CN112364235A

CN112364235A - Search processing method, model training method, device, medium and equipment

Info

Publication number: CN112364235A
Application number: CN202011303665.1A
Authority: CN
Inventors: 王鑫宇; 张永华
Original assignee: Beijing ByteDance Network Technology Co Ltd
Current assignee: Beijing ByteDance Network Technology Co Ltd
Priority date: 2020-11-19
Filing date: 2020-11-19
Publication date: 2021-02-12
Also published as: WO2022105775A1

Abstract

The present disclosure relates to a search processing method, a model training method, an apparatus, a medium, and a device, the search processing method including: receiving target search information; determining a target search result according to the target search information, and determining the target correlation between the target search result and the target search information through a correlation determination model; wherein, the correlation degree determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and the historical search information according to historical operation behavior information implemented by a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same; and taking the historical search information and the historical search result as the input of the model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain a correlation degree determination model. Therefore, the determined target correlation degree is more accurate.

Description

Search processing method, model training method, device, medium and equipment

Technical Field

The present disclosure relates to the field of search technologies, and in particular, to a search processing method, a model training method, an apparatus, a medium, and a device.

Background

In the search field, a search is generally performed according to search information, such as a search word or a search term, input by a user, to obtain a corresponding search result. The degree of correlation between the search result and the search information input by the user can reflect whether the search result meets the search intention of the user. Wherein, the higher the correlation degree between the search result and the search information is, the more the search result can be characterized to conform to the search intention of the user.

Currently, the degree of correlation between the search result and the search information is mainly determined by a model, and the training of the model depends on preset training data, which may include the degree of correlation between the search result and the search information that have been labeled. In the related art, technicians generally perform manual labeling on the correlation between search results and search information, however, the data size required for training is large, the efficiency is low through a manual labeling mode, and the manual labeling on the correlation is influenced by subjective judgment of the technicians and is not accurate enough.

Disclosure of Invention

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

In a first aspect, the present disclosure provides a search processing method, including: receiving target search information; determining a target search result according to the target search information, and determining a target correlation degree between the target search result and the target search information through a correlation degree determination model; wherein, the correlation determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

In a second aspect, the present disclosure provides a relevancy determination model training method, the method comprising: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein 7, the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

In a third aspect, the present disclosure provides a search processing apparatus, the apparatus comprising: the receiving module is used for receiving target searching information; the target relevancy determining module is used for determining a target searching result according to the target searching information and determining the target relevancy between the target searching result and the target searching information through a relevancy determining model; wherein, the correlation determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

In a fourth aspect, the present disclosure provides a relevancy determination model training apparatus, the apparatus including: the system comprises a relevancy determining module, a relevancy determining module and a relevancy determining module, wherein the relevancy determining module is used for respectively determining the relevancy between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, the historical search results with the same target text abstract information are the same as the relevancy between the historical search results and the historical search information, and the historical search results are obtained by searching according to the historical search information input by the user; and the training module is used for taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

In a fifth aspect, the present disclosure provides a computer readable medium having stored thereon a computer program which, when executed by a processing apparatus, performs the steps of the method provided by the first aspect of the present disclosure.

In a sixth aspect, the present disclosure provides a computer readable medium having stored thereon a computer program which, when executed by a processing apparatus, performs the steps of the method provided by the second aspect of the present disclosure.

In a seventh aspect, the present disclosure provides an electronic device, comprising: a storage device having a computer program stored thereon; processing means for executing the computer program in the storage means to implement the steps of the method provided by the first aspect of the present disclosure.

In an eighth aspect, the present disclosure provides an electronic device comprising: a storage device having a computer program stored thereon; processing means for executing the computer program in the storage means to implement the steps of the method provided by the second aspect of the present disclosure.

According to the technical scheme, the target search result is determined according to the target search information, and the target correlation degree between the target search result and the target search information is determined through the correlation degree determination model. In the training stage of the relevance determination model, under the condition that the historical search results conform to the search intention of the user, the user can operate the historical search results, so that the relevance between each historical search result and the historical search information is determined according to the historical operation behavior information of the user on the plurality of historical search results, and the relevance can be used as model training data to obtain the relevance determination model through training. Therefore, training data do not need to be labeled manually, data required by model training can be obtained quickly, and the problem that the correlation degree of manual labeling is inaccurate is avoided. Moreover, when the correlation between the historical search result and the historical search information is determined, the correlation between the historical search result with the same target text abstract information and the historical search information is the same, the influence of other factors such as the display sequence and the like on the operation behavior of the user can be avoided, the model training data and the trained correlation determination model are more accurate, the correlation between the target search result determined by the correlation determination model and the target search information is more accurate, and an accurate basis is provided for judging whether the target search result meets the search intention of the user.

