Information pushing method and device, computer equipment and storage medium
Technical Field
The present disclosure relates to the field of computer processing technologies, and in particular, to a method and an apparatus for pushing information, a computer device, and a storage medium.
Background
At present, in order to improve the search efficiency of a user, before the user inputs a keyword into a search box, a search word may be automatically recommended to the user in a search area, and if the recommended search word can arouse the interest of the user, the user may directly click the search word to obtain a corresponding search result.
Therefore, the existing search term pushing method only pushes the search terms, and the display effect is single.
Disclosure of Invention
The embodiment of the disclosure provides at least one scheme for information pushing, and the information pushing is realized in a combined display mode of texts and pictures, so that the display effect is improved, the display effect is more visual, and the selection by a user is facilitated.
Mainly comprises the following aspects:
in a first aspect, an embodiment of the present disclosure provides an information pushing method, where the method includes:
determining a target recommended word; acquiring N candidate media materials corresponding to the target recommended word, and extracting M candidate recommended pictures from the N candidate media materials; wherein M is greater than or equal to N;
determining a target recommended picture from the candidate recommended pictures based on the similarity between the candidate recommended pictures;
and associating the target recommended word with the target recommended picture, and sending the target recommended word to a terminal.
In an optional embodiment, the determining a target recommended picture from the candidate recommended pictures based on the similarity between the candidate recommended pictures includes:
grouping the M candidate recommended pictures based on the similarity among the candidate recommended pictures to obtain at least one candidate recommended picture set;
sorting the at least one candidate recommended picture set according to the number of candidate recommended pictures included in the candidate recommended picture set;
determining a first candidate recommended picture set with the highest ranking as a target recommended picture set;
and selecting the target recommended picture from the target recommended picture set.
In an optional implementation manner, if the number of the first candidate recommended picture set with the highest ranking is greater than or equal to 2, the determining, as the target recommended picture set, the first candidate recommended picture set with the highest ranking includes:
respectively obtaining candidate media materials from which each candidate recommended picture in the first candidate recommended picture set comes, and determining at least one candidate media material corresponding to each first candidate recommended picture set;
and determining the target recommended picture set from the first candidate recommended picture set based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word.
In an optional implementation manner, the determining the target recommended picture set from the first candidate recommended picture sets based on the correlation between the target recommended word and at least one candidate media material corresponding to each of the first candidate recommended picture sets includes:
for each first candidate recommended picture set, determining the number of the candidate recommended pictures in the first candidate recommended set included in the at least one candidate media material corresponding to the first candidate recommended picture set, and respectively generating the weight of each at least one candidate media material;
and determining the target recommended picture set from the first candidate recommended picture set based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word and the weight of each at least one candidate media material.
In an optional implementation manner, the grouping M candidate recommended pictures based on the similarity between the candidate recommended pictures to obtain at least one candidate recommended picture set includes:
aiming at any candidate recommended picture in the M candidate recommended pictures, forming a candidate recommended picture set by the candidate recommended picture and other candidate recommended pictures with the similarity meeting preset conditions;
wherein the preset condition comprises at least one of the following conditions:
the similarity is greater than a first threshold;
and the similarity between other different candidate recommended pictures except the candidate recommended picture in the formed candidate recommended picture set and the candidate recommended picture is the same.
In an optional embodiment, the selecting the target recommended picture from the target recommended picture set includes:
determining the similarity between any candidate recommended picture in the target recommended picture set and other candidate recommended pictures to obtain a similarity set corresponding to the target recommended picture set;
if the number of the similarities in the similarity set is 1, or the number of the similarities in the similarity set is greater than or equal to 2, and the absolute value of the difference between the similarities is less than or equal to a second threshold, determining any candidate recommended picture in the target recommended picture set as the target recommended picture;
if the number of the similarities in the similarity set is greater than or equal to 2 and the absolute value of the difference between the similarities is greater than the second threshold, attributing the candidate recommended pictures with the same similarity in the target recommended picture set to the same group to obtain at least one target recommended picture subset; sorting the at least one target recommended picture subset according to the number of candidate recommended pictures included in the target recommended picture subset; and determining any candidate recommended picture in the highest-ranked target recommended picture subset as the target recommended picture.
In an optional embodiment, if the target recommended picture subset is multiple; determining any candidate recommended picture in the highest-ranked subset of target recommended pictures as the target recommended picture, including:
aiming at any target recommended picture subset in each target recommended picture subset, determining the ranking result of candidate media materials to which any candidate recommended picture belongs in the target recommended picture subset in N candidate media materials; determining ranking results of the target recommended picture subsets in the target recommended picture subsets according to ranking results of the candidate media materials to which the candidate recommended pictures belong in the N candidate media materials;
and determining any target recommended picture in the target recommended picture subset with the highest ranking as the target recommended picture.
In an alternative embodiment, the similarity between the candidate recommended pictures is determined according to the following steps:
extracting a picture fingerprint feature from one candidate recommended picture in the two candidate recommended pictures;
and determining the similarity between the two candidate recommended pictures based on the similarity between the extracted picture fingerprint features.
In an optional implementation manner, the obtaining N candidate media materials corresponding to the target recommended word includes:
acquiring corresponding media materials according to the target recommended words;
ranking the media materials according to the sequence of the relevance from big to small;
and determining the media materials of the first N ranks as N candidate media materials corresponding to the target recommended word.
