CN110659404A - Information recommendation method and device and storage medium - Google Patents
Information recommendation method and device and storage medium Download PDFInfo
- Publication number
- CN110659404A CN110659404A CN201810608262.4A CN201810608262A CN110659404A CN 110659404 A CN110659404 A CN 110659404A CN 201810608262 A CN201810608262 A CN 201810608262A CN 110659404 A CN110659404 A CN 110659404A
- Authority
- CN
- China
- Prior art keywords
- information
- multimedia information
- multimedia
- piece
- pieces
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 76
- 238000003860 storage Methods 0.000 title claims abstract description 40
- 230000002776 aggregation Effects 0.000 claims abstract description 38
- 238000004220 aggregation Methods 0.000 claims abstract description 38
- 238000012544 monitoring process Methods 0.000 claims description 25
- 230000003993 interaction Effects 0.000 claims description 23
- 239000013598 vector Substances 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000004931 aggregating effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000010365 information processing Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 14
- 238000013210 evaluation model Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 230000009193 crawling Effects 0.000 description 10
- 230000006399 behavior Effects 0.000 description 9
- 238000007619 statistical method Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 235000019633 pungent taste Nutrition 0.000 description 4
- 238000009826 distribution Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000000977 initiatory effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Abstract
The embodiment of the invention discloses an information recommendation method, an information recommendation device and a storage medium, which are applied to the technical field of information processing. In the information recommendation method in the embodiment of the invention, when the information recommendation device carries out information recommendation, the information recommendation device carries out theme aggregation on a plurality of pieces of multimedia information to form a multimedia information group mainly according to the content abstract information of the plurality of pieces of multimedia information, and then carries out information recommendation to the application terminal according to the heat value of each aggregated multimedia information group. Therefore, according to the content of the same theme, the multimedia information in one multimedia information group aggregated by the content abstract information belongs to the content of the same theme, the frequency of the occurrence of the multimedia information of each theme can be represented according to the heat value of each multimedia information group calculated according to the number of the information, and the timeliness of the multimedia information in each multimedia information group on the content is further reflected, so that the information recommendation device can recommend the information with stronger timeliness on the content to the application terminal.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to an information recommendation method, an information recommendation apparatus, and a storage medium.
Background
With the development of the internet industry, more and more information acquisition modes are provided, people gradually change from actively acquiring information to passively receiving information, and the information quantity is explosively increased in a geometric multiple mode, so that more and more junk information is provided, the cost of a user for acquiring valuable information through terminal equipment is greatly increased, but the acquisition speed of effective information is extremely important in many times, and at the moment, an information recommendation system is rapidly developed.
The information recommendation system is an information filtering system and can comprise terminal equipment and a server, wherein a user can operate the terminal equipment to enable the terminal equipment to log in the server through a corresponding user identifier, so that the server can recommend information for the terminal equipment; and the terminal device can perform certain operations on the information, such as comment, praise, collection, forwarding or sharing, and the like, when receiving the information recommended by the server.
In the prior art, when a server performs information recommendation, the method mainly includes: acquiring information such as content actively subscribed by a user through terminal equipment, content of articles and cooperative media sent by a public number owner, and determining classification labels of the information according to a certain method; predicting preference information corresponding to each user identifier according to historical operation data of the terminal equipment corresponding to each user identifier based on the terminal equipment display information; and finally, referring to the preference information of the user identification and the obtained classification label of the information, recommending information with the type consistent with that of the preference information of the corresponding user identification and with more historical click rate for each terminal device, and displaying the information, thereby improving the user experience and the information conversion rate.
However, since some information, such as news and information, which are just released, are usually very time-sensitive, but the terminal device usually has few operation records on the information, the information recommended to the terminal device by the existing information recommendation method may not include the information, thereby affecting the time for the information to be effective in the terminal device.
Disclosure of Invention
The embodiment of the invention provides an information recommendation method, an information recommendation device and a storage medium, which realize that at least one piece of multimedia information is recommended to an application terminal according to content summary information of a plurality of pieces of multimedia information.
A first aspect of an embodiment of the present invention provides an information recommendation method, including:
acquiring content abstract information corresponding to a plurality of pieces of multimedia information respectively;
performing theme aggregation on a plurality of pieces of multimedia information according to content summary information respectively corresponding to the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups;
respectively counting the number of pieces of multimedia information in each multimedia information group;
calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group;
and recommending at least one piece of multimedia information in the plurality of pieces of multimedia information to an application terminal according to the heat value of each multimedia information group.
A second aspect of the embodiments of the present invention provides an information recommendation method, including:
monitoring push information sent by an application server to an application terminal, wherein the push information comprises content summary information of multimedia information;
when the application terminal is detected to receive the push information, adding the content abstract information of the multimedia information into a push notification message;
and sending the push notification message so that the information recommendation device can recommend the information according to the content summary information of the multimedia information in the push notification message.
A third aspect of an embodiment of the present invention provides an information recommendation apparatus, including:
the information acquisition unit is used for acquiring content abstract information corresponding to a plurality of pieces of multimedia information respectively;
the aggregation unit is used for performing theme aggregation on the plurality of pieces of multimedia information according to the content abstract information of the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups;
the parameter acquisition unit is used for respectively counting the number of pieces of multimedia information in each multimedia information group;
the calculating unit is used for calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group;
and the recommending unit is used for recommending at least one piece of multimedia information in the plurality of pieces of multimedia information to the application terminal according to the heat value of each multimedia information group.
A fourth aspect of the embodiments of the present invention provides an application terminal, including:
the monitoring unit is used for monitoring push information sent to the information recommendation device by the application server, and the push information comprises content summary information of the multimedia information;
the adding unit is used for adding the content abstract information of the multimedia information into a push notification message when the information recommending device is detected to receive the push information;
and the notification unit is used for sending the push notification message so that the information recommendation device can recommend the information according to the content summary information of the multimedia information in the push notification message.
A fifth aspect of the embodiments of the present invention provides a storage medium, where the storage medium stores a plurality of instructions, and the instructions are adapted to be loaded by a processor and to perform the information recommendation method according to the first or second aspect of the embodiments of the present invention.