Additional features and advantages of the disclosure will be set forth in the detailed description which follows.

Drawings

The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale. In the drawings:

FIG. 1 is a flow diagram illustrating a method of relevance determining model training in accordance with an exemplary embodiment.

FIG. 2 is a flow diagram illustrating a method of determining a degree of relevance between respective historical search results and historical search information, respectively, according to an example embodiment.

Fig. 3 is a flow diagram illustrating a method of determining target text summary information for historical search results, according to an example embodiment.

FIG. 4 is a flowchart illustrating a method of determining relevance between historical search results included in a search result set and historical search information, according to an example embodiment.

FIG. 5 is a flow diagram illustrating a search processing method in accordance with an exemplary embodiment.

Fig. 6 is a block diagram illustrating a search processing apparatus according to an example embodiment.

FIG. 7 is a block diagram illustrating a relevancy determination model training apparatus according to an exemplary embodiment.

Fig. 8 is a schematic structural diagram of an electronic device according to an exemplary embodiment.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.

It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order, and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.

The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description.

It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.

It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.

The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

First, a method for training a relevance determination model in an embodiment of the present disclosure is described, where the relevance determination model may be used to determine a relevance between a search result and search information.

Fig. 1 is a flowchart illustrating a relevancy determination model training method according to an exemplary embodiment, which may be applied to an electronic device with processing capability, and as shown in fig. 1, the method may include S101 and S102.

In S101, according to the history operation behavior information performed by the user on the plurality of history search results, the correlation between each history search result and the history search information is determined.

The historical search result can be obtained by searching according to the historical search information input by the user. Illustratively, the historical search information may include search information entered by the user over a historical period (e.g., the past week or month), which may be search terms or search sentences entered by the user. The plurality of historical search results can be searched according to the historical search information, for example, the historical search information input by the user is the name of the singer, and the plurality of historical search results can comprise a plurality of songs authored by the singer.

The user can perform operations on the historical search results according with the self search intention according to requirements, wherein the operations can comprise operations of clicking, browsing, sliding and the like, and in addition, when the user searches media files such as music, videos and the like, the operations can also comprise operations of playing the music or video files. The historical operational behavior information performed by the user on the historical search results may include one or more of the following items of information: user click behavior information, user browsing behavior information, user playing behavior information, and the like. The historical operation behavior information can be used for representing whether the user performs operation on the historical search result.

In one embodiment, the historical operation behavior information may include one of the above information items, for example, the user clicks on the historical search result, or the user browses the historical search result, and the user is considered to perform an operation on the historical search result. In another embodiment, the historical operation behavior information may include a plurality of information items, and it may be determined whether the user performed an operation on the historical search result through the plurality of information items. For example, the user clicks on the historical search result, and only after the page corresponding to the historical search result browses for a first preset time period, the user is considered to perform an operation on the historical search result. Or, the user clicks the historical search result, and plays the music or video file corresponding to the historical search result for a second preset time period, and then the user is considered to perform the operation on the historical search result. The first predetermined duration and the second predetermined duration may be pre-calibrated.

When the historical search result meets the search intention of the user, the user can operate the historical search result, so that the historical operation behavior information of the user on the historical search result can reflect the correlation degree between the historical search result and the historical search information to a certain extent. The historical search results which are operated more by the user are more consistent with the search intention of the user, namely, the correlation degree between the historical search information and the historical search information is relatively higher. According to the historical operation behavior information implemented by the user on the historical search results, the relevancy between each historical search result and the historical search information is respectively determined, and technicians do not need to manually mark the relevancy.

However, the operation behavior performed by the user on the historical search results is influenced by factors such as the display sequence of the historical search results, for example, the user is generally used to click the historical search results with the previous display sequence, which makes the determination of the correlation between the historical search results and the historical search information according to the historical operation behavior information easily influenced by other factors such as the display sequence, the playing amount of the media files, the attention degree, and the like.

Illustratively, taking a search user name as an example, for example, historical search information input by a user is aaa, and a plurality of historical search results are searched according to the historical search information, including: user name: aaa1, attention: 2000; user name aaa2, attention: 200 of a carrier; user name: aaa3, attention: 20. since the user name aaa1 is more focused and is shown in the first order, the click rate of the user with the user name aaa1 is higher than that of the other two users, the historical search result with the user name aaa1 is considered to have the highest correlation with the historical search information, the historical search result with the user name aaa2 has the second correlation with the historical search information, and the historical search result with the user name aaa3 has the lowest correlation with the historical search information. However, from the perspective of the text information, the three historical search results are matched with the historical search information aaa to the same extent.

