CN117454004A - Sample data generation method, model training method and search result determination method - Google Patents

Abstract

The disclosure provides a sample data generation method, relates to the technical field of artificial intelligence, and particularly relates to the technical field of deep learning and intelligent recommendation. The specific implementation scheme is as follows: identifying a core word of the target object from the search statement; identifying industry information of the target object according to the core word; determining a service type of the search statement according to at least one of the core word and the industry information, and taking the service type of the search statement as a type tag of the search statement, wherein the type tag comprises one of a first service type strongly related to enterprise-oriented service, a second service type cross-related to enterprise-oriented service and personal-oriented service, and a third service type unrelated to enterprise-oriented service; and determining the search statement with the type tag as sample data. The disclosure also provides a training method of the deep learning model, a search result determining method, a device, electronic equipment and a storage medium.

Description

Sample data generation method, model training method and search result determination method

Technical Field

The disclosure relates to the technical field of artificial intelligence, in particular to the technical field of deep learning and intelligent recommendation. More specifically, the present disclosure provides a sample data generating method, a training method of a deep learning model, a search result determining method, an apparatus, an electronic device, and a storage medium.

Background

ToB (Business to Business, enterprise-oriented) business refers to business targeting enterprise-level users, such as wholesale channel purchasing business, and the like. For the search statement input by the user on the search platform, whether the search statement is biased to ToB service or the strength of the ToB service is identified, and the page can be produced for the user in a targeted manner.

Disclosure of Invention

The present disclosure provides a sample data generating method, apparatus, device, and storage medium.

According to a first aspect, there is provided a sample data generation method, the method comprising: identifying a core word of the target object from the search statement; identifying industry information of the target object according to the core word; determining a service type of the search statement according to at least one of the core word and the industry information, and taking the service type of the search statement as a type tag of the search statement, wherein the type tag comprises one of a first service type strongly related to enterprise-oriented service, a second service type cross-related to enterprise-oriented service and personal-oriented service, and a third service type unrelated to enterprise-oriented service; and determining the search statement with the type tag as sample data.

According to a second aspect, there is provided a training method of a deep learning model, the method comprising: acquiring sample data, wherein the sample data is generated according to the training method of the deep learning model; and training the deep learning model by using the sample data to obtain the deep learning model for identifying the service type of the search statement.

According to a third aspect, there is provided a search result determination method, the method comprising: acquiring an input sentence; identifying a business type of the input sentence using the trained deep learning model, wherein the business type is one of a first business type strongly associated with the enterprise-oriented business, a second business type cross-associated with the enterprise-oriented business and the personal-oriented business, and a third business type unrelated to the enterprise-oriented business; generating a search result page according to the service type; the deep learning model is obtained by training according to the training method of the deep learning model.

According to a fourth aspect, there is provided a sample data generating apparatus comprising: the core word determining module is used for identifying the core word of the target object from the search statement; the industry information determining module is used for identifying the industry information of the target object according to the core word; the business type determining module is used for determining the business type of the search statement according to at least one of the core word and the industry information and taking the business type of the search statement as a type tag of the search statement, wherein the type tag comprises one of a first business type strongly associated with the enterprise-oriented business, a second business type cross-associated with the enterprise-oriented business and the personal-oriented business and a third business type irrelevant with the enterprise-oriented business; and a sample data generation module for determining the search statement with the type tag as sample data.

According to a fifth aspect, there is provided a training apparatus of a deep learning model, the apparatus comprising: the sample acquisition module is used for acquiring sample data, wherein the sample data is generated according to the sample data generation device; and the training module is used for training the deep learning model by using the sample data to obtain the deep learning model for identifying the service type of the search statement.

According to a sixth aspect, there is provided a search result determination apparatus comprising: the input sentence acquisition module is used for acquiring an input sentence; a processing module for identifying a business type of the input sentence using the trained deep learning model, wherein the business type is one of a first business type strongly associated with the enterprise-oriented business, a second business type cross-associated with the enterprise-oriented business and the personal-oriented business, and a third business type unrelated to the enterprise-oriented business; the search result determining module is used for generating a search result page according to the service type; the deep learning model is obtained by training according to the training device of the deep learning model.

According to a seventh aspect, there is provided an electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method provided in accordance with the present disclosure.

According to an eighth aspect, there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform a method provided according to the present disclosure.