A sixth aspect of the embodiments of the present invention provides a server, including a processor and a storage medium, where the processor is configured to implement each instruction;
the storage medium is used for storing a plurality of instructions, and the instructions are used for being loaded by a processor and executing the information recommendation method according to the first aspect of the embodiment of the invention.
A seventh aspect of the embodiments of the present invention provides a terminal device, including a processor and a storage medium, where the processor is configured to implement each instruction;
the storage medium is used for storing a plurality of instructions, and the instructions are used for being loaded by a processor and executing the information recommendation method according to the second aspect of the embodiment of the invention.
As can be seen, in the information recommendation method according to the embodiment of the present invention, when information recommendation is performed by the information recommendation device, the plurality of pieces of multimedia information are subject-aggregated to form a multimedia information group according to the content summary information of the plurality of pieces of multimedia information, and then information recommendation is performed to the application terminal according to the heat value of each aggregated multimedia information group. Therefore, the multimedia information in one multimedia information group aggregated according to the content summary information belongs to the content of the same theme, and the counted number of pieces of information of each multimedia information group can represent the condition of the multimedia information under the same theme, so that the heat value of each multimedia information group calculated according to the number of pieces of information can represent the occurrence frequency of the multimedia information of each theme, and further reflect the timeliness of the multimedia information in each multimedia information group on the content, and the information recommendation device can recommend the information with stronger timeliness on the content to the application terminal.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1a is a schematic diagram of a scenario to which an information recommendation method according to an embodiment of the present invention is applied;
fig. 1b is a schematic diagram of another scenario in which an information recommendation method according to an embodiment of the present invention is applied;
FIG. 2 is a flow chart of an information recommendation method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a user interface displayed by an application terminal in one embodiment of the invention;
FIG. 4 is a diagram illustrating an information recommendation method according to another embodiment of the present invention;
FIG. 5a is a diagram of a user interface implemented by a terminal according to another embodiment of the present invention;
FIG. 5b is a diagram of another embodiment of a user interface for implementing a terminal;
fig. 6 is a schematic structural diagram of a system to which an information recommendation method according to an embodiment of the present invention is applied;
FIG. 7 is a diagram illustrating an information recommendation method according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating crawling multimedia information in an embodiment of the present invention;
fig. 9 is a schematic structural diagram of an information recommendation apparatus according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of an application terminal according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
An embodiment of the present invention provides an information recommendation method, which may be applied to a scenario shown in fig. 1a, where the scenario may include an application server and an application terminal, where the application server is configured to obtain information from multiple channels, for example, respectively obtain content summary information of corresponding multimedia information from multiple (illustrated in fig. 1a by taking n as an example) information source systems, determine timeliness of the multimedia information according to the obtained content summary information of the multimedia information, and recommend multimedia information whose actual content has stronger timeliness to the application terminal; the application terminal is used for receiving the multimedia information actively recommended by the application server for displaying, and can also actively acquire the latest multimedia information from the application server for displaying.
It can be seen that, in the scenario shown in fig. 1a, the information recommendation function of the embodiment of the present invention is integrated in the application server.
The method of the embodiment of the present invention may also be applied to a scenario as shown in fig. 1b, where the scenario may include at least one application terminal and an application server (1 is illustrated in fig. 1b as an example) in an application system, and an information recommendation system, where the information recommendation system is configured to obtain information from multiple channels, for example, respectively obtain content summary information of corresponding multimedia information from multiple (n is illustrated in fig. 1 b) information source systems, determine timeliness of the multimedia information according to the obtained content summary information of the multimedia information, and may also recommend multimedia information whose actual content has stronger timeliness to each application terminal; any application terminal is used for receiving the multimedia information actively recommended by the information recommendation system through the corresponding application server for displaying, and can also actively obtain the latest multimedia information from the information recommendation system through the corresponding application server for displaying.
It can be seen that in such a scenario as shown in fig. 1b, the information recommendation function of the embodiment of the present invention is independent of each application server, so that application terminals in multiple application systems can be merged.
The information source system in the scene can be a system for actively providing and operating information by a merchant user, such as news, headlines or commodity display, and can also be a self-media system such as a microblog, a blog, a WeChat, a post bar or a forum; the application server and the application terminal may be a server and a terminal device in the same application system, for example, the application terminal is an instant messaging client and the application server is an instant messaging server, or the application terminal is a news client and the application server is a news server.
The method of the embodiment of the present invention may also be applied to other scenarios, which are not illustrated herein.
The application server in the scenario shown in fig. 1a or the information recommendation system shown in fig. 1b may implement information recommendation by the following steps:
acquiring content abstract information corresponding to a plurality of pieces of multimedia information respectively, wherein the plurality of pieces of multimedia information come from a plurality of information source systems; performing theme aggregation on the plurality of pieces of multimedia information according to the content summary information respectively corresponding to the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups; respectively counting the number of pieces of multimedia information in each multimedia information group; calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group; and recommending at least one piece of multimedia information in the plurality of pieces of multimedia information to the application terminal according to the heat value of each multimedia information group.
Therefore, the multimedia information in one multimedia information group aggregated according to the content summary information belongs to the content of the same theme, and the counted number of the information pieces of each multimedia information group can represent the condition of the multimedia information under the same theme, so that the heat value of each multimedia information group calculated according to the number of the information pieces can represent the frequency of the multimedia information of each theme in the information source system, and further reflect the timeliness of the multimedia information in each multimedia information group in the content, and the information recommendation device can recommend the information with stronger timeliness in the content to the application terminal.
An information recommendation method provided by an embodiment of the present invention is a method executed by an information recommendation apparatus (for example, the application server shown in fig. 1a or the information recommendation system shown in fig. 1 b), and a flowchart is shown in fig. 2, and includes:
Here, a piece of multimedia information may include information in at least one of the following forms: information in the form of text, pictures, audio and video, for example, a piece of multimedia information may be an article. The content summary information corresponding to a piece of multimedia information may represent the specific content of the piece of multimedia information, and specifically may include a title of the multimedia information, and may further include information such as a publishing source, publishing time, and a content summary of the multimedia information, where the publishing source may include identification information of an information source system to which the multimedia information belongs, and the content summary may represent the main content of the piece of multimedia information.