By way of further example, taking the historical search information as search information for a media file as an example, the historical search information input by the user is word a, the word a is the name of a singer, the singer has a plurality of musical compositions, for example, music 1 and music 2, when the historical search result is displayed to the user, the display order of the music 1 may be the front, the display order of the music 2 is the back, and in general, the user play rate of the music 1 is high, and the user play rate of the music 2 is low. However, the creators of music 1 and music 2 both have the same subject, and the degree of matching between music 1 and word a and the degree of matching between music 2 and word a are the same.

In view of this, in order to avoid the influence of other factors such as the presentation order, the play amount, and the like on the user operation behavior, in the present disclosure, in determining the degree of correlation between the historical search result and the historical search information, the degree of correlation between the historical search result and the historical search information having the same target text summary information is the same. The target text summary information of the historical search result may be text information of a part of the historical search result, which has an association with the historical search information. If the target text abstract information is the same, the characteristic that the matching degree between the historical search result and the historical search information is the same from the perspective of the text can be represented.

In S102, the historical search information and the historical search result are used as inputs of the model, the correlation between the historical search result and the historical search information is used as a target output of the model, and the model is trained to obtain a correlation determination model. The relevance determination model may be any network model, such as a neural network model, and the form of the relevance determination model is not particularly limited by the present disclosure.

According to the scheme, under the condition that the historical search results conform to the search intention of the user, the user can operate the historical search results, therefore, according to the historical operation behavior information of the user on the plurality of historical search results, the correlation degree between each historical search result and the historical search information is determined, the correlation degree can be used as model training data, and a correlation degree determination model is obtained through training. Therefore, training data do not need to be labeled manually, data required by model training can be obtained quickly, and the problem that the correlation degree of manual labeling is inaccurate is avoided. Moreover, when the correlation degree between the historical search result and the historical search information is determined, the correlation degree between the historical search result with the same target text abstract information and the historical search information is the same, the influence of other factors such as the display sequence on the operation behavior of the user can be avoided, the model training data is more accurate, and the trained correlation degree determination model is more accurate.

Fig. 2 is a flowchart illustrating a method of determining a degree of correlation between each historical search result and historical search information, respectively, according to an exemplary embodiment, and as shown in fig. 2, S101 may include S201 to S203.

In S201, for each historical search result, target text summary information of the historical search result is determined.

In an embodiment, an exemplary implementation of step S201 may be as shown in fig. 3, including S2011 to S2014.

In S2011, text information belonging to a preset topic in the history search result is acquired. The preset theme can be a preset label used for describing the historical search result from different dimensions.

For example, in the case where the history search information is search information for a user name, the preset topic may include a user ID, a user name, a degree of attention, a user signature, and the like. As another example, in the case where the historical search information is search information for a document, the preset topic may include a title of the document, an author of the document, a content of the document, and the like. For example, in the case where the historical search information is search information for media files, the preset theme may include names of the media files, creators of the media files, lyrics of the media files, albums to which the media files belong, genres of the media files, and the like. It should be noted that, in the following description of the present disclosure, the history search information is taken as the search information for the media file as an example for illustration, but the present disclosure is not limited to the embodiment.

The text information belonging to the preset theme in the acquired historical search result may include all theme contents or partial theme contents under the preset theme, and under the condition that the text information includes partial theme contents under the preset theme, a plurality of text information belonging to the preset theme are provided.

Taking the historical search information as the search information for the media file as an example, for a preset topic, such as the name of the media file, the topic content under the preset topic is generally shorter, and is usually the song name of the song, so the text information belonging to the preset topic may include all the topic contents under the preset topic. In addition, the preset themes, such as the creator of the media file, the album to which the media file belongs, and the style of the media file, may be included in the acquired text information. For the preset theme of the lyrics of the media file, since the content of the lyrics is generally more, the text information belonging to the preset theme may include a part of the theme content under the preset theme, and the part of the theme content may be, for example, a sentence in the lyrics. In this way, the text information belonging to the preset theme is plural, and each sentence in the lyrics can be used as the text information belonging to the preset theme.

In S2012, candidate text digest information is determined. The candidate text summary information may include text information belonging to each preset topic, and text combination information of the text information.