According to a ninth aspect, there is provided a computer program product comprising a computer program stored on at least one of a readable storage medium and an electronic device, which, when executed by a processor, implements a method provided according to the present disclosure.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1 is a schematic diagram of an exemplary system architecture to which at least one of a sample data generation method, a training method of a deep learning model, a search result determination method may be applied, according to one embodiment of the present disclosure;

FIG. 2 is a flow chart of a sample data generation method according to one embodiment of the present disclosure;

FIG. 3 is a schematic diagram of a method of business type identification of a search term according to one embodiment of the present disclosure;

FIG. 4 is a flow chart of a training method of a deep learning model according to one embodiment of the present disclosure;

FIG. 5 is a flow chart of a search result determination method according to one embodiment of the present disclosure;

FIG. 6 is a schematic diagram of a method of determining a traffic type of a search statement according to one embodiment of the disclosure;

FIG. 7 is a block diagram of a sample data generating device according to one embodiment of the present disclosure;

FIG. 8 is a block diagram of a training apparatus of a deep learning model according to one embodiment of the present disclosure;

FIG. 9 is a block diagram of a search result determination apparatus according to one embodiment of the present disclosure;

fig. 10 is a block diagram of an electronic device of at least one of a sample data generating method, a training method of a deep learning model, a search result determining method according to one embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

ToB (Business to Business, enterprise-oriented) business refers to business targeting enterprise-level users, such as wholesale channel purchasing business, and the like. The ToC (To Customer) service refers To a service targeting individual users. For the search statement input by the user on the search platform, whether the search statement is biased to the ToB service or the strength of the ToB service is identified, and the service type of the search statement can be determined. For example, the traffic types of the search statement may include a strong ToB traffic type, a medium ToB traffic type, a non-ToB traffic type, and so on.

Currently, the identification of the type of service of a search statement includes the following several ways.

One method of searching for the business type of a sentence is to train a classification model using manually labeled data. For example, manually marking the service type of the sample Query. A model for identifying the type of service of the search term is trained using the annotation data. The manual labeling method is high in cost, a large number of samples are difficult to obtain, and therefore the accuracy and recall rate of the trained model are difficult to improve.

One search requirement recognition method is to directly infer the service type using a pre-trained ToB Query recognition model. For example, a pre-trained ToB Query recognition model is used to recognize whether a Query belongs to a ToB service type, and if so, the determined service type of the Query is the ToB service type. Otherwise, determining that the service type of the Query is a non-ToB service type. However, this ToB Query recognition model can only classify Query into ToB traffic type and non-ToB traffic type, and cannot meet the classification requirements herein. And, the classification accuracy and recall rate are not high enough.

In view of this, embodiments of the present disclosure provide a method for identifying a service type of a search statement, which identifies that a Query searched by a user belongs to the strength of a ToB service type. And embodiments of the present disclosure more finely divide the intensity of Query belonging to ToB traffic types into strong ToB traffic types, medium ToB traffic types, and non-ToB traffic types. The Query belonging to the ToB service type, the Query belonging to the ToB and ToC cross service type, and the Query not belonging to the ToB service type are defined. Compared with the division of ToB service types and non-ToB service types in the related art, after the ToB strength of the Query is identified to belong to one of the strong ToB service types, the medium ToB service types and the non-ToB service types, the method can display the page and the customized service individually for the Query with different strengths, so that the user experience is improved, and the coverage rate, the flow rate ratio and the flow guiding magnitude of the platform are improved.

The method for identifying the service type of the search statement provided by the embodiment of the disclosure relates to a sample data generating method, a training method of a deep learning model and a search result determining method, and the methods are described below.

In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.

In the technical scheme of the disclosure, the authorization or consent of the user is obtained before the personal information of the user is obtained or acquired.

FIG. 1 is a schematic diagram of an exemplary system architecture to which at least one of a sample data generation method, a training method of a deep learning model, a search result determination method may be applied, according to one embodiment of the present disclosure. It should be noted that fig. 1 is only an example of a system architecture to which embodiments of the present disclosure may be applied to assist those skilled in the art in understanding the technical content of the present disclosure, but does not mean that embodiments of the present disclosure may not be used in other devices, systems, environments, or scenarios.

As shown in fig. 1, a system architecture 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired and/or wireless communication links, and the like.

At least one of the sample data generating method and the search result determining method provided by the embodiments of the present disclosure may be generally performed by the terminal devices 101, 102, 103. Accordingly, at least one of the sample data generating means and the search result determining means provided by the embodiments of the present disclosure may be generally provided in the terminal device 101, 102, 103.

The training method of the deep learning model provided by the embodiments of the present disclosure may be generally performed by the server 105. Accordingly, the training apparatus of the deep learning model provided by the embodiments of the present disclosure may be generally disposed in the server 105.