It is understood that the information recommendation device may initiate the execution of the process of this embodiment according to a certain period, in one case, the information recommendation device may obtain the content summary information of the multimedia information from the plurality of information source systems and store the content summary information to the local before initiating the process of this embodiment, and call the content summary information of the multimedia information from the local when the information recommendation device initiates the process of this embodiment. In another case, the information recommendation device may directly obtain the content summary information of the multimedia information from the multiple information source systems when initiating the process of the embodiment.
The information recommending device can obtain content summary information of at least one piece of multimedia information from one information source system, and the at least one piece of multimedia information can be information with stronger timeliness in the information source system, such as newly generated news in one news system, or news with more user operation numbers, such as reading numbers, sharing numbers, collection numbers and the like.
When the information recommendation device obtains the multimedia information from the information source system, the following ways may be specifically included, but are not limited to:
(1) the information recommending device can directly acquire the content summary information respectively corresponding to the plurality of pieces of multimedia information in each information source system from each information source system according to a certain period. For example, content summary information corresponding to 10 pieces of newly released multimedia information is acquired from a news system.
(2) The information recommendation device receives Push notification messages respectively sent by each application terminal, wherein the Push notification messages comprise Push (Push) information sent by an application server to the corresponding application terminal, and the Push information comprises content summary information of the multimedia information.
In this case, the application terminal in each application system monitors the push information sent by the corresponding application server to the application terminal, and sends a push notification message to the information recommendation device, where the push notification message is used to notify the information recommendation device, and the application server sends the information to the application terminal.
For example, the news headline server pushes information such as the title, the release source, and the release time of a certain article to the news headline terminal, so that the news headline terminal sends the information such as the title, the release source, and the release time of the article to the information recommendation device through a push notification message.
It can be understood that, when performing topic aggregation, mainly aggregating a plurality of pieces of multimedia information belonging to the same topic into the same cluster, for example, 2 pieces of multimedia information are obtained from the information source system 1, where the 2 pieces of multimedia information are specifically different articles related to an entertainment event, and 1 piece of multimedia information related to the entertainment event is obtained from the information source system 2, the information recommendation device may aggregate the 3 pieces of multimedia information into the same cluster.
Specifically, the information recommendation device may perform topic aggregation by using a k-means algorithm, a Latent Dirichlet Allocation (LDA) algorithm based on a topic model, or a text clustering algorithm such as a Single-Pass (Single-Pass) algorithm, or may perform topic aggregation by using an extracted feature vector algorithm such as a word vector analysis (word2 vec).
When the information recommendation device adopts the feature vector extraction algorithm to perform theme aggregation, specifically: if the content abstract information of any piece of multimedia information comprises the title of the piece of multimedia information, the title of each piece of multimedia information needs to be segmented to obtain the title segmentation of each piece of multimedia information, and then the feature vector of each piece of multimedia information is determined according to the title segmentation of each piece of multimedia information; and finally, when the aggregation is carried out, if the distance between the feature vectors of the two pieces of multimedia information is smaller than a preset value, the two pieces of multimedia information can be aggregated into a multimedia information group with the same theme.
Since the information amount of the title of a certain piece of multimedia information is small, in order to improve the accuracy of topic aggregation, the information recommendation device can acquire the content summary of any piece of multimedia information when acquiring the content summary information of the piece of multimedia information.
And 103, respectively counting the number of the multimedia information in each multimedia information group.
And 104, calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group.
Specifically, for a certain multimedia information group, the information recommendation device may directly use the number of pieces of information of the multimedia information group as the heat value of the multimedia information group, or use a weighted value of the number of pieces of information as the heat value of the multimedia information group. Wherein, the weighted value of the number of the pieces of information of the multimedia information group is specifically as follows: the product of the number of pieces of information and the corresponding weight value.
The weight value of a certain multimedia information group can be determined according to information of at least one dimension of each piece of multimedia information in the multimedia information group, such as the dimension of an information source, the dimension of a theme type of the multimedia information, and the like. In an actual application process, an integral corresponding to each dimension of at least one piece of multimedia information in a multimedia information group may be determined according to information of at least one dimension of each piece of multimedia information in the at least one dimension, and then a weight value corresponding to the multimedia information group may be determined according to the determined integral, where the information of at least one dimension of a certain piece of multimedia information may be obtained according to the content summary information of the piece of multimedia information obtained in step 101. Specifically, the method comprises the following steps:
if one multimedia information group comprises one piece of multimedia information, the information recommendation device obtains corresponding integral according to the information of at least one dimensionality of the piece of multimedia information, and then determines the weight value of the multimedia information group according to the integral corresponding to the at least one dimensionality. For example, the score a1 is obtained according to the publishing time of the piece of multimedia information, the score a2 is obtained according to the publishing source of the piece of multimedia information, and the score a3 is obtained according to the topic type (for example, the topic type belongs to the time, science, technology, or finance) of the piece of multimedia information, and then the information recommendation device may add the score a1, a2 and a3 and then normalize the added value to obtain a weighted value of the multimedia information group.
If a multimedia information group comprises a plurality of pieces of multimedia information, the information recommendation device obtains average integrals corresponding to at least one dimension according to information of at least one dimension of each piece of multimedia information in the plurality of pieces of multimedia information, and then determines the weight value of the multimedia information group according to the average integrals corresponding to the at least one dimension. For example, if the multimedia information group includes multimedia information 1 and multimedia information 2, the information recommendation device may obtain the score b1 according to the distribution time of the multimedia information 1, obtain the score b2 according to the distribution source of the multimedia information 1, and obtain the score b3 according to the topic type of the multimedia information 1; and get the score c1 according to the release time of the multimedia information 2, get the score c2 according to the release source of the multimedia information 2, and get the score c3 according to the theme type of the multimedia information 2; obtaining an average integral d1 of the dimension of the publishing time according to the integrals b1 and c1, obtaining an average integral d2 of the dimension of the publishing source according to the integrals b2 and c2, and obtaining an average integral d3 of the dimension of the topic type according to the integrals b3 and c 3; finally, the information recommending device can add the integrals d1, d2 and d3 and then normalize the integrals to obtain values as the weight values of the multimedia information group.
And 105, recommending at least one piece of multimedia information in the plurality of pieces of multimedia information to the application terminal according to the heat value of each multimedia information group.