When the user inputs the historical search information, the user may input contents of a plurality of preset topics, for example, the user searches in a manner of singer plus singer name, so the candidate text summary information may include text combination information of the text information, wherein the text information of different preset topics may be combined, and the situation that the user inputs the contents of the plurality of preset topics can be covered, so that the candidate text summary information is more comprehensive. The text combination mode is not particularly limited in the present disclosure, and for example, two text messages may be combined.

In S2013, the matching degree between each candidate text summary information and the history search information is determined respectively.

For example, the matching degree between the candidate text summary information and the historical search information can be determined by the number of matched characters between the two, wherein the matching can mean the same or consistent. The matching degree between the candidate text summary information and the historical search information can be determined by, for example, the following formula (1):

wherein, M represents the matching degree between the candidate text summary information and the historical search information, hit _ terms represents the number of characters matching between the candidate text summary information and the historical search information, query _ length represents the number of characters of the historical search information, and doc _ length represents the number of characters of the candidate text summary information.

In S2014, the candidate text summary information having the highest matching degree with the historical search information is determined as the target text summary information.

In S202, the historical search results with the same target text summary information are aggregated into a search result group.

After the target text abstract information of each historical search result is determined, the historical search results with the same target text abstract information can be determined. For example, if the target text summary information of the historical search result 1 is the same as that of the historical search result 2, the historical search result 1 and the historical search result 2 may be aggregated into a search result group. It is worth noting that the present disclosure is not particularly limited with respect to the number of historical search results included in the search result set, and the above examples are merely illustrative.

In S203, a degree of correlation between the historical search result included in the search result group and the historical search information is determined according to the historical operation behavior information performed by the user on the historical search result included in the search result group.

In one embodiment, the degree of correlation between the historical search results and the historical search information may be determined from one historical search activity performed by the user based on the historical search information. However, the correlation degree is determined only through one time of historical search behavior of the user, the reference data is less, effective data support cannot be provided, and the determined correlation degree may not be accurate enough. The present disclosure provides another preferred embodiment, which may include S2031 to S2033 as shown in fig. 4, where the correlation between the historical search result and the historical search information is determined by a plurality of historical search behaviors performed by the user according to the historical search information.

In S2031, for each of a plurality of historical search behaviors performed by the user based on the historical search information, the target historical operation behavior information performed by the user on the search result group in the historical search behavior is determined based on the historical operation behavior information performed by the user on the historical search results included in the search result group in the historical search behavior.

Wherein the multiple historical search activities may be initiated by different users. The target historical operational behavior information may be used to characterize whether the user performed an operation on the set of search results. For example, in one historical search behavior of the user, as long as the user performs an operation on any historical search result in the search result group, the target historical operation behavior information corresponding to the search result group may be marked as 1.

In S2032, the historical operation behavior feature information applied by the user to the search result group is determined according to the target historical operation behavior information applied by the user to the search result group in the plurality of times of historical search behaviors.

The historical operation behavior characteristic information can represent the historical click rate, the historical play rate, the historical browse rate and the like of the search result group. For example, data of X times of historical search behaviors performed by a user according to historical search information is counted, target historical operation behavior information is Y, and Y is smaller than or equal to X, then historical operation behavior feature information can be represented by a ratio of Y to X.

In S2033, a degree of correlation between the history search result included in the search result group and the history search information is determined based on the history operation behavior feature information.

For example, the historical operation behavior feature information may be directly used as the correlation between the historical search result included in the search result group and the historical search information. Alternatively, the degree of correlation may be determined according to a correspondence relationship between the historical operation behavior feature information quantified in advance and the degree of correlation.

In the scheme, the correlation degree between the historical search result and the historical search information can be accurately determined through multiple historical search behaviors performed by the user according to the historical search information. Moreover, historical search results with the same target text abstract information are aggregated into a search result group, the correlation degree between the historical search results and the historical search information in the search result group is the same, the influence of other factors such as the display sequence on the operation behavior of the user can be avoided, the model training data is more accurate, and the trained correlation degree determination model is more accurate.

The present disclosure also provides a search processing method, and fig. 5 is a flowchart illustrating a search processing method according to an exemplary embodiment, which may be applied to an electronic device having a processing capability, such as a terminal or a server, as shown in fig. 5, and which may include S501 and S502.

In S501, target search information is received.

The target search information may be information such as a search word or a search sentence input by a user, the user may be the same as or different from a user who inputs the historical search information, the target search information may be the same as or different from the historical search information, and the disclosure is not particularly limited.

In S502, a target search result is determined according to the target search information, and a target degree of correlation between the target search result and the target search information is determined by the degree of correlation determination model.

