention, at least one of the obtaining module 210, the extracting module 220, the searching module 230, the classifying module 240, the storing module 250, the first determining module, and the second determining module may be implemented as a computer program module described with reference to fig. 5, which, when executed by the processor 510, may implement the respective operations described above.
The present disclosure also provides a computer-readable storage medium, which may be contained in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It will be appreciated by those skilled in the art that various combinations and/or combinations of the features recited in the various embodiments of the disclosure and/or the claims may be made even if such combinations or combinations are not explicitly recited in the disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments of the present disclosure and/or the claims may be made without departing from the spirit and teachings of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
While the disclosure has been shown and described with reference to certain exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims and their equivalents. Accordingly, the scope of the present disclosure should not be limited to the above-described embodiments, but should be defined not only by the appended claims, but also by equivalents thereof.
Claims (9)
1. An entity classification method comprising:
obtaining information to be processed based on request information input by a user; or
Acquiring network information in a preset time period, and acquiring the information to be processed from the network information; the step of obtaining the information to be processed from the network information is to directly use the network information as the information to be processed or extract specific information in the network information as the information to be processed;
extracting entity features in the information to be processed;
determining the category of the entity feature based on a data set to obtain an initial category corresponding to the entity feature;
under the condition that the initial category does not meet the preset condition, calling a search engine, searching based on the entity characteristics, and obtaining search result information;
determining the category of the entity feature at least based on the search result information to obtain a target category corresponding to the entity feature;
and storing the entity characteristics and the target categories into the data set.
2. The method of claim 1, further comprising:
determining whether recorded information matched with the entity characteristics exists in the data set;
determining the category of the entity feature based on the data set under the condition that the record information matched with the entity feature exists in the data set, and obtaining the initial category; and
and under the condition that the record information matched with the entity features does not exist in the data set, calling a search engine and searching based on the entity features so as to obtain a target category corresponding to the entity features based on the search result information.
3. The method of claim 1, further comprising:
determining whether the initial category satisfies a predetermined condition; and
if the initial category meets a predetermined condition, taking the initial category as a target category corresponding to the entity feature,
wherein the determining whether the initial category meets a predetermined condition comprises:
inputting the entity features and the initial categories into a particular neural network model;
obtaining score information about the initial category based on a particular neural network model; and
determining whether the score information satisfies a predetermined score range, in which case the initial category satisfies the predetermined condition.
4. The method of claim 1, wherein the searching based on the entity characteristics, obtaining search result information comprises:
searching a webpage by taking the entity characteristics as a search target to obtain search result information about the entity characteristics; or alternatively
And searching a webpage by taking the entity characteristics and the information to be processed as search targets to obtain search result information about the entity characteristics and the information to be processed.
5. The method of claim 1, wherein:
the search result information includes a plurality of entry information;
determining the category of the entity feature based on at least the search result information, and obtaining a target category corresponding to the entity feature comprises:
extracting at least one item information satisfying a specific condition from the search result information; and
and inputting the at least one item information into a specific classification model to obtain a classification result, and taking the classification result as a target class corresponding to the entity characteristics.
6. The method of claim 1, wherein the determining a category of the entity feature based at least on the search result information, the deriving a target category corresponding to the entity feature comprises:
inputting at least part of entry information in the search result information into a specific classification model to obtain a plurality of classification results; and
determining one classification result from the plurality of classification results as a target class based on a predetermined selection condition.
7. The method of claim 2, further comprising:
generating response information corresponding to the request information based on the target category of the entity characteristics; and
and outputting the response information to the user.
8. An entity classification apparatus comprising:
the acquisition module is used for acquiring information to be processed based on request information input by a user; the information to be processed is obtained from the network information, namely the network information is directly used as the information to be processed or specific information in the network information is extracted as the information to be processed; or
Acquiring network information in a preset time period, and acquiring the information to be processed from the network information;
the extraction module is used for extracting entity characteristics in the information to be processed;
the searching module is used for calling a searching engine under the condition that the initial category does not meet the preset condition, searching based on the entity characteristics and obtaining searching result information;
the classification module is used for determining the category of the entity characteristics at least based on the search result information to obtain a target category corresponding to the entity characteristics; and
and the storage module is used for storing the entity characteristics and the target categories into a data set.