Specifically, the information recommending device recommends at least one piece of multimedia information in the multimedia information group with the heat value larger than the preset value to the application terminal. If a certain multimedia information group with the heat value larger than the preset value comprises a piece of multimedia information, the information recommending device can directly recommend the piece of multimedia information to the application terminal; if the multimedia information group with the heat value larger than the preset value comprises a plurality of pieces of multimedia information, the information recommending device can select at least one piece of multimedia information to recommend to the application terminal.
When at least one piece of multimedia information is selected, the information recommendation device can select the piece of multimedia information by combining subscription information, preference information and the like of a user identifier corresponding to the application terminal, and if a publishing source of one piece of multimedia information in the pieces of multimedia information is published by another user identifier in an information source system (such as a microblog system), and subscription information of the user identifier corresponding to the application terminal also comprises another user identifier, the information recommendation device can select the piece of multimedia information to recommend to the application terminal; if the theme type of a piece of multimedia information in the pieces of multimedia information is consistent with the preference type included in the preference information of the user identifier corresponding to the application terminal, the information recommendation device also selects the piece of multimedia information to recommend to the application terminal.
Specifically, when recommending a piece of multimedia information to the application terminal, the information recommending device may send address information for storing specific content of the piece of multimedia information and content summary information of the piece of multimedia information to the application terminal. And if the corresponding heat value of the piece of multimedia information is greater than the preset value, the information recommendation device determines that the content of the piece of multimedia information has stronger timeliness and is hot content, and can also send information for indicating that the piece of multimedia information belongs to the hot content to the application terminal.
And after receiving the information recommended by the information recommendation device, the application terminal stores the address information therein, displays the identification information of a link interface on the user interface, associates the link interface with the received address information, and the identification information of the link interface can be the content summary information of the piece of multimedia information. Further, if the application terminal further receives information sent by the information recommendation device and used for indicating that the piece of multimedia information belongs to the hot content, the application terminal also displays a hot mark of the piece of multimedia information on the user interface, wherein the hot mark is used for indicating that the content of the piece of multimedia information has stronger timeliness and belongs to the hot content. Therefore, the user can intuitively know the timeliness of the piece of multimedia information, when the user clicks the content abstract information of the piece of multimedia information displayed on the user interface, the application terminal can link the corresponding address information, acquire the specific content of the piece of multimedia information and display the specific content.
For example, fig. 3 shows a user interface displayed by the application terminal, where the user interface includes identification information of n linked interfaces, which are content summary information of n pieces of multimedia information, respectively, and a hotness mark of the multimedia information 1 is also displayed on the user interface, specifically, a position before the content summary information of the multimedia information 1 is displayed
As a heat mark of the multimedia information 1. The position and shape of the heat mark may also be in other forms, such as displaying a pentagram at a position before displaying the content summary information of the multimedia information 1, and the like, which is not exemplified here.
Moreover, it should be noted that the application terminal may actively acquire the information to be recommended from the information recommendation device, or passively receive the information recommended by the information recommendation device.
As can be seen, in the information recommendation method according to the embodiment of the present invention, when information recommendation is performed by the information recommendation device, the plurality of pieces of multimedia information are subject-aggregated to form a multimedia information group according to the content summary information of the plurality of pieces of multimedia information, and then information recommendation is performed to the application terminal according to the heat value of each aggregated multimedia information group. Therefore, the multimedia information in one multimedia information group aggregated according to the content summary information belongs to the content of the same theme, and the counted number of the information pieces of each multimedia information group can represent the condition of the multimedia information under the same theme, so that the heat value of each multimedia information group calculated according to the number of the information pieces can represent the frequency of the multimedia information of each theme in the information source system, and further reflect the timeliness of the multimedia information in each multimedia information group in the content, and the information recommendation device can recommend the information with stronger timeliness in the content to the application terminal.
In the above embodiment, the information recommendation device calculates the heat value of each multimedia information group mainly according to the content summary information of each multimedia information. Further, in a specific embodiment, the information recommendation device may further calculate the heat value in combination with other parameters, such as user interaction parameters and the like, specifically:
before the information recommendation device calculates the heat value in step 104, it may further obtain the parameter value of the user interaction parameter corresponding to each multimedia information group.
Here, if a multimedia information group includes a piece of multimedia information, the user interaction parameters of the multimedia information group refer to: after the multimedia information is respectively displayed on a plurality of terminal devices in the corresponding information source system, a user operates parameters of the multimedia information through each terminal device, such as the sum of operation quantities of forwarding number, collection number, sharing number, comment number, praise number and the like.
If a multimedia information group comprises a plurality of pieces of multimedia information, the user interaction parameters of the multimedia information group refer to: after a plurality of pieces of multimedia information are respectively displayed on a plurality of terminal devices in a corresponding information source system, a user respectively operates parameters of the plurality of pieces of multimedia information through each terminal device, such as average operands of forwarding number, collection number, sharing number, comment number, praise number and the like. For example, the pieces of multimedia information of a multimedia information group are denoted as: the sum of operands such as l1, l2, … …, lm and multimedia information l1 is M1, the sum of operands such as comment number, and like is M1, the sum of operands such as the comment number, collection number, share number, comment number, and like is M2, and the sum of operands such as the forwarding number, collection number, share number, comment number, and like, corresponding to … … multimedia information lm is Mm, and then the parameter values of the user interaction parameters corresponding to the multimedia information group may specifically be: m1, M2, … …, mean value of Mm.
When the information recommendation device obtains the parameter value of the user interaction parameter of a certain multimedia information group, the parameter value can be obtained according to the operation log of the terminal equipment in the information source system to each piece of multimedia information in the multimedia information group.
Then, the information recommendation device calculates the heat value of each multimedia information group according to the number of information items corresponding to each multimedia information group and the parameter value of the user interaction parameter when executing step 104.
Specifically, the heat value of a certain multimedia information group may specifically be: the number of pieces of information corresponding to the multimedia information group and the parameter value of the user interaction parameter are calculated by any function, such as the product of the number of pieces of information and the parameter value of the user interaction parameter, or the sum value.
Or, the heat value of a certain multimedia information group may specifically be: the weighted value of the number of the information pieces of the multimedia information group and the random function calculation value of the parameter value of the user interaction parameter are as follows: the product of the number of pieces of information and the corresponding weight value, and the method for determining the weight value of a certain multimedia information group can be described in the above embodiments, which is not repeated herein.