Fig. 2 is a flow chart of a sample data generation method according to one embodiment of the present disclosure.

As shown in fig. 2, the sample data generating method 200 includes operations S210 to S240.

In operation S210, a core word of a target object is identified from a search sentence.

The search statement, query, may be a collection of search records for a large number of users. The core words in Query include entity words of the target object (e.g., the article body), as well as entity words, modifier words, etc. associated with the target object, such as brand names, place name, domain names, brand recommendation information, etc. The target object may be an item (e.g., a mechanical product, an electronic product, etc.) or a service (e.g., an advertisement, a marketing service, etc.). Specifically, a core word in the Query can be identified by using a word labeling model.

In operation S220, industry information to which the target object belongs is identified according to the core word.

The industry to which the target object belongs can be identified based on the core word using a pre-trained industry classification model. For example, core words of the target object (e.g., target object body, brand name, domain name, etc.) may be input into a pre-trained industry classification model to obtain the industry to which the target object belongs.

Specifically, the primary industry, the secondary industry, and the like of the target object can also be finely identified. For example, the primary industry may be manufacturing, the secondary industry may be machine manufacturing, and so on.

In operation S230, a service type of the search term is determined as a type tag of the search term according to at least one of the core word and the industry information.

For example, the traffic types are classified into a first traffic type strongly associated with enterprise-oriented traffic (strong ToB traffic type), a second traffic type cross-associated with enterprise-oriented traffic and personal-oriented traffic (medium ToB traffic type), and a third traffic type unrelated to enterprise-oriented traffic (non-ToB traffic type).

Correspondingly, category ranges corresponding to the respective business types may be set. For example, the category range of the strong ToB business type includes articles sold for business, goods services, and the like. The category range of ToB business types in (1) includes items where ToB crosses ToC, such as ToC category items that may have wholesale purchase needs. The category range of non-ToB traffic types may include the following: one is a ToC class item without wholesale purchase demand; one is an item that is not normally sold by the ToB platform. And also may be article independent such as movies, entertainment, information, etc.

According to the embodiment of the disclosure, matching core words and industry information with the category range of each business type; and determining the service type of the search statement according to the category range hit by at least one of the core word and the industry information.

For example, after determining the category scope of each business type, the core word and industry information may be matched with the category scope of each type, and which category scope of each type the core word and industry information hit may determine which business type the search term belongs to.

In operation S240, a search statement having a type tag is determined as sample data.

After obtaining the type tag of the Query (strong ToB traffic type, medium ToB traffic type, non-ToB traffic type), the Query with the type tag can be taken as a sample. And training an end-to-end classification model by using the sample, wherein the classification model has the capability of identifying the service type of the Query.

According to the embodiment of the disclosure, the core words and the corresponding industry information in the search statement are identified, the service types are determined based on the core words and the industry information and are used as type labels, sample data are obtained, training of the model is conveniently carried out by using the sample data, a deep learning model for identifying the service types of the search statement is obtained, and therefore the identification efficiency, recall rate and accuracy rate of the service types of the search statement are improved.

The division of the category ranges for each service type is described below.

According to an embodiment of the present disclosure, the industry information includes a first industry category (strong ToB industry category) strongly associated with the business-oriented business, a second industry category (middle ToB industry category) cross-associated with the business-oriented business and the personal business-oriented, and a third industry category (ToC industry category) strongly associated with the personal business-oriented; dividing object entities belonging to a first business category into category ranges of the first business type; dividing object entities under a second industry class into class ranges of a second service type; and dividing the object entities under the third industry category, the object entities irrelevant to all the industry categories and the object entities without transaction attributes under all the industry categories into the category range of the third business type.

For example, the ToB intensity classification criteria may be explicit based on industry information. For object entities obviously belonging to ToB industry classes, the object entities are divided into class ranges of strong ToB service types. Including, for example, articles sold for business, business services, rental operation services, second hand vehicles, and transportation vehicles, among others.

For an object entity that belongs to the category of the ToC industry, but has ToB business requirements (e.g., wholesale procurement requirements), it can be classified into a category range of ToB business types. Including, for example, consumer goods, furniture, office supplies, and the like.

In addition to the two cases described above, the category range of the non-ToB traffic type can be divided. For example, it includes three cases, one is a ToC-like article such as cosmetics, jewelry, etc., which does not require wholesale purchase. One is that there are no transaction attributes, i.e., items that the platform is not generally vending, such as cigarettes, western medicines, non-secondhand cell phones, cars, etc. And also can be object entities unrelated to articles, such as film and television works, life emotion problems and the like.