In an alternative embodiment, the candidate media materials include candidate videos, and the extracting M candidate recommended pictures from the N candidate media materials includes:
determining all candidate pictures included in the N candidate videos as M candidate recommended pictures; alternatively, the first and second electrodes may be,
intercepting any candidate video in the N candidate videos according to a preset intercepting duration and/or a preset intercepting position to obtain an intercepted candidate video, and determining all candidate pictures included in the intercepted N candidate videos as M candidate recommended pictures; alternatively, the first and second electrodes may be,
extracting key frame candidate pictures from all candidate pictures included in any one of the N candidate videos, and determining the key frame candidate pictures in the extracted N candidate videos as M candidate recommended pictures.
In an alternative embodiment, the candidate media material comprises web page content; determining N candidate media materials corresponding to the target recommended word according to the following steps:
capturing N webpage contents corresponding to the target recommended words; the N webpage contents carry M candidate recommended pictures;
extracting M candidate recommended pictures from the N candidate media materials according to the following steps:
and extracting M candidate recommended pictures from the N webpage contents.
In a second aspect, the present disclosure further provides a method for pushing information, where the method includes:
receiving a target recommended word and a target recommended picture; the target recommended picture is determined from each candidate recommended picture based on the similarity among M candidate recommended pictures, wherein the M candidate recommended pictures are extracted from N candidate media materials, and M is greater than or equal to N;
and performing associated display on the target recommended word and the target recommended picture.
In a third aspect, the present disclosure further provides an information pushing apparatus, where the apparatus includes:
the acquisition module is used for determining a target recommended word; acquiring N candidate media materials corresponding to the target recommended word, and extracting M candidate recommended pictures from the N candidate media materials; wherein M is greater than or equal to N;
the determining module is used for determining a target recommended picture from the candidate recommended pictures based on the similarity between the candidate recommended pictures;
and the pushing module is used for associating the target recommended word with the target recommended picture and sending the associated target recommended word to a terminal.
In a fourth aspect, the present disclosure further provides an information pushing apparatus, where the apparatus includes:
the receiving module is used for receiving the target recommended words and the target recommended pictures; the target recommended picture is determined from each candidate recommended picture based on the similarity among M candidate recommended pictures, wherein the M candidate recommended pictures are extracted from N candidate media materials, and M is greater than or equal to N;
and the display module is used for carrying out associated display on the target recommended words and the target recommended pictures.
In a fifth aspect, the present disclosure also provides a computer device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, the processor and the storage medium communicate with each other through the bus when a computer device runs, and the machine-readable instructions, when executed by the processor, perform the steps of the method for pushing information according to the first aspect and any one of the various embodiments thereof.
In a sixth aspect, the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the information pushing method according to the first aspect and any one of the various implementation manners thereof.
By adopting the scheme, the server firstly determines the target recommended word, then obtains N candidate media materials corresponding to the target recommended word, and can determine a target recommended picture from the M candidate recommended pictures based on the similarity between the candidate recommended pictures, and finally can associate the target recommended word with the target recommended picture and send the target recommended word to a terminal for display, therefore, the terminal can display the target recommendation words and the target recommendation pictures at the same time, relatively only recommend the search words, the proposal improves the display effect by utilizing the target recommended picture, in addition, because the target recommended picture is extracted from the candidate media materials corresponding to the target recommended word, therefore, the target recommendation picture and the target recommendation word have larger relevance, so that the display effect is further improved, the display effect is more visualized, and the user can conveniently select the target recommendation picture and the target recommendation word.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly described below, and the drawings herein incorporated in and forming a part of the specification illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the technical solutions of the present disclosure. It is appreciated that the following drawings depict only certain embodiments of the disclosure and are therefore not to be considered limiting of its scope, for those skilled in the art will be able to derive additional related drawings therefrom without the benefit of the inventive faculty.
Fig. 1 shows a flowchart of a method for pushing information according to a first embodiment of the present disclosure;
2(a) -2 (b) show a specific application diagram of the method for pushing information provided by the first embodiment of the present disclosure;
fig. 3 is a flowchart illustrating a specific method for selecting a target recommended picture in a method for pushing information according to a first embodiment of the present disclosure;
fig. 4 is a flowchart illustrating a specific method for determining a target recommended picture set in a method for pushing information according to a first embodiment of the present disclosure;
fig. 5 is a flowchart illustrating a specific method for determining a target recommended picture in a method for pushing information according to a first embodiment of the present disclosure;
fig. 6 is a flowchart illustrating a specific method for determining target candidate media materials in a method for pushing information according to a first embodiment of the present disclosure;
fig. 7 shows a flowchart of a method for pushing information provided by the second embodiment of the present disclosure;
fig. 8 is a schematic diagram illustrating an apparatus for pushing information according to a third embodiment of the present disclosure;
fig. 9 is a schematic diagram of another information pushing apparatus provided in the third embodiment of the present disclosure;
FIG. 10 is a schematic diagram of a computer device provided in the fourth embodiment of the present disclosure;
fig. 11 shows a schematic diagram of another computer device provided in the fourth embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. The components of the embodiments of the present disclosure, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure, presented in the figures, is not intended to limit the scope of the claimed disclosure, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.
Research shows that, in the related art, before a user inputs a keyword into a search box, the search word can be automatically recommended to the user in a search area, so that the user can obtain a corresponding search result by directly clicking the search word.
Based on the research, the information pushing scheme is provided, the information pushing is achieved through a combined display mode of texts and pictures, the display effect is improved, the display effect is more visual, the user selection is facilitated, and the search requirement of the user can be met.
The above-mentioned drawbacks are the results of the inventor after practical and careful study, and therefore, the discovery process of the above-mentioned problems and the solutions proposed by the present disclosure to the above-mentioned problems should be the contribution of the inventor in the process of the present disclosure.