In another specific embodiment, after performing step 101, the information recommendation apparatus may further determine at least one attribute information of each piece of multimedia information, such as a region attribute, a topic classification attribute, and the like, according to the content summary information of each piece of multimedia information. In this way, when the information recommendation apparatus executes step 102, the information recommendation apparatus specifically performs topic aggregation on the plurality of pieces of multimedia information according to the content summary information and the attribute information respectively corresponding to the plurality of pieces of multimedia information, so as to obtain a plurality of multimedia information groups. Specifically, the following cases can be included but not limited to:
(1) the determined attribute information of each piece of multimedia information includes a region attribute, and the region attribute can be specifically determined according to information used for indicating a region in the content summary information of each piece of multimedia information. The region attribute of a piece of multimedia information may indicate the region information of the subject described by the piece of multimedia information, and the like.
When the information recommendation device carries out theme aggregation, at least one piece of multimedia information which belongs to the same theme and has the same regional attribute can be aggregated into the same cluster.
(2) The determined attribute information of each piece of multimedia information comprises a theme classification attribute, wherein the theme classification attribute specifically refers to a theme type, such as theme types of entertainment, science and technology, society, sports, finance, military, real-time affairs and the like. The topic classification attribute may be specifically determined according to the title of the multimedia information in the content summary information of each piece of multimedia information.
When the information recommendation device carries out theme aggregation, at least one piece of multimedia information which belongs to the same theme and has the same theme classification attribute can be aggregated into the same cluster.
(3) The determined attribute information of each piece of multimedia information comprises a region attribute and a subject classification attribute.
When the information recommendation device carries out theme aggregation, at least one piece of multimedia information which belongs to the same theme, has the same regional attribute and has the same theme classification attribute can be aggregated into the same cluster.
Therefore, in this embodiment, the information recommendation device may automatically and respectively determine one or more attribute information of each piece of multimedia information according to the content summary information of each piece of multimedia information, so that multimedia information with high timeliness under a certain attribute information can be finally determined according to the at least one attribute information. Such as multimedia information with strong timeliness belonging to the theme of "society".
An information recommendation method provided in an embodiment of the present invention is a method executed by an application terminal in fig. 1a or fig. 1b, where a flowchart is shown in fig. 4, and includes:
It can be understood that the application terminal can actively acquire the content summary information of the multimedia information updated by the application server from the corresponding application server; the content summary information of the multimedia information sent by the application server can also be passively received, in which case, the information sent by the application server is called push information.
The application terminal can automatically start and execute the process of the embodiment after being started; or, after the application terminal is started, the user enables the application terminal to start the pushed monitoring function by operating the application terminal, and then the application terminal executes the process of this embodiment.
In addition, when the application terminal executes step 101, different methods may be used for monitoring due to different application terminals. In one case, the application terminal includes a program of a corresponding application, and in this embodiment, a push monitoring program, that is, a hook program, needs to be deployed in the application terminal in a hook (hook) manner. Specifically, the application terminal may intercept the push information sent by the application server through an interface of a hook program in the application terminal, that is, a remote notification service (RemoteNotificationServer) interface. The remote notification service Interface is an Application Programming Interface (API) of hook, and when receiving the push information sent by the Application server, the Application terminal transmits the push information to the remote notification service Interface, so that the push monitoring program deployed in the Application terminal executes the following steps 202 and 203; meanwhile, the application terminal can also transmit the push information to the program of the corresponding application, and the program of the corresponding application displays the push information.
The application terminal in this case may be an application terminal having an Internet Operating System (IOS).
In another case, the application terminal includes the program of the corresponding application, and no other program needs to be deployed, specifically, the application terminal directly intercepts the push information sent by the application server through a program interface of the corresponding application, that is, a remote listening service (Notification listener service) interface. The remote monitoring service interface is a program API of a corresponding application, and after receiving the push information sent by the application server, the application terminal transmits the push information to the remote monitoring service interface, so that the program of the corresponding application in the application terminal executes the following steps 202 and 203, and the program of the corresponding application displays the push information.
The application terminal in this case may be an application terminal having an Android (Android) operating system.
It can be seen that, in this embodiment, the application terminal monitors the push information sent by the application server, and reports a piece of push information of the application server as content summary information of a piece of multimedia information, where one application server may be an information source system. Therefore, the content abstract information of each piece of multimedia information is monitored and obtained only when each application terminal operates, and an additional server is not needed for monitoring, so that the content abstract information of the multimedia information is conveniently obtained.
In a specific embodiment, when the application terminal receives information recommended by the information recommendation device, the recommended information includes content summary information of at least one piece of multimedia information and information for indicating that the at least one piece of multimedia information belongs to hot content, and the recommended information may further include address information for storing the at least one piece of multimedia information.
In this way, the application terminal displays the identification information of at least one link interface on the user interface according to the recommended information, the identification information of at least one link interface is the content summary information of at least one piece of multimedia information, and the application terminal can also associate the identification information of at least one link interface with the corresponding address information; and displaying the heat mark of at least one piece of multimedia information on the user interface according to the information indicating that at least one piece of multimedia information belongs to the hot content in the recommended information.
Further, in one case, if the recommended information further includes at least one piece of multimedia information respectively corresponding to a region attribute, if the region attribute of a piece of multimedia information in the at least one piece of multimedia information is consistent with the region to which the application terminal belongs, for example, the region attribute of a piece of multimedia information is "shenzhen", and the region to which the application terminal is located is also "shenzhen", etc., a local mark of a piece of multimedia information is also displayed on the user interface.
Thus, the piece of multimedia information has a local mark and a hot mark, and in this case, the application terminal can only display the local mark on the user interface to inform the user that the content of the piece of multimedia information belongs to the local hot content.
For example, fig. 5a shows a user interface displayed by an application terminal, where the user interface includes n link interfaces, each identified by content summary information of n pieces of multimedia information, where a hotness mark of multimedia information 1 and a local mark of multimedia information 2 are also displayed on the user interface, and specifically, a position before the content summary information of multimedia information 1 is displayed
As the hotness mark of the multimedia information 1, the position before the content summary information of the multimedia information 2 is displayed
As a heat mark of the multimedia information 1.