It can be understood that, in this implementation, for all kinds of items covered by the search platform, the types of strong ToB service, medium ToB service and non-ToB service are classified according to the service types, and for a search sentence input by a user, if a core word or industry information in the search sentence hits a range of a certain service type, it can be determined that the service type of the search sentence is the service type corresponding to the hit range of the service type.

Fig. 3 is a schematic diagram of a method of traffic type identification of a search term according to one embodiment of the present disclosure.

As shown in fig. 3, the present embodiment includes a Query service type identification module 301 based on core words, a Query service type identification module 302 based on industry information, a special class Query identification module 303, and a search requirement identification module 304.

The respective identification modules are explained below.

First, the Query service type recognition module 301 based on the core word will be described. Core words in Query can be identified using the lexical labeling model. And matching the core word with the category range of each service type to determine the service type of the search statement.

The Query service type identification module 302 based on industry information is described below.

And identifying the industry information of the Query by using an industry classification model. Then matching the industry of the Query with the category range of each business type, wherein the category range of each business type is divided based on industry information, so that the category in each category range corresponds to the industry information. If the industry to which the Query belongs hits the industry information of a certain class, and if the core word of the Query is combined with the object in the class, the service type corresponding to the hit class is the service type of the Query.

Next, the special class Query recognition module 303 will be described. After the Query business type identification module 302 based on industry information identifies the industry information, some special categories may be mixed with normal categories, which cannot be distinguished only by the industry information.

According to an embodiment of the present disclosure, the first industry category includes a first special category range; the special class Query identification includes determining, based on the core word of the target object, that the service type of the search statement is one of a first service type and a third service type in response to the target object belonging to the first special class scope.

Aiming at the mixed situation in the strong ToB industry. For example, trucks and cars are both in the new industry, trucks are strong ToB items, and platforms do not sell cars. For this case, some strong ToB industry categories (first special category range) where special categories may exist may be defined, if Query belongs to the first special category range, the core word of Query is input into the existing ToB Query recognition model, if recognition is successful, the core word is classified as a strong ToB service type, otherwise, the core word is classified as a special category in the strong ToB industry, and is classified as a non-ToB service type.

According to an embodiment of the present disclosure, the second industry category includes a second special category; the special class Query identification includes matching a core word of the target object with an object entity in a second special category range in response to the target object belonging to the second special category range; and dividing the service type of the search statement into one of a second service type and a third service type according to the matching result.

For mixed cases in the ToB industry. For example, both paper towels and cigarettes are in the daily consumer industry, paper towels are medium ToB items, while cigarettes are items that the platform does not sell. For such cases, some middle ToB industry categories (second special category ranges) may be defined for which special categories may exist, and if Query belongs to the second special category range, regular matching is used to determine if Query hits the second special category range. If the Query hits the second special category range (e.g., paper towel), then the Query is determined to be a strong ToB service type, otherwise the Query is a special category in the ToB industry, and is classified as a non-ToB service type.

The search requirement identification module 304 is described below.

According to an embodiment of the present disclosure, a search requirement of a search sentence is identified according to at least one of core words and industry information, wherein the search requirement includes one of a sourcing class, a content class, and other classes; and determining the service type of the search statement as a third service type in response to the search requirement of the search statement being of the other class.

For some Query, while the body portion is related to the item, the intended need for the overall expression is not within the reach of the platform. For example, if Query is for disease control of agricultural crops, although belonging to the agriculture, forestry, animal and fish industry, the knowledge of disease control is not what the platform target covers, so such Query should be of non-ToB business type.

The search requirements of Query may be identified using a trained requirement identification model. Search requirements may include sourcing classes, content classes, and other classes. The source searching type refers to the purchasing price inquiring requirement of the user, the content type refers to the information searching requirement of the user, and the searching requirements except the two requirements are classified into other types.

For example, the search requirement identification module 304 identifies a search requirement for a Query using a trained requirement identification model, and if the search requirement is of another class, the traffic type for the Query may be determined to be a non-ToB traffic type. For the situation that Query inquires about disease control of agricultural crops, the Query can be well identified as a non-ToB business type by using a demand identification model.

In the embodiment, the service types of the sample Query are classified by using a plurality of pre-training models and a regular matching method, so that the service type label of the sample Query is obtained, the manual marking with high use cost is avoided, the problem of poor identification effect of a single pre-training model is also avoided, and the identification accuracy of the Query of each service type is improved.

Fig. 4 is a flowchart of a training method of a deep learning model according to one embodiment of the present disclosure.