The technical solutions in the present disclosure will be described clearly and completely with reference to the accompanying drawings in the present disclosure, and it is to be understood that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The components of the present disclosure, as generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure, presented in the figures, is not intended to limit the scope of the claimed disclosure, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
To facilitate understanding of the present embodiment, first, a detailed description is given of an information pushing method disclosed in an embodiment of the present disclosure, where an execution subject of the information pushing method provided in the embodiment of the present disclosure is generally a computer device with certain computing capability, and the computer device includes, for example: a server or other processing device. In some possible implementations, the information pushing method may be implemented by a processor calling a computer readable instruction stored in a storage medium.
The following describes a method for pushing information provided by the embodiments of the present disclosure by taking an execution subject as a server.
Example one
Referring to fig. 1, which is a flowchart of a method for pushing information according to a first embodiment of the present disclosure, the method includes steps S101 to S103, where:
s101, determining a target recommended word; acquiring N candidate media materials corresponding to the target recommended word, and extracting M candidate recommended pictures from the N candidate media materials; wherein M is greater than or equal to N;
s102, determining a target recommended picture from the candidate recommended pictures based on the similarity between the candidate recommended pictures;
and S103, associating the target recommended word with the target recommended picture, and sending the target recommended word to the terminal.
Here, in the information submission method provided by the embodiment of the present disclosure, after determining a target recommended word, a server may determine a candidate media material corresponding to the target recommended word, extract candidate recommended pictures from the candidate media material, then determine the target recommended picture based on a similarity between the candidate recommended pictures, and finally send the target recommended word and a related target recommended picture to a terminal, so that the terminal may simultaneously display the target recommended word and the target recommended picture, and the overall display effect is improved by using a visual response effect of the target recommended picture.
In order to facilitate the realization of different search requirements of different users, the target recommended word can be generated based on the search popularity information so that the user can know information with higher popularity at present, and can also be generated based on the historical search record information of the terminal so that the personalized requirements of different users can be met.
In the embodiment of the present disclosure, only the target recommended word related to the search popularity information may be determined for the terminal, only the target recommended word for the search record of the terminal may be determined for the terminal, and the terminal may also be determined to include the above two types of target recommended words, which is not limited in this embodiment of the present disclosure. In specific application, the determination mode of which target recommended word is adopted can be determined by acquiring a search mode preset by a terminal.
The target recommended word may be one or more than one. In a specific Application, the server in the embodiment of the present disclosure may determine the target recommended word automatically determined after a search Application (APP) set on the terminal is opened, or may determine the target recommended word after receiving a target recommended word presentation instruction sent by the terminal.
The target recommended word display instruction may be initiated by the terminal to the server after responding to the target triggering operation, where the target triggering operation may be performed by using a search button set on a search display interface presented by the terminal by the user, that is, after the search button is triggered by the user, the target recommended word display instruction may be sent to the server by using a communication channel between the terminal and the server. The trigger operation may be a single click, a double click, a long press, and the like, which is not specifically limited in this disclosure.
In order to improve timeliness of information search, the target recommended word may be updated as the search time goes by. For example, it is determined that the user's history search record information is related to driving of a car within the history time, and in this case, the target recommended word thereof may be a text including driving of a car. With the forward progress of the search time, if the automobile beauty with higher heat appears, after the server pushes the automobile beauty to the terminal, the interest of the user may be aroused, so that the search of the user for the automobile beauty is promoted, and at this time, after the user triggers the target recommended word display instruction again, the text of the automobile beauty can be determined as the target recommended word of the user.
In the embodiment of the disclosure, while the target recommended word is determined for the terminal, a target recommended picture associated with the target recommended word may also be determined for the terminal, and the target recommended picture may be determined from each candidate recommended picture based on the similarity between M candidate recommended pictures.
The M candidate recommended pictures may be extracted from N candidate media materials, where M is greater than or equal to N, that is, at least one candidate recommended picture may be extracted from any one of the candidate media materials in the embodiment of the present disclosure. The candidate media materials may be video materials, web page materials, or other multimedia materials containing pictures, which is not limited in this disclosure. In addition, because the candidate media materials correspond to the target recommended words, the target recommended pictures extracted from the candidate media materials have a relatively large relevance with the target recommended words, and therefore the display effect is further improved.
Based on the determined target recommended word and the target recommended picture, the information pushing method provided by the embodiment of the disclosure can associate the target recommended word with the target recommended picture and then send the target recommended word to the terminal. Therefore, the terminal can simultaneously display each target recommendation word and the corresponding target recommendation picture, at this time, if it is determined that the user performs the target selection operation aiming at one target recommendation word in each target recommendation word, the terminal can initiate a search request to the server, then can receive the media material fed back by the server based on the search request, and can display the media material. The content of the media material is similar to that of the candidate media material, and is not described herein again.
In order to facilitate understanding of the information pushing method provided by the embodiment of the present disclosure, an example is described below with reference to fig. 2(a) to 2 (b).
As shown in fig. 2(a), the first search display interface is presented by the terminal after the server determines and sends the target recommendation pictures corresponding to the 3 target recommendation words to the terminal. On the first search display interface, the macro-detective pick-up is a target recommendation word, and the picture arranged on the left side of the macro-detective pick-up is the target recommendation picture. After the user selects the target recommendation word of the large detective picnic chum, the server can feed back the media material based on the search request of the terminal, and can display the media material, as shown in fig. 2 (b).
It should be noted that the setting position of the search button and the relative position of the target recommendation picture to the target recommendation word are only a specific example, and in an actual application, the setting may be performed based on different application requirements, and the setting is not limited specifically here.