In another case, if the recommended information further includes at least one topic classification attribute corresponding to each piece of multimedia information, the application terminal displays the content summary information of the corresponding multimedia information as the identification information of the link interface on the user interface corresponding to the topic classification attribute when displaying the identification information of at least one link interface on the user interface according to the recommended information.
For example, fig. 5b shows a "social" user interface displayed by the application terminal, where the "social" user interface includes n link interfaces respectively identified by content summary information of n pieces of multimedia information, and a hotness mark of the multimedia information 1 is also displayed on the "social" user interface, specifically, a position before the content summary information of the multimedia information 1 is displayed
As a heat mark of the multimedia information 1. The theme classification attributes of the n pieces of multimedia information are social, and the application terminal can also display other types of user interfaces, such as user interfaces of financial, scientific, temporal and the like, and display the multimedia information with the corresponding theme classification attributes on the other types of user interfaces.
The information recommendation method according to the embodiment of the present invention is described below with a specific embodiment, and the information recommendation method according to the embodiment may be applied to the system shown in fig. 6, and specifically includes the following parts:
1. an instant communication client terminal, which is the application terminal
Multimedia information such as text, pictures, video and audio can be sent and received through an instant messaging server (not shown in the figure), and can be displayed;
acquiring index information of recommended multimedia information from a recommendation engine through a content acquisition and push interface service module, and then actively acquiring specific content of the multimedia information from a content storage center according to the index information;
for the multimedia information recommended by the recommendation engine, displaying a link interface of the multimedia information on a user interface according to a timeline dimension, and if the multimedia information is hot content, displaying a hot mark of the hot content on the user interface for displaying the multimedia information;
reporting a push notification message to a statistical analysis system through a statistical reporting interface service module, and reporting a log of multimedia information operation by a user through an instant messaging client by using the statistical reporting interface service module, wherein the log comprises: the method comprises the steps of commenting, sharing, praise, like or dislike, forwarding and other operation information, and can further comprise reading duration, clicking times, video content watching times, buffering and stopping information in the playing process and the like.
2. Content acquisition and push interface service module
The content obtaining and pushing interface service module is an interface for the instant messaging client to communicate with the recommendation engine, obtains the index information of the multimedia information requested by the user through the instant messaging client from the recommendation engine, and the index information comprises content abstract information of the multimedia information, such as the title of the multimedia information, a publisher, a content abstract, a cover page picture, publishing time, multimedia identification and the like;
acquiring corresponding multimedia information from a content storage center according to the acquired index information and providing the multimedia information to the instant messaging client; or directly sending the acquired index information to the instant messaging client, and acquiring the corresponding multimedia information from the content storage center by the instant messaging client according to the index information.
3. Content caching service module
In some cases, for example, if the instant messaging client is not on-line, the multimedia information that is not sent to the instant messaging client by the content acquisition and push interface service module in time can be temporarily stored in the content cache service module; and after the instant messaging client is on line, sending the stored multimedia information to the instant messaging client. The storage time is usually customized, typically 48 hours.
4. Statistical reporting interface service module
And the client terminal communicates with the instant messaging client terminal, receives the log information reported by the instant messaging client terminal and sends the log information to the statistical analysis system. The log information may include operation information of the instant messaging client on the displayed multimedia information, such as information of access, browsing, sharing, approval, forwarding or collection, and may also include information of a user identifier of the instant messaging client.
It should be noted that the content obtaining and pushing interface service module and the statistics reporting interface service module may be integrated into an instant messaging server.
5. Statistical analysis system
Performing statistical analysis according to log information of each application terminal (including application terminals such as an instant messaging client and the like) reported by a statistical reporting interface service module, and calculating the click rate, the achievement rate and the conversion rate of multimedia information, and basic result data such as user portrait, user operation behavior and the like;
communicate with the recommendation engine (not shown in fig. 6), and provide a series of parameters for information recommendation, such as user profile, user operation behavior, etc. to the recommendation engine.
Wherein the user portrayal is as follows: each application terminal corresponds to a set of information such as the age, sex, region, education level, topic type of typical interests (such as science and technology, entertainment, military and the like) and the like of a user. The user operation behavior is as follows: each application terminal corresponds to a set of information of browsing, forwarding, praise or collection of the multimedia information by the user identification.
6. Recommendation engine
Receiving multimedia information pushed by a content center, and reading hot content of each dimension, such as global hot content, regional hot content, topic classification hot content and the like, from a hot content recommendation interface service module;
acquiring a series of parameters from a statistical analysis system, such as user images of user identifications of all instant messaging clients, user operation behavior characteristics and other information, and determining preference information of all user identifications according to the user operation behavior characteristics, the user images and other information;
referring to the preference information of the user identification and the hot content of each dimension, recommending the hot content of each dimension with the type consistent with the preference information of the corresponding user identification for each instant messaging client; non-hot content can be recommended to each instant messaging client, wherein the non-hot content is multimedia information acquired from a content center;
when the instant messaging client actively acquires the multimedia information, hot content and non-hot content of each dimension are pushed to the instant messaging client through the content acquisition and pushing interface service module.
7. Auditing system
The multimedia information pushed to the content center can be ensured to meet the access permitted by local laws and policies by combining manual work and machine review; the audited multimedia information comes from a plurality of information source systems, such as: an editorial contribution platform, or a self-media distribution system, etc.
8. Content center
Receiving legal multimedia information which is approved by the auditing system; processing and filtering the multimedia information, establishing a content grading and quality evaluation system, and ensuring that the multimedia information provided for a recommendation engine is high-quality information suitable for exposure;
and providing the acquired content abstract information of each piece of multimedia information for the hotspot information evaluation model.
9. Content assembly service module
And acquiring the pushed multimedia information from the content center, assembling according to the font, the style and the layout suitable for the instant messaging client, and pushing the assembled information to the content storage center.
10. Content storage center
Usually, a group of servers which are widely distributed and are close to the instant messaging client cache the latest multimedia information; the multimedia information can be directly provided for the instant communication client, and the multimedia information can also be provided for the instant communication client through the content acquisition and push interface service module.
11. Self-media publishing system
The user can publish the multimedia information to the self-media publishing system through the self-media terminal, for example, the user publishes an article in a microblog system.
12. Editing manuscript supplying platform
The multimedia information is usually obtained by editing and processing through an editing and contribution platform by an operator, such as an article published by a certain publishing company.