As shown in fig. 4, the training method 400 of the deep learning model includes operations S410 to S420.

In operation S410, sample data is acquired.

In operation S420, a deep learning model is trained using the sample data, resulting in a deep learning model for identifying a traffic type of the search statement.

Sample data is generated according to the sample data generation method described above. The sample data has a type tag, and the traffic type tag of each sample data is one of a strong ToB traffic type, a medium ToB traffic type, and a non-ToB traffic type.

Training is performed on the basis of the pre-training language model by using sample data, so that a trained deep learning model is obtained, and the trained deep learning model has the capability of identifying the service type of Query. The pre-trained language model may be a Bert model.

The Query with the business type label is used for model training, and the model contains the information of semantic similarity among texts after the model is pre-trained by massive text data, so that the model is used for fine tuning training, the model and the method used in the sample generation stage can be combined with semantic information stored in the pre-training process, and the trained model can achieve higher classification accuracy and recall rate than the sample. Another advantage is that the model can make end-to-end predictions, which can be used for on-line classification predictions with greatly improved speed of inference compared to methods using multiple models in the sample generation stage.

According to the embodiment of the disclosure, for samples used in training, the number proportion can be set for samples of different service type labels, so that the model reaches a balanced standard rate.

Specifically, most of the whole sample Query is related to education, movies, life and the like, and is not related to articles, so that the generated sample is most of the Query which is not of ToB business type. And the number of Query of the middle ToB service type is obviously smaller than that of the Query of the strong ToB service type. When training with such unbalanced class samples, the most significant problem is that the majority of class samples get more attention during training, while the minority of class samples do not get enough attention during training, and finally do not reach satisfactory accuracy and recall.

In this embodiment, a certain balance is performed on the ratio of the Query types. The strong ToB and non-ToB Query may be downsampled such that, after sampling, the ratio of the number of strong ToB, medium ToB, non-ToB in the sample is a preset ratio, for example, 1:1:2. when the model is trained by using the samples with the category proportion, the model can achieve balanced accuracy and recall rate for three categories.

Fig. 5 is a flow chart of a search result determination method according to one embodiment of the present disclosure.

As shown in fig. 5, the training method 500 of the deep learning model includes operations S510 to S520.

In operation S510, an input sentence is acquired.

In operation S520, a service type of the input sentence is identified using the trained deep learning model. The deep learning model is obtained by training according to the training method of the deep learning model.

In operation S530, a search result page is generated according to the service type.

The input statement Query may be entered by a user at the search platform. The input sentence is input into the trained deep learning model, and the service type of the input sentence can be obtained, wherein the service type is one of a strong ToB service type, a medium ToB service type and a non-ToB service type.

According to the service type of the input statement Query, pages can be displayed and produced for users in a targeted manner. For example, for an input statement Query of a strong ToB business type, a large number of pages of the ToB business type, such as related pages for purchase, price enquiry, brand recommendation, etc., may be produced. For the input statement Query of ToB service type, a page of partial ToB service type can be produced. For input statement Query of non-ToB service type, the page of ToB service type may not be produced.

According to the embodiment, the service type is determined according to the input statement, and the page can be displayed and produced for the user in a targeted manner, so that the user experience, the coverage rate of the platform, the flow rate duty ratio and the diversion magnitude are improved.

According to an embodiment of the present disclosure, the method further comprises: identifying a core word of an input object from an input sentence; identifying industry information of the input object according to the core word; determining a search requirement of an input sentence according to at least one of core words and industry information, wherein the search requirement is one of a source searching class, a content class and other classes; and generating a search result page according to the service type and the search requirement of the input sentence.

The core words in the Query can be identified by using the word labeling model, and the core words are input into the pre-trained industry classification model to obtain the industry information of the Query. The intention recognition can be carried out on the search sentences according to the core words, so that the search requirement of the search sentences is obtained; and determining the search requirement of the search statement according to the matching rule corresponding to the industry information.

For example, a pre-trained intent recognition model may be used to perform intent recognition based on core words, recognize the user's needs for purchase, price-ask, question-answer, text, video, etc., determine the purchase and price-ask needs as a sourcing class, and determine the question-answer, text, and video needs as a content class.

For example, a corresponding content recognition rule and a source-seeking recognition rule may be set for each industry, and for a Query whose target object belongs to a certain industry, whether the Query is a content class or a source-seeking class is determined using the content recognition rule and the source-seeking recognition rule. For Query that does not fall within the content range, nor the source range, the type of demand may be determined to be other classes.