Considering that determining a target recommended picture based on the similarity between candidate recommended pictures is a key step for implementing information push in the embodiments of the present disclosure, a detailed description may be given below with reference to fig. 3.
As shown in fig. 3, the step of determining the target recommended picture specifically includes:
s301, grouping the M candidate recommended pictures based on the similarity among the candidate recommended pictures to obtain at least one candidate recommended picture set;
s302, sequencing at least one candidate recommended picture set according to the number of the candidate recommended pictures included in the candidate recommended picture set;
s303, determining a first candidate recommended picture set with the highest ranking as a target recommended picture set;
s304, selecting the target recommended pictures from the target recommended picture set.
Here, in the embodiment of the present disclosure, first, M candidate recommended pictures may be grouped based on the similarity between the candidate recommended pictures to obtain a plurality of candidate recommended picture sets, and then, a first candidate recommended picture set with the largest number of candidate recommended pictures is selected from each candidate recommended picture set as a target recommended picture set to implement selection of the target recommended picture. The number of the same or similar images can be maximized by the selected target recommended picture set, so that the association degree between the selected target recommended pictures and the target recommended words can be further ensured, and the display effect is improved.
In the embodiment of the present disclosure, the M candidate recommended pictures may be grouped based on the similarity between the candidate recommended pictures, that is, for each candidate recommended picture, a candidate recommended picture set may be determined from the candidate recommended picture and other candidate recommended pictures whose similarities with the candidate recommended picture satisfy the preset condition.
The preset condition may be that the similarity is greater than a first threshold, that is, other candidate recommended pictures with greater similarity to the candidate recommended pictures to be grouped may be divided into a group with the candidate recommended pictures to be grouped; the preset condition may also be that the similarity between the other different candidate recommended pictures except the candidate recommended picture in the composed candidate recommended picture set and the candidate recommended picture is the same, that is, the other candidate recommended pictures with the similarity to the candidate recommended picture to be grouped and the candidate recommended picture to be grouped may be divided into a group.
After the candidate recommended pictures are grouped and at least one candidate recommended picture set is obtained, the at least one candidate recommended picture set may be ranked based on the number of candidate recommended pictures included in each candidate recommended picture set, and a first candidate recommended picture set with the highest ranking may be determined as a target recommended picture set.
There may be one or more first candidate recommended picture sets with the highest ranking. When the first candidate recommended picture set is multiple, the target recommended picture set may be determined from the multiple first candidate recommended picture sets based on the correlation between the target recommended word and at least one candidate media material corresponding to the first candidate recommended picture set.
In the embodiment of the present disclosure, regarding the correlation between each candidate media material corresponding to the first candidate recommended picture set and the target recommended word, the correlation between the candidate media material and the target recommended word may be determined based on the feature information corresponding to the candidate media material.
For example, when candidate videos are used as candidate media materials, the correlation between each candidate video and a target recommended word may be determined based on the correlation between the target recommended word and characteristic information such as summary information and video content of the candidate videos, and when the target recommended word is included in the summary information of one candidate video, it may be determined that the correlation between the candidate video and the target recommended word is relatively large, and otherwise, the correlation is relatively small. For another example, when the web page content is used as a candidate media material, the correlation between each web page content and the target recommended word may be determined by searching an inverted index established by the web page content, and the like.
To facilitate understanding of the above process of determining a target recommended picture set from a plurality of first candidate recommended picture sets based on the correlation between a candidate media material and a target recommended word, a description will be given with reference to a specific example, with a candidate video as a candidate media material.
For example, there are 10 candidate videos and two first candidate recommended picture sets with the highest ranking, each candidate recommended picture in one first candidate recommended picture set is from candidate video a, candidate video B, and candidate video C, and the ranking results respectively correspond to 1, 2, and 3, each candidate recommended picture in the other first candidate recommended picture set is from candidate video C, candidate video D, and candidate video F, and the ranking results respectively correspond to 3, 4, and 5, since each candidate video in the first candidate recommended picture set may be in sequence, the higher the ranking is, the greater the corresponding correlation is, and therefore, the first candidate recommended picture set with the ranking results of 1, 2, and 3 may be determined as the target recommended picture set.
Considering that different first candidate recommended picture sets may be from different candidate videos or from the same candidate video, when different first candidate recommended picture sets are from the same candidate video, the target recommended picture set may be determined by combining the weight (corresponding to the number of candidate recommended pictures appearing) of at least one candidate media material corresponding to each first candidate recommended picture set.
As shown in fig. 4, the target recommended picture set may be determined as follows.
S401, aiming at each first candidate recommended picture set, determining the number of candidate recommended pictures in the first candidate recommended set, wherein the at least one candidate recommended picture set corresponds to the first candidate recommended picture set, and respectively generating the weight of each at least one candidate media material;
s402, determining a target recommended picture set from the first candidate recommended picture set based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word and the weight of each at least one candidate media material.
Here, for each first candidate recommended picture set, firstly, a weight of each at least one candidate media material may be generated based on the number of candidate recommended pictures in the first candidate recommended set included in the at least one candidate media material corresponding to the first candidate recommended picture set, that is, the weight of the at least one candidate media material corresponding to the first candidate recommended picture set having the larger number of candidate recommended pictures is higher.
In this way, the target recommended picture set with the best correlation and the largest weight can be selected from the first candidate recommended picture set by combining the correlation between the at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word and the weight of each at least one candidate media material.
After the target recommended picture set is determined, the target recommended pictures can be selected from the target recommended picture set. In the embodiment of the disclosure, a similarity set corresponding to the target recommended picture set may be obtained first based on the similarity between each candidate recommended picture in the target recommended picture set and each other candidate recommended picture, and then the target recommended picture may be selected based on the number and size of at least one similarity included in the similarity set. Next, a detailed description is given by the following two aspects.