The target search result and the target search information can be input into a relevance determining model trained in advance, so that the target relevance between the target search result and the target search information output by the relevance determining model can be obtained.

The correlation determination model may be trained as follows: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of the model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain a correlation degree determination model.

The training process of the correlation determination model is described in detail above, and is not described herein again.

Optionally, the determining, according to historical operation behavior information that is implemented by a user on a plurality of historical search results, a degree of correlation between each historical search result and the historical search information includes: for each historical search result, determining the target text abstract information of the historical search result; aggregating the historical search results with the same target text abstract information into a search result group; and determining the correlation degree between the historical search results included in the search result group and the historical search information according to the historical operation behavior information implemented by the user on the historical search results included in the search result group.

Optionally, the determining, according to the historical operation behavior information performed by the user on the historical search results included in the search result group, a degree of correlation between the historical search results included in the search result group and the historical search information includes: for each historical search behavior in a plurality of historical search behaviors performed by a user according to the historical search information, determining target historical operation behavior information, which is implemented by the user on the search result group in the historical search behavior, according to the historical operation behavior information, which is implemented by the user on the historical search results included in the search result group in the historical search behavior; determining historical operation behavior characteristic information of the user on the search result group according to the target historical operation behavior information of the user on the search result group in multiple historical search behaviors; and determining the correlation degree between the historical search result and the historical search information in the search result group according to the historical operation behavior characteristic information.

Optionally, the determining the target text summary information of the historical search result includes: acquiring text information belonging to a preset theme in the historical search result, wherein the text information comprises all theme contents or partial theme contents under the preset theme, and under the condition that the text information comprises partial theme contents under the preset theme, a plurality of text information belonging to the preset theme are acquired; determining candidate text abstract information, wherein the candidate text abstract information comprises the text information belonging to each preset theme and text combination information of the text information; respectively determining the matching degree between each candidate text abstract information and the historical search information; and determining the candidate text abstract information with the highest matching degree with the historical search information as the target text abstract information.

The specific implementation of the above steps has been described in detail in the above relevancy determination model training method.

The search processing method provided by the present disclosure may further include: and determining the display sequence of the target search results according to the target correlation between the target search results and the target search information.

Generally, a plurality of target search results can be searched according to the target search information, and the correlation degree between the plurality of target search results and the target search information may be different. For example, the presentation order of the target search results having a high degree of correlation with the target search information may be ranked before the presentation order of the target search results having a low degree of correlation with the target search information. Therefore, the user can browse the search results which are more relevant to the target search information input by the user, and the user experience is improved.

Based on the same inventive concept, the present disclosure also provides a search processing apparatus, and fig. 6 is a block diagram illustrating a search processing apparatus according to an exemplary embodiment, where the apparatus 600 may include:

a receiving module 601, configured to receive target search information;

a target relevance determining module 602, configured to determine a target search result according to the target search information, and determine a target relevance between the target search result and the target search information through a relevance determining model;

wherein, the correlation determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Optionally, the apparatus 600 may further include: and the display sequence determining module is used for determining the display sequence of the target search result according to the target correlation degree between the target search result and the target search information.

With regard to the apparatus in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be described in detail here.

The present disclosure also provides a relevancy determination model training apparatus, and fig. 7 is a block diagram illustrating a relevancy determination model training apparatus according to an exemplary embodiment, where the apparatus 700 may include:

a relevance determining module 701, configured to determine, according to historical operation behavior information that is implemented by a user on multiple historical search results, relevance between each historical search result and historical search information, where the relevance between the historical search result having the same target text summary information and the historical search information is the same, and the historical search result is obtained by searching according to the historical search information input by the user;

a training module 702, configured to use the historical search information and the historical search result as inputs of a model, use a correlation between the historical search result and the historical search information as a target output of the model, and train the model to obtain the correlation determination model.

Optionally, the correlation determining module 701 includes: a target text summary information determining submodule, configured to determine, for each historical search result, the target text summary information of the historical search result; the aggregation sub-module is used for aggregating the historical search results with the same target text abstract information into a search result group; a first relevancy determination sub-module, configured to determine relevancy between the historical search result included in the search result group and the historical search information according to the historical operation behavior information that is implemented by the user on the historical search result included in the search result group.