13. Hot content crawling service module
Directionally monitoring various information source systems, such as a portal consultation website and the like, and crawling content summary information of a plurality of pieces of multimedia information in the various information source systems according to a certain period (such as 5 minutes), such as titles, authors, release time, texts and the like of the multimedia information; the text of the crawled multimedia information can be cleaned, namely, the information such as advertisements and the like is removed. The hot content crawling service module can mainly crawl hot contents in each information source system, such as content summary information of multimedia information in a popular ranking list.
Meanwhile, information sent by each application terminal and pushed to the corresponding application terminal by each information source system (such as a news system) is received, and the information may specifically include index information of multimedia information and the like.
14. Hot spot information evaluation model
Reading content abstract information of the multimedia information which enters a content center and is obtained from each information source system, and obtaining parameter values of user interaction parameters of each multimedia information from a statistical analysis system; hot spot evaluation can be carried out according to the method executed by the information recommendation device, namely, the multimedia information is subject aggregated to obtain a plurality of multimedia information groups, and the heat value of each multimedia information group is calculated;
providing content abstract information of the multimedia information in the multimedia information group with the heat value larger than the preset value, namely global hot content, to a recommendation engine through a hot content recommendation interface service module;
according to content abstract information and attribute information of the multimedia information entering the content center, hot content of other dimensions, such as region hot content, theme classification hot content and the like, is expanded, wherein the theme classification hot content refers to the multimedia information in a multimedia information group with the heat value larger than a preset value under each theme classification attribute, and the region hot content refers to the multimedia information in the multimedia information with the heat value larger than the preset value under each region attribute;
and scheduling the hot content crawling service module, so that the hot content crawling service module regularly updates the multimedia information from each information source system.
15. Hot content recommendation interface service module
Hot content of each dimension, such as global hot content, regional hot content or topic classification hot content, is mainly read from the hot information evaluation model to form different hot content queues for a recommendation engine to read.
It should be noted that, the combination of the hot content crawling service module, the hot information evaluation model and the hot content recommendation interface service module in this embodiment is the information recommendation device in the above embodiment.
Referring to fig. 7, in this embodiment, the information recommendation between the above parts may be implemented by the following method, including:
step 301, the auditing system acquires multimedia information from each information source system, for example, from an editing and providing platform, a media publishing system, or a hot content crawling service module and other channels; the auditing system audits the acquired multimedia information and pushes the multimedia information which accords with the preset rule to the content center.
Specifically, as shown in fig. 8, when monitoring information pushed by each information source system, the hot content crawling server module may specifically be implemented in the following manner, and the hot content crawling service module may specifically include a monitoring server, where:
when an application server in an application system sends push information to an application terminal, the push information may include content summary information of a plurality of pieces of multimedia information.
The application terminal receives the push information and sends a push notification message to the monitoring server, wherein the push notification message includes content summary information corresponding to the plurality of pieces of multimedia information, and the content summary information of each piece of multimedia information may include identification information of the piece of multimedia information and the like. Therefore, when the monitoring server receives the push notification message, the specific content of each piece of multimedia information is respectively transmitted to the external data source according to the identification information of each piece of multimedia information.
After the monitoring server obtains the specific content of each piece of multimedia information, format cleaning is carried out on each piece of multimedia information, for example, the contents such as advertisement content and diversion two-dimensional codes in the multimedia information are removed; and uploading the multimedia information subjected to format cleaning to a content center through an auditing system.
The application terminal and the application server can be terminals and servers in any application system, and the application terminal can be an instant messaging client or a news headline or other clients.
Step 302, the hotspot information evaluation model obtains content summary information corresponding to the plurality of pieces of multimedia information entering the content center, and determines attribute information of each piece of multimedia information.
Performing theme aggregation on the plurality of pieces of multimedia information according to the content abstract information to obtain a plurality of multimedia information groups; and counting the number of pieces of multimedia information in each multimedia information group. Therefore, the hotspot information evaluation model can obtain the occurrence frequency of the multimedia information belonging to the same theme in each information source system.
Step 303, the hotspot information evaluation model obtains the user operation behavior of each piece of multimedia information in each multimedia information group from the statistical analysis system, and determines the parameter value of the user interaction parameter corresponding to each multimedia information group according to the obtained user operation behavior.
Step 304, the hot spot information evaluation model calculates the heat value of each multimedia information group according to the number of the information pieces and the parameter value of the user interaction parameter, determines the multimedia information in the multimedia information group with the heat value larger than the preset value as global hot spot content, and can send the information of the global hot spot content to the recommendation engine through the hot spot content recommendation interface service module.
Furthermore, the hotspot information evaluation model also performs theme aggregation on a plurality of pieces of multimedia information according to the content summary information and attribute information (such as region attributes or theme classification attributes) to obtain a plurality of multimedia information groups, and counts the number of pieces of multimedia information in each multimedia information group, so that the hotspot information evaluation model can obtain the frequency of occurrence of the multimedia information belonging to the same theme and the same attribute in each information source system. And then the hot spot information evaluation model calculates the heat value of each multimedia information group according to the number of the information pieces and the parameter value of the user interaction parameter, determines the multimedia information in the multimedia information group with the heat value larger than the preset value as the hot spot content with a certain attribute, and can send the information of the hot spot content with the certain attribute to a recommendation engine through a hot spot content recommendation interface service module.
Step 305, the recommendation engine obtains a series of parameters from the statistical analysis system, such as user portraits corresponding to the user identifiers of the instant messaging client, user operation behaviors and other information, and determines preference information of the user identifiers according to the user operation behaviors, the user portraits and other information.
Then, the recommendation engine refers to the preference information of the user identifier and the hot content of each dimension acquired from the hot content recommendation interface service module, and determines the hot content of each dimension, such as global hot content, regional hot content or topic classification hot content, with the type consistent with the preference information of the user identifier corresponding to the instant messaging client.
And the recommendation engine can also determine the non-hot content with the type consistent with the preference information of the user identification corresponding to the instant communication client according to the multimedia information acquired from the content center.
Step 306, the recommendation engine sends the information of the hot content of each dimension determined in the step 305 to the instant messaging client through the content obtaining and pushing interface service module, and the instant messaging client displays the hot content.