In a first aspect: when determining that at least one similarity is the same similarity (that is, the number of similarities in the corresponding similarity set is 1), the embodiment of the disclosure may determine any candidate recommended picture in the target recommended picture set as the target recommended picture. In addition, when it is determined that at least one similarity is multiple similarities and the absolute value of the difference between different similarities is smaller than or equal to the second threshold (that is, the number of the similarities in the corresponding similarity set is greater than or equal to 2, and the similarities are relatively close), any one candidate recommended picture in the target recommended picture set may also be determined as the target recommended picture.
In a second aspect, when it is determined that at least one similarity is multiple similarities and the absolute value of the difference between different similarities is greater than a second threshold (that is, the number of the similarities in the corresponding similarity set is greater than or equal to 2, and the differences between the similarities are large), the target recommended picture sets may be grouped again to obtain multiple target recommended picture subsets, and then the target recommended picture subset with the largest number of candidate recommended pictures is selected from the target recommended picture subsets to be used as the target recommended picture subset, so as to select the target recommended picture. As shown in fig. 5, the above process of selecting the target recommended picture may be implemented as follows:
s501, attributing candidate recommended pictures with the same similarity in a target recommended picture set to the same group to obtain at least one target recommended picture subset;
s502, sequencing at least one target recommended picture subset according to the number of candidate recommended pictures included in the target recommended picture subset;
s503, determining any candidate recommended picture in the highest-ranked target recommended picture subset as a target recommended picture.
Here, based on the similarity between any two candidate recommended pictures in the target recommended picture set, the corresponding candidate recommended pictures with the same similarity may be assigned to the same target recommended picture subset, and then any one candidate recommended picture in the target recommended picture subset including the largest number of candidate recommended pictures may be determined as the target recommended picture.
In the information pushing method provided by the embodiment of the disclosure, the similarity between any two candidate recommended pictures can be determined based on the similarity between the picture fingerprint features extracted from the two candidate recommended pictures, that is, the more similar the picture fingerprint features are, the higher the corresponding similarity is.
The picture fingerprint feature in the embodiment of the present disclosure, as the identification feature of the picture, may be implemented by using a picture hash (hash) principle, that is, a candidate recommended picture may be used as an input, and after applying a hash function, a hash value of the picture may be calculated based on picture vision, and hash values of similar pictures are also similar. In this way, the similarity between two candidate recommended pictures can be determined based on the hash value similarity between any two candidate recommended pictures.
In the embodiment of the present disclosure, there may be one or more target recommended picture subsets with the highest ranking. When the target recommended picture subsets are multiple, firstly, for any target recommended picture subset in each target recommended picture subset, determining ranking results of candidate media materials to which any candidate recommended picture in the target recommended picture subset belongs in the N candidate media materials, then determining ranking results of the target recommended picture subset in each target recommended picture subset according to ranking results of candidate videos to which each candidate recommended picture belongs in the N candidate media materials, and finally determining any target recommended picture in the highest-ranking target recommended picture subset as a target recommended picture.
In the embodiment of the present disclosure, the N candidate media materials may be arranged in sequence, so that once the ranking results of the candidate media materials, to which any candidate recommended picture included in one target recommended picture subset belongs, in the N candidate media materials are determined, the ranking results of the target recommended picture subset in each target recommended picture subset may be determined.
For example, there are 10 candidate media materials and two target recommendation picture subsets, the ranking result of the candidate media material to which each candidate recommendation picture included in one target recommendation picture subset belongs among the 10 candidate media materials is distributed from 1 st to 5 th, and the ranking result of the candidate media material to which each candidate recommendation picture included in the other target recommendation picture subset belongs among the 10 candidate media materials is distributed from 6 th to 10 th, at this time, the target recommendation picture subsets whose ranking results are distributed from 1 st to 5 th may be determined as the final target recommendation picture subset, and the target recommendation picture may be selected from the final target recommendation picture subset.
Considering that the N candidate media materials determined in the method for pushing information provided by the embodiment of the present disclosure may be arranged in sequence, here, as shown in fig. 6, the N candidate media materials may be determined as follows:
s601, acquiring corresponding media materials according to the target recommended words;
s602, ranking the media materials according to the sequence of the relevance from big to small;
s603, determining the first N media materials with the ranking as N candidate media materials corresponding to the target recommended word.
Here, first, the corresponding media materials may be obtained according to the target recommended word, then, the media materials may be ranked in order from high to low based on the degree of correlation, and finally, the first N ranked media materials may be determined as N candidate media materials corresponding to the target recommended word.
The media material obtained according to the target recommended word may be obtained from an offline database, or may also be a media material displayed on a display interface of a search APP by each terminal, which is obtained in real time after the terminal initiates a search request.
After the media materials are obtained, the relevance between the target recommended word and the media materials can be determined based on the characteristic information of the media materials, the higher the relevance is, the higher the rank of the corresponding media materials is, and N candidate media materials corresponding to the target recommended word can be determined by using the ranking result.
For the specific representation forms of the media materials and the process of determining the correlation between the media materials and the target recommended word, reference may be specifically made to the specific representation forms of the candidate media materials and the related description contents of determining the correlation between the candidate media materials and the target recommended word, which are not described herein again.
In the information push method provided by the embodiment of the present disclosure, different candidate media materials have different ways of extracting candidate recommended pictures, and then a method of extracting candidate recommended pictures by using candidate videos as candidate media materials and a scheme of extracting candidate recommended pictures by using candidate web page contents as candidate media materials are described in the following two aspects.