Optionally, the first correlation determination sub-module includes: a behavior information determining submodule, configured to determine, for each historical search behavior of multiple historical search behaviors performed by the user according to the historical search information, target historical operation behavior information that the user performs on the search result group in the historical search behavior according to the historical operation behavior information that the user performs on the historical search result included in the search result group in the historical search behavior; the behavior characteristic information determining submodule is used for determining the historical operation behavior characteristic information which is implemented by the user on the search result group according to the target historical operation behavior information which is implemented by the user on the search result group in multiple historical search behaviors; and the second correlation degree determining submodule is used for determining the correlation degree between the historical search results and the historical search information included in the search result group according to the historical operation behavior characteristic information.

Optionally, the target text summary information determining sub-module includes: a text information obtaining sub-module, configured to obtain text information belonging to a preset topic in the historical search result, where the text information includes all topic content or partial topic content under the preset topic, and under a condition that the text information includes partial topic content under the preset topic, there are multiple text information belonging to the preset topic; the candidate text abstract information determining submodule is used for determining candidate text abstract information, wherein the candidate text abstract information comprises the text information belonging to each preset theme and text combination information of the text information; the matching degree determining submodule is used for respectively determining the matching degree between each candidate text abstract information and the historical search information; and the abstract information determining submodule is used for determining the candidate text abstract information with the highest matching degree with the historical search information as the target text abstract information.

Referring now to FIG. 8, shown is a schematic diagram of an electronic device 800 suitable for use in implementing embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 8 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. 8, an electronic device 800 may include a processing means (e.g., central processing unit, graphics processor, etc.) 801 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)802 or a program loaded from a storage means 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data necessary for the operation of the electronic apparatus 800 are also stored. The processing apparatus 801, the ROM 802, and the RAM 803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

Generally, the following devices may be connected to the I/O interface 805: input devices 806 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 807 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; storage 808 including, for example, magnetic tape, hard disk, etc.; and a communication device 809. The communication means 809 may allow the electronic device 800 to communicate wirelessly or by wire with other devices to exchange data. While fig. 8 illustrates an electronic device 800 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication means 809, or installed from the storage means 808, or installed from the ROM 802. The computer program, when executed by the processing apparatus 801, performs the above-described functions defined in the methods of the embodiments of the present disclosure.

It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having 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. 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. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.

The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.

The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: receiving target search information; determining a target search result according to the target search information, and determining a target correlation degree between the target search result and the target search information through a correlation degree determination model; wherein, the correlation determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Alternatively, the computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Computer program code for carrying out operations for the present disclosure may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

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 and/or flowchart illustration, and combinations of blocks in the block diagrams and/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.

The modules described in the embodiments of the present disclosure may be implemented by software or hardware. The name of a module does not in some cases constitute a limitation of the module itself, and for example, a receiving module may also be described as a "target search information receiving module".

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.

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. A 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.

Example 1 provides a search processing method according to one or more embodiments of the present disclosure, the method including: receiving target search information; determining a target search result according to the target search information, and determining a target correlation degree between the target search result and the target search information through a correlation degree determination model; wherein, the correlation determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Example 2 provides the method of example 1, wherein determining the correlation between each historical search result and the historical search information according to the historical operation behavior information implemented by the user on the plurality of historical search results respectively comprises: for each historical search result, determining the target text abstract information of the historical search result; aggregating the historical search results with the same target text abstract information into a search result group; and determining the correlation degree between the historical search results included in the search result group and the historical search information according to the historical operation behavior information implemented by the user on the historical search results included in the search result group.

Example 3 provides the method of example 2, the determining a degree of correlation between the historical search results included in the search result group and the historical search information according to the historical operational behavior information implemented by the user on the historical search results included in the search result group, including: for each historical search behavior in a plurality of historical search behaviors performed by a user according to the historical search information, determining target historical operation behavior information, which is implemented by the user on the search result group in the historical search behavior, according to the historical operation behavior information, which is implemented by the user on the historical search results included in the search result group in the historical search behavior; determining historical operation behavior characteristic information of the user on the search result group according to the target historical operation behavior information of the user on the search result group in multiple historical search behaviors; and determining the correlation degree between the historical search result and the historical search information in the search result group according to the historical operation behavior characteristic information.

Example 4 provides the method of example 2, the determining the target text summary information of the historical search results, including, in accordance with one or more embodiments of the present disclosure: acquiring text information belonging to a preset theme in the historical search result, wherein the text information comprises all theme contents or partial theme contents under the preset theme, and under the condition that the text information comprises partial theme contents under the preset theme, a plurality of text information belonging to the preset theme are acquired; determining candidate text abstract information, wherein the candidate text abstract information comprises the text information belonging to each preset theme and text combination information of the text information; respectively determining the matching degree between each candidate text abstract information and the historical search information; and determining the candidate text abstract information with the highest matching degree with the historical search information as the target text abstract information.