9. An electronic device, comprising:
one or more processors;
a computer-readable storage medium for storing one or more programs,
wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010247706.3A CN111428506B (en) | 2020-03-31 | 2020-03-31 | Entity classification method, entity classification device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010247706.3A CN111428506B (en) | 2020-03-31 | 2020-03-31 | Entity classification method, entity classification device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111428506A CN111428506A (en) | 2020-07-17 |
| CN111428506B true CN111428506B (en) | 2023-02-21 |
Family
ID=71550381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010247706.3A Active CN111428506B (en) | 2020-03-31 | 2020-03-31 | Entity classification method, entity classification device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111428506B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112163415A (en) * | 2020-09-30 | 2021-01-01 | 北京猎豹移动科技有限公司 | User intention identification method and device for feedback content and electronic equipment |
| CN112735124B (en) * | 2020-12-16 | 2022-05-20 | 阿波罗智联(北京)科技有限公司 | Traffic data analysis method, device, equipment, vehicle and storage medium |
| CN113254635B (en) * | 2021-04-14 | 2021-11-05 | 腾讯科技（深圳）有限公司 | Data processing method, device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617239A (en) * | 2013-11-26 | 2014-03-05 | 百度在线网络技术（北京）有限公司 | Method and device for identifying named entity and method and device for establishing classification model |
| CN104615687A (en) * | 2015-01-22 | 2015-05-13 | 中国科学院计算技术研究所 | Entity fine granularity classifying method and system for knowledge base updating |
| CN110032625A (en) * | 2019-03-28 | 2019-07-19 | 腾讯科技（上海）有限公司 | A kind of interactive method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9443021B2 (en) * | 2011-12-30 | 2016-09-13 | Microsoft Technology Licensing, Llc | Entity based search and resolution |
| CN107480450A (en) * | 2017-08-15 | 2017-12-15 | 京东方科技集团股份有限公司 | A kind of intelligence point examines method and system |
-
2020
- 2020-03-31 CN CN202010247706.3A patent/CN111428506B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617239A (en) * | 2013-11-26 | 2014-03-05 | 百度在线网络技术（北京）有限公司 | Method and device for identifying named entity and method and device for establishing classification model |
| CN104615687A (en) * | 2015-01-22 | 2015-05-13 | 中国科学院计算技术研究所 | Entity fine granularity classifying method and system for knowledge base updating |
| CN110032625A (en) * | 2019-03-28 | 2019-07-19 | 腾讯科技（上海）有限公司 | A kind of interactive method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111428506A (en) | 2020-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110582025B (en) | Method and apparatus for processing video | |
| CN111428506B (en) | Entity classification method, entity classification device and electronic equipment | |
| US20200034370A1 (en) | Internet search result intention | |
| CN111984779B (en) | Dialogue text analysis method, device, equipment and readable medium | |
| CN108763952B (en) | Data classification method and device and electronic equipment | |
| US10997249B2 (en) | Search query intent | |
| US10956469B2 (en) | System and method for metadata correlation using natural language processing | |
| CN110245475B (en) | Identity verification method and device | |
| CN112949297B (en) | Intent recognition method, apparatus, computer device, and computer-readable storage medium | |
| CN110322281B (en) | Similar user mining method and device | |
| US20160171111A1 (en) | Method and system to detect use cases in documents for providing structured text objects | |
| US11182605B2 (en) | Search device, search method, search program, and recording medium | |
| CN109299276B (en) | Method and device for converting text into word embedding and text classification | |
| CN111435369B (en) | Music recommendation method, device, terminal and storage medium | |
| CN116882372A (en) | Text generation method, device, electronic equipment and storage medium | |
| US20200065589A1 (en) | Automatic tagging of images using speech recognition | |
| US9619558B2 (en) | Method and system for entity recognition in a query | |
| US11574024B2 (en) | Method and system for content bias detection | |
| CN117409419A (en) | Image detection method, device and storage medium | |
| CN109492401B (en) | Content carrier risk detection method, device, equipment and medium | |
| CN112818206A (en) | Data classification method, device, terminal and storage medium | |
| CN107609094B (en) | Data disambiguation method and device and computer equipment | |
| CN114139547B (en) | Knowledge fusion method, device, equipment, system and medium | |
| US9910921B2 (en) | Keyword refinement in temporally evolving online media | |
| CN112836612B (en) | Method, device and system for user real-name authentication |
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 |