Having identified the search requirements of Query, a more refined and accurate page can be produced for the user. For example, for a Query that is of the strong ToB business type and is a sourcing requirement, more pages containing wholesale purchase channel information may be produced. For Query, which is of a non-ToB traffic type and is a content requirement, more pages containing content introductions can be produced.

According to the implementation, according to the input sentences and the service types and the demand categories, pages are displayed and produced for the user in a targeted manner, the user searching demands can be met more precisely, and therefore user experience is improved.

Fig. 6 is a schematic diagram of a method of determining a traffic type of a search statement according to one embodiment of the disclosure.

As shown in fig. 6, the present embodiment includes a sample generation stage, a model refinement stage, and a traffic type prediction stage. And obtaining a Query service type label by using the service type identification method of the search statement, and generating a sample Query. The model after fine tuning is obtained by using a sample Query fine tuning pre-trained language model (for example, a Bert model). And carrying out demand recognition on the Query newly input by the user by using the fine-tuning model to obtain whether the service type of the search statement is strong ToB, medium ToB or non-ToB.

Fig. 7 is a block diagram of a sample data generating device according to one embodiment of the present disclosure.

As shown in fig. 7, the sample data generating apparatus 700 includes a core word determining module 701, an industry information determining module 702, a service type determining module 703, and a sample data generating module 704.

The core word determining module 701 is configured to identify a core word of a target object from the search statement.

The industry information determining module 702 is configured to identify, according to the core word, industry information to which the target object belongs.

The service type determining module 703 is configured to determine a service type of the search statement according to at least one of the core word and the industry information, as a type tag of the search statement, where the type tag includes one of a first service type strongly associated with the enterprise-oriented service, a second service type cross-associated with the enterprise-oriented service and the personal-oriented service, and a third service type unrelated to the enterprise-oriented service.

The sample data generation module 704 is configured to determine a search statement with a type tag as sample data.

The traffic type determination module 703 comprises a matching sub-module and a first determination sub-module.

The matching sub-module is used for matching the core words and industry information with the category range of each business type.

The first determining submodule is used for determining the service type of the search statement according to the category range hit by at least one of the core word and the industry information.

According to an embodiment of the present disclosure, the industry information includes a first industry category strongly associated with the enterprise-oriented business, a second industry category cross-associated with the enterprise-oriented business and the personal-oriented business, and a third industry category strongly associated with the personal-oriented business. The sample data generating device further comprises a first dividing module, a second dividing module and a third dividing module.

The first dividing module is used for dividing the object entity belonging to the first business category into category ranges of the first business type.

The second dividing module is used for dividing the object entity under the second industry class into the class range of the second service type.

The third dividing module is used for dividing the object entity under the third industry category, the object entity irrelevant to all the industry categories and the object entity without transaction attribute under all the industry categories into category ranges of the third business type.

The first industry category includes a first special category range; the sample data generating device further comprises a first screening module, wherein the first screening module is used for responding that the target object belongs to the first special category range, and determining that the service type of the search statement is one of a first service type and a third service type according to the core word of the target object.

The second industry category includes a second special category; the sample data generating device further comprises a second screening module, wherein the second screening module is used for responding that the target object belongs to the second special category range and matching the core word of the target object with the object entity in the second special category range; and dividing the service type of the search statement into one of a second service type and a third service type according to the matching result.

The traffic type determination module 703 includes a search demand determination sub-module and a second determination sub-module.

The search requirement determination submodule is used for identifying the search requirement of the search statement according to at least one of the core word and the industry information, wherein the search requirement comprises one of a source searching class, a content class and other classes.

The second determining submodule is used for determining the service type of the search statement as a third service type in response to the search requirement of the search statement as other types.

Fig. 8 is a block diagram of a training apparatus of a deep learning model according to one embodiment of the present disclosure.

As shown in fig. 8, the training apparatus 800 of the deep learning model includes a sample acquisition module 801 and a training module 802.

The sample acquisition module 801 is configured to acquire sample data, where the sample data is generated according to the sample data generating device.

Training module 802 is configured to train a deep learning model using the sample data to obtain a deep learning model that is used to identify a type of business for the search term.

Fig. 9 is a block diagram of a search result determination apparatus according to one embodiment of the present disclosure.

As shown in fig. 9, the search result determination apparatus 900 includes an input sentence acquisition module 901, a processing module 902, and a search result determination module 903.

The input sentence acquisition module 901 is configured to acquire an input sentence.