Example 5 provides the method of example 4, wherein the historical search information is search information for media files, and accordingly the preset topic includes names of the media files, creators of the media files, lyrics of the media files, albums to which the media files belong, and genres of the media files.

Example 6 provides the method of any one of examples 1 to 5, further comprising, in accordance with one or more embodiments of the present disclosure: and determining the display sequence of the target search results according to the target correlation between the target search results and the target search information.

Example 7 provides a relevance determination model training method, according to one or more embodiments of the present disclosure, the method including: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Example 8 provides the method of example 7, in accordance with historical operational behavior information performed by a user on a plurality of historical search results, the determining a degree of correlation between each historical search result and the historical search information, respectively, including: for each historical search result, determining the target text abstract information of the historical search result; aggregating the historical search results with the same target text abstract information into a search result group; and determining the correlation degree between the historical search results included in the search result group and the historical search information according to the historical operation behavior information implemented by the user on the historical search results included in the search result group.

Example 9 provides the method of example 8, the determining a degree of correlation between the historical search results included in the search result group and the historical search information according to the historical operational behavior information implemented by the user on the historical search results included in the search result group, including: for each historical search behavior in a plurality of historical search behaviors performed by a user according to the historical search information, determining target historical operation behavior information, which is implemented by the user on the search result group in the historical search behavior, according to the historical operation behavior information, which is implemented by the user on the historical search results included in the search result group in the historical search behavior; determining historical operation behavior characteristic information of the user on the search result group according to the target historical operation behavior information of the user on the search result group in multiple historical search behaviors; and determining the correlation degree between the historical search result and the historical search information in the search result group according to the historical operation behavior characteristic information.

Example 10 provides the method of example 8, the determining the target text summary information of the historical search results, including, in accordance with one or more embodiments of the present disclosure: acquiring text information belonging to a preset theme in the historical search result, wherein the text information comprises all theme contents or partial theme contents under the preset theme, and under the condition that the text information comprises partial theme contents under the preset theme, a plurality of text information belonging to the preset theme are acquired; determining candidate text abstract information, wherein the candidate text abstract information comprises the text information belonging to each preset theme and text combination information of the text information; respectively determining the matching degree between each candidate text abstract information and the historical search information; and determining the candidate text abstract information with the highest matching degree with the historical search information as the target text abstract information.

Example 11 provides the method of example 10, the historical search information is search information for media files, and accordingly, the preset topic includes names of the media files, creators of the media files, lyrics of the media files, albums to which the media files belong, and genres of the media files.

Example 12 provides a search processing apparatus according to one or more embodiments of the present disclosure, the apparatus comprising: the receiving module is used for receiving target searching information; the target relevancy determining module is used for determining a target searching result according to the target searching information and determining the target relevancy between the target searching result and the target searching information through a relevancy determining model; wherein, the correlation determination model is obtained by training in the following way: respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user; and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Example 13 provides a correlation determination model training apparatus according to one or more embodiments of the present disclosure, the apparatus including: the system comprises a relevancy determining module, a relevancy determining module and a relevancy determining module, wherein the relevancy determining module is used for respectively determining the relevancy between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, the historical search results with the same target text abstract information are the same as the relevancy between the historical search results and the historical search information, and the historical search results are obtained by searching according to the historical search information input by the user; and the training module is used for taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

Example 14 provides a computer readable medium having stored thereon a computer program that, when executed by a processing apparatus, performs the steps of the method of any one of examples 1-6, in accordance with one or more embodiments of the present disclosure.

Example 15 provides a computer readable medium having stored thereon a computer program that, when executed by a processing apparatus, implements the steps of the method of any one of examples 7-11, in accordance with one or more embodiments of the present disclosure.

Example 16 provides, in accordance with one or more embodiments of the present disclosure, an electronic device, comprising: a storage device having a computer program stored thereon; processing means for executing the computer program in the storage means to carry out the steps of the method of any one of examples 1-6.

Example 17 provides, in accordance with one or more embodiments of the present disclosure, an electronic device, comprising: a storage device having a computer program stored thereon; processing means for executing the computer program in the storage means to carry out the steps of the method of any one of examples 7-11.

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.

Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims. With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. A method of search processing, the method comprising:

receiving target search information;

determining a target search result according to the target search information, and determining a target correlation degree between the target search result and the target search information through a correlation degree determination model;

wherein, the correlation determination model is obtained by training in the following way:

respectively determining the correlation degree between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, wherein the correlation degrees between the historical search results with the same target text abstract information and the historical search information are the same, and the historical search results are obtained by searching according to the historical search information input by the user;

and taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

2. The method of claim 1, wherein the determining the correlation between each historical search result and the historical search information according to the historical operation behavior information of the plurality of historical search results implemented by the user comprises:

for each historical search result, determining the target text abstract information of the historical search result;

aggregating the historical search results with the same target text abstract information into a search result group;

and determining the correlation degree between the historical search results included in the search result group and the historical search information according to the historical operation behavior information implemented by the user on the historical search results included in the search result group.

3. The method of claim 2, wherein determining the degree of correlation between the historical search results included in the search result set and the historical search information according to the historical operational behavior information performed by the user on the historical search results included in the search result set comprises:

for each historical search behavior in a plurality of historical search behaviors performed by a user according to the historical search information, determining target historical operation behavior information, which is implemented by the user on the search result group in the historical search behavior, according to the historical operation behavior information, which is implemented by the user on the historical search results included in the search result group in the historical search behavior;

determining historical operation behavior characteristic information of the user on the search result group according to the target historical operation behavior information of the user on the search result group in multiple historical search behaviors;

and determining the correlation degree between the historical search result and the historical search information in the search result group according to the historical operation behavior characteristic information.

4. The method of claim 2, wherein the determining the target text summary information of the historical search results comprises:

acquiring text information belonging to a preset theme in the historical search result, wherein the text information comprises all theme contents or partial theme contents under the preset theme, and under the condition that the text information comprises partial theme contents under the preset theme, a plurality of text information belonging to the preset theme are acquired;

determining candidate text abstract information, wherein the candidate text abstract information comprises the text information belonging to each preset theme and text combination information of the text information;

respectively determining the matching degree between each candidate text abstract information and the historical search information;

and determining the candidate text abstract information with the highest matching degree with the historical search information as the target text abstract information.

5. The method of claim 4, wherein the historical search information is search information for media files, and accordingly, the preset topic comprises names of the media files, creators of the media files, lyrics of the media files, albums to which the media files belong, and styles of the media files.

6. The method according to any one of claims 1-5, further comprising:

and determining the display sequence of the target search results according to the target correlation between the target search results and the target search information.

7. A method for training a relevance-determining model, the method comprising:

8. The method of claim 7, wherein the determining the correlation between each historical search result and the historical search information according to the historical operation behavior information performed by the user on the plurality of historical search results comprises:

9. The method of claim 8, wherein determining the degree of correlation between the historical search results included in the search result set and the historical search information according to the historical operational behavior information performed by the user on the historical search results included in the search result set comprises:

10. The method of claim 8, wherein the determining the target text summary information of the historical search results comprises:

11. The method of claim 10, wherein the historical search information is search information for media files, and accordingly, the preset topic comprises names of the media files, creators of the media files, lyrics of the media files, albums to which the media files belong, and styles of the media files.

12. A search processing apparatus, characterized in that the apparatus comprises:

the receiving module is used for receiving target searching information;

the target relevancy determining module is used for determining a target searching result according to the target searching information and determining the target relevancy between the target searching result and the target searching information through a relevancy determining model;

13. A correlation determination model training apparatus, characterized in that the apparatus comprises:

the system comprises a relevancy determining module, a relevancy determining module and a relevancy determining module, wherein the relevancy determining module is used for respectively determining the relevancy between each historical search result and historical search information according to historical operation behavior information of a user on a plurality of historical search results, the historical search results with the same target text abstract information are the same as the relevancy between the historical search results and the historical search information, and the historical search results are obtained by searching according to the historical search information input by the user;

and the training module is used for taking the historical search information and the historical search result as the input of a model, taking the correlation degree between the historical search result and the historical search information as the target output of the model, and training the model to obtain the correlation degree determination model.

14. A computer-readable medium, on which a computer program is stored, characterized in that the program, when being executed by processing means, carries out the steps of the method of any one of claims 1 to 6.

15. A computer-readable medium, on which a computer program is stored, which, when being executed by processing means, carries out the steps of the method according to any one of claims 7 to 11.

16. An electronic device, comprising:

a storage device having a computer program stored thereon;

processing means for executing the computer program in the storage means to carry out the steps of the method according to any one of claims 1 to 6.

17. An electronic device, comprising:

a storage device having a computer program stored thereon;

processing means for executing the computer program in the storage means to carry out the steps of the method according to any one of claims 7 to 11.