On the first hand, when determining the candidate video as the candidate media material, the candidate video can be directly subjected to framing processing, and all the obtained candidate pictures are determined as candidate recommended pictures; the candidate video can be further intercepted according to a preset intercepting duration and/or a preset intercepting position to obtain the intercepted candidate video, all candidate pictures included in the intercepted candidate video are determined as candidate recommended pictures, and for example, for a candidate video with a video total length of 5s, the candidate video in the first 2 seconds can be intercepted to extract the candidate recommended pictures; the key frame candidate pictures can be extracted from all candidate pictures included in the candidate video based on a key frame extraction mode, and then the extracted key frame candidate pictures are determined as candidate recommended pictures.
In a second aspect, when determining web page content as a candidate media material, N web page contents corresponding to a target recommended word may be first crawled; the N webpage contents carry M candidate recommended pictures, and then the M candidate recommended pictures are extracted from the N webpage contents.
In the embodiment of the disclosure, the web page content corresponding to each target recommended word can be crawled based on a web crawler technology, and the higher the relevance between the key information contained in the web page content and the target recommended word is, the higher the corresponding ranking is.
Example two
As shown in fig. 7, a flowchart of an information pushing method provided in the second embodiment of the present disclosure is provided, where an execution subject of the information pushing method may be a terminal, and the information pushing method includes the following steps:
s701, receiving a target recommended word and a target recommended picture; the target recommended pictures are determined from the candidate recommended pictures based on the similarity among the M candidate recommended pictures, wherein the M candidate recommended pictures are extracted from N candidate media materials, and M is greater than or equal to N;
and S702, performing associated display on the target recommended word and the target recommended picture.
Here, after receiving the target recommended word and the target recommended picture, the terminal may perform the association display.
For the determination process of the target recommended picture, reference is made to the related description in the first embodiment, and details are not repeated here. The associated display form of the target recommended word and the target recommended picture may be as shown in fig. 2 (a).
The terminal in the embodiment of the disclosure may also display the corresponding media material based on the selection operation of the user for the target recommended word, that is, after determining the search request sent by the user for the target recommended word and the target icon recommended picture, the server may feed back the media material to the terminal, so that the terminal performs material display.
For the process of displaying the media material, reference is made to the related description in the first embodiment, which is not described herein again, and the manner of displaying the media material may be as shown in fig. 2 (b).
EXAMPLE III
Based on the above embodiments, the embodiments of the present disclosure further provide an information pushing device, and the following various implementations of the device may refer to the implementation of the method, and repeated details are not described again.
As shown in fig. 8, a schematic diagram of an information pushing apparatus provided in a third embodiment of the present disclosure is shown, where the apparatus includes:
an obtaining module 801, configured to determine a target recommended word; acquiring N candidate media materials corresponding to the target recommended word, and extracting M candidate recommended pictures from the N candidate media materials; wherein M is greater than or equal to N;
a determining module 802, configured to determine a target recommended picture from the candidate recommended pictures based on similarity between the candidate recommended pictures;
and the pushing module 803 is configured to associate the target recommended word with the target recommended picture, and send the associated target recommended word to the terminal.
In one embodiment, the determining module 802 is configured to determine the target recommended picture from the candidate recommended pictures according to the following steps:
grouping the M candidate recommended pictures based on the similarity among the candidate recommended pictures to obtain at least one candidate recommended picture set;
sorting at least one candidate recommended picture set according to the number of candidate recommended pictures included in the candidate recommended picture set;
determining a first candidate recommended picture set with the highest ranking as a target recommended picture set;
and selecting the target recommended picture from the target recommended picture set.
In an embodiment, if the number of the first candidate recommended picture set with the highest ranking is greater than or equal to 2, the determining module 802 is configured to determine the first candidate recommended picture set with the highest ranking as the target recommended picture set according to the following steps:
respectively obtaining candidate media materials from which each candidate recommended picture in the first candidate recommended picture set comes, and determining at least one candidate media material corresponding to each first candidate recommended picture set;
and determining a target recommended picture set from the first candidate recommended picture sets based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word.
In one embodiment, the determining module 802 is configured to determine the target recommended picture set from the first candidate recommended picture set according to the following steps:
for each first candidate recommended picture set, determining the number of candidate recommended pictures in the first candidate recommended set, which are included in at least one candidate media material corresponding to the first candidate recommended picture set, and respectively generating the weight of each at least one candidate media material;
and determining a target recommended picture set from the first candidate recommended picture set based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word and the weight of each at least one candidate media material.
In one embodiment, the determining module 802 is configured to obtain at least one candidate recommended picture set according to the following steps:
aiming at any candidate recommended picture in the M candidate recommended pictures, forming a candidate recommended picture set by the candidate recommended picture and other candidate recommended pictures with the similarity meeting preset conditions;
wherein the preset condition comprises at least one of the following conditions:
the similarity is greater than a first threshold;
and the similarity between other different candidate recommended pictures except the candidate recommended picture in the formed candidate recommended picture set and the candidate recommended picture is the same.
In one embodiment, the determining module 802 is configured to select a target recommended picture from the target recommended picture set according to the following steps:
determining the similarity between any candidate recommended picture in the target recommended picture set and other candidate recommended pictures to obtain a similarity set corresponding to the target recommended picture set;
if the number of the similarities in the similarity set is 1, or the number of the similarities in the similarity set is greater than or equal to 2, and the absolute value of the difference between the similarities is less than or equal to a second threshold, determining any candidate recommended picture in the target recommended picture set as a target recommended picture;
if the number of the similarities in the similarity set is greater than or equal to 2 and the absolute value of the difference between the similarities is greater than a second threshold, attributing candidate recommended pictures with the same similarity in the target recommended picture set to the same group to obtain at least one target recommended picture subset; sequencing at least one target recommended picture subset according to the number of candidate recommended pictures included in the target recommended picture subset; and determining any candidate recommended picture in the highest-ranked target recommended picture subset as a target recommended picture.