The processing module 902 is configured to identify a traffic type of the input sentence using the trained deep learning model, wherein the traffic type is one of a first traffic type strongly associated with the enterprise-oriented traffic, a second traffic type cross-associated with the enterprise-oriented traffic and the personal-oriented traffic, and a third traffic type unrelated to the enterprise-oriented traffic.

The search result determining module 903 is configured to generate a search result page according to a service type;

the deep learning model is obtained by training according to the training device of the deep learning model.

According to an embodiment of the present disclosure, the search result determining apparatus 900 further includes an input object core word determining module, an input object industry information determining module, and a search requirement determining module.

The input object core word determining module is used for identifying the core word of the input object from the input sentence.

The input object industry information determining module is used for identifying industry information of the input object according to the core word.

The search requirement determining module is used for determining the search requirement of the input sentence according to at least one of the core word and the industry information, wherein the search requirement is one of a source searching class, a content class and other classes.

The search result determining module 903 is configured to generate a search result page according to the service type and the search requirement of the input sentence.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

Fig. 10 shows a schematic block diagram of an example electronic device 1000 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 10, the apparatus 1000 includes a computing unit 1001 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a Random Access Memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other by a bus 1004. An input/output (I/O) interface 1005 is also connected to bus 1004.

Various components in device 1000 are connected to I/O interface 1005, including: an input unit 1006 such as a keyboard, a mouse, and the like; an output unit 1007 such as various types of displays, speakers, and the like; a storage unit 1008 such as a magnetic disk, an optical disk, or the like; and communication unit 1009 such as a network card, modem, wireless communication transceiver, etc. Communication unit 1009 allows device 1000 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks.

The computing unit 1001 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 1001 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The calculation unit 1001 performs the respective methods and processes described above, for example, at least one of a sample data generation method, a training method of a deep learning model, and a search result determination method. For example, in some embodiments, at least one of the sample data generation method, the training method of the deep learning model, the search result determination method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 1008. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 1000 via ROM 1002 and/or communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of at least one of the sample data generating method, the training method of the deep learning model, and the search result determining method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to perform at least one of the sample data generation method, the training method of the deep learning model, the search result determination method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, 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), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims (23)

1. A sample data generation method, comprising:

identifying a core word of the target object from the search statement;

identifying industry information of the target object according to the core word;

determining a service type of the search statement as a type tag of the search statement according to at least one of the core word and the industry information, wherein the type tag comprises one of a first service type strongly associated with enterprise-oriented services, a second service type cross-associated with enterprise-oriented services and personal-oriented services, and a third service type unrelated to enterprise-oriented services; and

the search statement with the type tag is determined as sample data.

2. The method of claim 1, wherein the determining the type of service of the search term based on at least one of the core word and the industry information comprises:

Matching the core word and the industry information with the category range of each business type; and

and determining the service type of the search statement according to the category range hit by at least one of the core word and the industry information.

3. The method of claim 2, wherein the industry information includes a first industry category strongly associated with enterprise-oriented business, a second industry category cross-associated with enterprise-oriented business and personal-oriented business, and a third industry category strongly associated with personal-oriented business; the method further comprises the steps of:

dividing object entities belonging to the first business category into category ranges of the first business type;

dividing the object entity under the second industry class into class ranges of the second business type;

and dividing the object entities under the third industry category, the object entities irrelevant to all the industry categories and the object entities without transaction attributes under all the industry categories into category ranges of the third business type.

4. A method according to claim 3, wherein the first industry category comprises a first special category range; the method further comprises the steps of:

And responding to the target object belonging to the first special category range, and determining that the service type of the search statement is one of a first service type and a third service type according to the core word of the target object.

5. The method of claim 3 or 4, wherein the second industry category comprises a second special category; the method further comprises the steps of:

responding to the fact that the target object belongs to the second special category range, and matching the core word of the target object with an object entity in the second special category range; and

and dividing the service type of the search statement into one of a second service type and a third service type according to the matching result.

6. The method of any of claims 1-5, the determining a business type of the search statement from at least one of the core word and the industry information comprising:

identifying a search requirement of the search statement according to at least one of the core word and the industry information, wherein the search requirement comprises one of a sourcing class, a content class and other classes; and

and determining the service type of the search statement as a third service type in response to the search requirement of the search statement being other types.

7. A training method of a deep learning model, comprising:

obtaining sample data, wherein the sample data is generated according to the method of any one of claims 1 to 6; and

and training a deep learning model by using the sample data to obtain the deep learning model for identifying the service type of the search statement.