In one embodiment, if the target recommended picture subset is multiple; a determining module 802, configured to determine a target recommended picture according to the following steps:
aiming at any target recommended picture subset in each target recommended picture subset, determining the ranking result of candidate media materials to which any candidate recommended picture belongs in the target recommended picture subset in N candidate media materials; determining ranking results of the target recommended picture subsets in the target recommended picture subsets according to ranking results of the candidate media materials to which the candidate recommended pictures belong in the N candidate videos;
and determining any target recommended picture in the target recommended picture subset with the highest ranking as a target recommended picture.
In one embodiment, the determining module 802 is configured to determine the similarity between the candidate recommended pictures according to the following steps:
extracting a picture fingerprint feature from one candidate recommended picture in the two candidate recommended pictures;
and determining the similarity between the two candidate recommended pictures based on the similarity between the extracted picture fingerprint features.
In one embodiment, the obtaining module 801 is configured to obtain N candidate media materials corresponding to the target recommended word according to the following steps:
acquiring corresponding media materials according to the target recommended words;
ranking the media materials according to the sequence of the relevance from big to small;
and determining the first N media materials with the ranking as N candidate media materials corresponding to the target recommended word.
In one embodiment, the candidate media materials include candidate videos, and the obtaining module 801 is configured to extract M candidate recommended pictures from the N candidate media materials according to the following steps:
determining all candidate pictures included in the N candidate videos as M candidate recommended pictures; alternatively, the first and second electrodes may be,
intercepting any candidate video in the N candidate videos according to a preset intercepting duration and/or a preset intercepting position to obtain an intercepted candidate video, and determining all candidate pictures included in the intercepted N candidate videos as M candidate recommended pictures; alternatively, the first and second electrodes may be,
and extracting key frame candidate pictures from all candidate pictures included in any one of the N candidate videos, and determining the key frame candidate pictures in the extracted N candidate videos as M candidate recommended pictures.
In one embodiment, the candidate media material includes web page content; an obtaining module 801, configured to determine M candidate recommended pictures according to the following steps:
capturing N webpage contents corresponding to the target recommended words; the N webpage contents carry M candidate recommended pictures;
and extracting M candidate recommended pictures from the N webpage contents.
As shown in fig. 9, a schematic view of another information pushing apparatus provided in the third embodiment of the present disclosure is shown, where the apparatus includes:
a receiving module 901, configured to receive a target recommended word and a target recommended picture; the target recommended pictures are determined from the candidate recommended pictures based on the similarity among the M candidate recommended pictures, wherein the M candidate recommended pictures are extracted from N candidate media materials, and M is greater than or equal to N;
and the display module 902 is configured to perform associated display on the target recommended word and the target recommended picture.
Example four
A fourth embodiment of the present disclosure provides a computer device, where the computer device may be a server or a terminal, and when the server is used as a computer device, as shown in fig. 10, the computer device includes: the information push system comprises a processor 1001, a storage medium 1002 and a bus 1003, wherein the storage medium 1002 stores machine-readable instructions executable by the processor 1001 (such as execution instructions corresponding to the obtaining module 801, the determining module 802 and the pushing module 803 in the information push apparatus in fig. 8, and the like), when the computer device runs, the processor 1001 and the storage medium 1002 communicate through the bus 1003, and when the machine-readable instructions are executed by the processor 1001, the following processes are performed:
determining a target recommended word; acquiring N candidate media materials corresponding to the target recommended word, and extracting M candidate recommended pictures from the N candidate media materials; wherein M is greater than or equal to N;
determining a target recommended picture from the candidate recommended pictures based on the similarity between the candidate recommended pictures;
and associating the target recommended word with the target recommended picture, and sending the target recommended word to the terminal.
In an optional implementation manner, in the instructions executed by the processor 1001, determining a target recommended picture from the candidate recommended pictures based on a similarity between the candidate recommended pictures includes:
grouping the M candidate recommended pictures based on the similarity among the candidate recommended pictures to obtain at least one candidate recommended picture set;
sorting at least one candidate recommended picture set according to the number of candidate recommended pictures included in the candidate recommended picture set;
determining a first candidate recommended picture set with the highest ranking as a target recommended picture set;
and selecting the target recommended picture from the target recommended picture set.
In an optional implementation manner, if the number of the first candidate recommended picture sets with the highest rank is greater than or equal to 2, in the instruction executed by the processor 1001, determining the first candidate recommended picture set with the highest rank as the target recommended picture set includes:
respectively obtaining candidate media materials from which each candidate recommended picture in the first candidate recommended picture set comes, and determining at least one candidate media material corresponding to each first candidate recommended picture set;
and determining a target recommended picture set from the first candidate recommended picture sets based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word.
In an optional implementation manner, in the instructions executed by the processor 1001, determining a target recommended picture set from the first candidate recommended picture sets based on a correlation between at least one candidate media material corresponding to each first candidate recommended picture set and a target recommended word includes:
for each first candidate recommended picture set, determining the number of candidate recommended pictures in the first candidate recommended set, which are included in at least one candidate media material corresponding to the first candidate recommended picture set, and respectively generating the weight of each at least one candidate media material;
and determining a target recommended picture set from the first candidate recommended picture set based on the correlation between at least one candidate media material corresponding to each first candidate recommended picture set and the target recommended word and the weight of each at least one candidate media material.
In an optional implementation manner, in the instructions executed by the processor 1001, grouping the M candidate recommended pictures based on the similarity between the candidate recommended pictures to obtain at least one candidate recommended picture set includes:
aiming at any candidate recommended picture in the M candidate recommended pictures, forming a candidate recommended picture set by the candidate recommended picture and other candidate recommended pictures with the similarity meeting preset conditions;
wherein the preset condition comprises at least one of the following conditions:
the similarity is greater than a first threshold;
and the similarity between other different candidate recommended pictures except the candidate recommended picture in the formed candidate recommended picture set and the candidate recommended picture is the same.
In an optional implementation manner, the instructions executed by the processor 1001 to select a target recommended picture from a target recommended picture set includes:
determining the similarity between any candidate recommended picture in the target recommended picture set and other candidate recommended pictures to obtain a similarity set corresponding to the target recommended picture set;
if the number of the similarities in the similarity set is 1, or the number of the similarities in the similarity set is greater than or equal to 2, and the absolute value of the difference between the similarities is less than or equal to a second threshold, determining any candidate recommended picture in the target recommended picture set as a target recommended picture;
if the number of the similarities in the similarity set is greater than or equal to 2 and the absolute value of the difference between the similarities is greater than a second threshold, attributing candidate recommended pictures with the same similarity in the target recommended picture set to the same group to obtain at least one target recommended picture subset; sequencing at least one target recommended picture subset according to the number of candidate recommended pictures included in the target recommended picture subset; and determining any candidate recommended picture in the highest-ranked target recommended picture subset as a target recommended picture.
In an optional embodiment, if the target recommended picture subset is multiple; in the instructions executed by the processor 1001, determining any candidate recommended picture in the highest-ranked subset of target recommended pictures as a target recommended picture includes:
aiming at any target recommended picture subset in each target recommended picture subset, determining the ranking result of candidate media materials to which any candidate recommended picture belongs in the target recommended picture subset in N candidate media materials; determining the ranking results of the target recommended picture subsets in the target recommended picture subsets according to the ranking results of the candidate media materials to which the candidate recommended pictures belong in the N candidate media materials;
and determining any target recommended picture in the target recommended picture subset with the highest ranking as a target recommended picture.
In an alternative embodiment, in the instructions executed by the processor 1001, the similarity between the candidate recommended pictures is determined according to the following steps:
extracting a picture fingerprint feature from one candidate recommended picture in the two candidate recommended pictures;
and determining the similarity between the two candidate recommended pictures based on the similarity between the extracted picture fingerprint features.
In an optional implementation manner, the obtaining N candidate media materials corresponding to the target recommended word in the instruction executed by the processor 1001 includes:
acquiring corresponding media materials according to the target recommended words;
ranking the media materials according to the sequence of the relevance from big to small;
and determining the first N media materials with the ranking as N candidate media materials corresponding to the target recommended word.
In an alternative embodiment, the candidate media materials include candidate videos, and the instructions executed by the processor 1001 to extract M candidate recommended pictures from N candidate media materials include:
determining all candidate pictures included in the N candidate videos as M candidate recommended pictures; alternatively, the first and second electrodes may be,
intercepting any candidate video in the N candidate videos according to a preset intercepting duration and/or a preset intercepting position to obtain an intercepted candidate video, and determining all candidate pictures included in the intercepted N candidate videos as M candidate recommended pictures; alternatively, the first and second electrodes may be,
and extracting key frame candidate pictures from all candidate pictures included in any one of the N candidate videos, and determining the key frame candidate pictures in the extracted N candidate videos as M candidate recommended pictures.
In an alternative embodiment, the candidate media material comprises web page content; in the instructions executed by the processor 1001, the N candidate media materials corresponding to the target recommended word are determined according to the following steps:
capturing N webpage contents corresponding to the target recommended words; the N webpage contents carry M candidate recommended pictures;
extracting M candidate recommended pictures from the N candidate media materials according to the following steps:
and extracting M candidate recommended pictures from the N webpage contents.
When the terminal is used as a computer device, as shown in fig. 11, the computer device includes: the information push device comprises a processor 1101, a storage medium 1102 and a bus 1103, wherein the storage medium 1102 stores machine-readable instructions executable by the processor 1101 (such as execution instructions corresponding to a receiving module 901 and a presenting module 902 in the information push device in fig. 9, and the like), when the computer device runs, the processor 1101 communicates with the storage medium 1102 through the bus 1103, and the machine-readable instructions, when executed by the processor 1101, perform the following processing:
receiving a target recommended word and a target recommended picture; the target recommended pictures are determined from the candidate recommended pictures based on the similarity among the M candidate recommended pictures, wherein the M candidate recommended pictures are extracted from N candidate media materials, and M is greater than or equal to N;
and performing associated display on the target recommended words and the target recommended pictures.
The disclosed embodiment also provides a computer readable storage medium, on which a computer program is stored, where the computer program is executed by the processor 1001 to execute the steps of the information pushing method provided in the first embodiment, or the processor 1101 is executed to execute the steps of the information pushing method provided in the second embodiment.
Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, and the like, and when a computer program on the storage medium is run, the information push method can be executed, so that the problem of poor display effect caused by only recommending search terms in the related art is solved, information push is realized in a combined display mode of texts and pictures, and the display effect is improved.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to corresponding processes in the method embodiments, and are not described in detail in this disclosure. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and there may be other divisions in actual implementation, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some communication interfaces, and may be in an electrical, mechanical or other form.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.
The above are only specific embodiments of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present disclosure, and shall be covered by the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.