8. A search result determination method, comprising:

acquiring an input sentence;

identifying a business type of the input sentence using a trained deep learning model, wherein the business type is one of a first business type strongly associated with an enterprise-oriented business, a second business type cross-associated with an enterprise-oriented business and a personal-oriented business, and a third business type unrelated to the enterprise-oriented business; and

generating a search result page according to the service type;

wherein the deep learning model is trained according to the method of claim 7.

9. The method of claim 8, the method further comprising:

identifying a core word of an input object from an input sentence;

identifying industry information of the input object according to the core word;

and determining the search requirement of the input sentence according to at least one of the core word and the industry information, wherein the search requirement is one of a source searching class, a content class and other classes.

10. The method of claim 9, the generating a search results page according to the traffic type comprising:

and generating a search result page according to the service type and the search requirement of the input sentence.

11. A sample data generating apparatus comprising:

the core word determining module is used for identifying the core word of the target object from the search statement;

the industry information determining module is used for identifying the industry information of the target object according to the core word;

a service type determining module, configured to determine, according to at least one of the core word and the industry information, a service type of the search statement as a type tag of the search statement, where the type tag includes one of a first service type strongly associated with an enterprise-oriented service, a second service type cross-associated with an enterprise-oriented service and a personal service, and a third service type unrelated to the enterprise-oriented service; and

and the sample data generation module is used for determining the search statement with the type tag as sample data.

12. The apparatus of claim 11, wherein the traffic type determination module comprises:

The matching sub-module is used for matching the core word and the industry information with the category range of each business type; and

and the first determining submodule is used for determining the service type of the search statement according to the category range hit by at least one of the core word and the industry information.

13. The apparatus of claim 12, wherein the industry information comprises a first industry category strongly associated with enterprise-oriented business, a second industry category cross-associated with enterprise-oriented business and personal-oriented business, and a third industry category strongly associated with personal-oriented business; the apparatus further comprises:

the first dividing module is used for dividing the object entity belonging to the first business category into category ranges of the first business category;

the second dividing module is used for dividing the object entity under the second industry class into class ranges of the second service type;

and the third dividing module is used for dividing the object entity under the third industry category, the object entity irrelevant to all the industry categories and the object entity without transaction attribute under all the industry categories into the category range of the third business type.

14. The apparatus of claim 13, wherein the first industry category comprises a first special category range; the apparatus further comprises:

and the first screening module is used for responding that the target object belongs to the first special category range and determining that the service type of the search statement is one of a first service type and a third service type according to the core word of the target object.

15. The apparatus of claim 13 or 14, wherein the second industry category comprises a second special category; the apparatus further comprises:

the second screening module is used for responding that the target object belongs to the second special category range and matching the core word of the target object with the object entity in the second special category range; and dividing the service type of the search statement into one of a second service type and a third service type according to the matching result.

16. The apparatus of any of claims 11 to 15, wherein the traffic type determination module comprises:

a search requirement determination sub-module for identifying a search requirement of the search sentence according to at least one of the core word and the industry information, wherein the search requirement includes one of a sourcing class, a content class, and other classes; and

And the second determining submodule is used for determining the service type of the search statement as a third service type in response to the search requirement of the search statement as other types.

17. A training device for a deep learning model, comprising:

a sample acquisition module for acquiring sample data, wherein the sample data is generated according to the apparatus of any one of claims 11 to 16; and

and the training module is used for training the deep learning model by using the sample data to obtain the deep learning model for identifying the service type of the search statement.

18. A search result determination apparatus, comprising:

the input sentence acquisition module is used for acquiring an input sentence;

a processing module, configured to identify a service type of the input sentence using a trained deep learning model, where the service type is one of a first service type strongly associated with an enterprise-oriented service, a second service type cross-associated with an enterprise-oriented service and a personal-oriented service, and a third service type unrelated to the enterprise-oriented service; and

the search result determining module is used for generating a search result page according to the service type;

wherein the deep learning model is trained from the apparatus of claim 17.

19. The apparatus of claim 18, the apparatus further comprising:

the input object core word determining module is used for identifying the core word of the input object from the input sentence;

the input object industry information determining module is used for identifying industry information of the input object according to the core word;

and the search requirement determining module is used for determining the search requirement of the input sentence according to at least one of the core word and the industry information, wherein the search requirement is one of a source searching class, a content class and other classes.

20. The apparatus of claim 19, wherein the search result determination module is configured to generate a search result page according to a service type and a search requirement of the input sentence.

21. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 10.

22. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1 to 10.

23. A computer program product comprising a computer program stored on at least one of a readable storage medium and an electronic device, which, when executed by a processor, implements the method according to any one of claims 1 to 10.