And the recommendation engine may further send the information of the non-hot content determined in step 305 to the instant messaging client through the content obtaining and pushing interface service module, and the instant messaging client displays the hot content.
An embodiment of the present invention further provides an information recommendation apparatus, for example, an application server shown in fig. 1a or an information recommendation system shown in fig. 1b, where a schematic structural diagram of the information recommendation apparatus is shown in fig. 9, and the information recommendation apparatus may specifically include:
the information obtaining unit 10 is configured to obtain content summary information corresponding to a plurality of pieces of multimedia information, where the plurality of pieces of multimedia information are from a plurality of information source systems.
An information obtaining unit 10, configured to obtain content summary information of the newly released multimedia information from the plurality of information source systems, respectively; or, receiving push notification information respectively sent by each application terminal, where the push notification information includes content summary information of a plurality of pieces of multimedia information pushed to the corresponding application terminal by the application server.
The aggregation unit 11 is configured to perform theme aggregation on the multiple pieces of multimedia information according to the content summary information of the multiple pieces of multimedia information acquired by the information acquisition unit 10, so as to obtain multiple multimedia information groups.
If the content summary information of any piece of multimedia information includes a title of the piece of multimedia information, the aggregation unit 11 is specifically configured to perform word segmentation on the title of each piece of multimedia information to obtain a title word segmentation of each piece of multimedia information; determining the characteristic vector of each piece of multimedia information according to the title word segmentation of each piece of multimedia information; and if the distance between the feature vectors of the two pieces of multimedia information is smaller than a preset value, aggregating the two pieces of multimedia information into a multimedia information group with the same theme.
And the parameter obtaining unit 12 is configured to count the number of pieces of multimedia information in each multimedia information group obtained by the aggregation unit 11.
A calculating unit 13, configured to calculate a heat value of each multimedia information group according to the number of information items corresponding to each multimedia information group acquired by the parameter acquiring unit 12.
A recommending unit 14, configured to recommend at least one piece of multimedia information in the pieces of multimedia information to the application terminal according to the heat value of each multimedia information group calculated by the calculating unit 13.
Further, the parameter obtaining unit 12 is further configured to obtain parameter values of the user interaction parameters respectively corresponding to the multimedia information groups; the calculating unit 13 is specifically configured to calculate a heat value of each multimedia information group according to the number of information corresponding to each multimedia information group and the parameter value of the user interaction parameter.
When calculating the heat value of a certain multimedia information group according to the number of pieces of information corresponding to the certain multimedia information group and the parameter value of the user interaction parameter, the calculating unit 13 takes the weighted value of the number of pieces of information of the certain multimedia information group and an arbitrary function calculation value of the parameter value of the user interaction parameter as the heat value of the certain multimedia information group. The weighted value of the number of the information pieces of the certain multimedia information group is the product of the number of the information pieces of the certain multimedia information group and the corresponding weighted value.
Further, the information recommendation apparatus may further include: the attribute determining unit 15 is configured to determine at least one piece of attribute information of each piece of multimedia information according to content summary information of each piece of multimedia information;
the aggregation unit 11 is further configured to perform theme aggregation on the multiple pieces of multimedia information according to the content summary information and the attribute information determined by the attribute determining unit 15, where the content summary information corresponds to the multiple pieces of multimedia information, respectively, to obtain multiple multimedia information groups.
If the at least one piece of attribute information includes a region attribute, the aggregating unit 11 aggregates at least one piece of multimedia information belonging to the same topic and having the same region attribute into the same cluster; if the at least one attribute information includes a topic classification attribute, the aggregation unit 11 aggregates at least one piece of multimedia information belonging to the same topic and having the same topic classification attribute into the same cluster.
As can be seen, in the information recommendation apparatus in the embodiment of the present invention, the aggregation unit 11 mainly performs topic aggregation on a plurality of pieces of multimedia information according to content summary information of the plurality of pieces of multimedia information to form a multimedia information group, and then the recommendation unit 14 performs information recommendation to the application terminal according to the heat value of each aggregated multimedia information group. Therefore, the multimedia information in one multimedia information group aggregated according to the content summary information belongs to the content of the same theme, and the counted number of the information pieces of each multimedia information group can represent the condition of the multimedia information under the same theme, so that the heat value of each multimedia information group calculated according to the number of the information pieces can represent the frequency of the multimedia information of each theme in the information source system, and further reflect the timeliness of the multimedia information in each multimedia information group in the content, and the information recommendation device can recommend the information with stronger timeliness in the content to the application terminal.
An embodiment of the present invention further provides an application terminal, a schematic structural diagram of which is shown in fig. 10, and the application terminal may specifically include:
the monitoring unit 20 is configured to monitor push information sent by the application server to the information recommendation device, where the push information includes content summary information of the multimedia information.
The monitoring unit 20 is specifically configured to intercept push information sent by the application server through an interface of a hook program in the application terminal; or intercepting the push information sent by the application server through a program interface of a corresponding application in the application terminal.
An adding unit 21, configured to add content summary information of the multimedia information to a push notification message when the monitoring unit 20 detects that the information recommendation device receives the push information;
a notifying unit 22, configured to send the push notification message added by the adding unit 21, so that the information recommending apparatus recommends information according to the content summary information of the multimedia information in the push notification message.
Further, the application terminal may further include:
a recommendation receiving unit 23, configured to receive information recommended by the information recommendation apparatus, where the recommended information includes content summary information of at least one piece of multimedia information, and information used to indicate that the at least one piece of multimedia information belongs to hot content;
a display unit 24, configured to display, on a user interface, the identification information of the at least one link interface according to the recommended information received by the recommendation receiving unit 23, where the identification information of the at least one link interface is content summary information of the at least one piece of multimedia information; and displaying the heat mark of the at least one piece of multimedia information on the user interface according to the information indicating that the at least one piece of multimedia information belongs to the hot content in the recommended information.
The display unit 24 is further configured to display a local mark of a certain piece of multimedia information on the user interface if the recommended information further includes a region attribute corresponding to each of the at least one piece of multimedia information, and the region attribute of the certain piece of multimedia information is consistent with the region to which the application terminal belongs.
If the recommended information further includes the topic classification attribute corresponding to the at least one piece of multimedia information, the display unit 24 is specifically configured to display content summary information of the corresponding multimedia information on the user interface corresponding to the topic classification attribute, as identification information of the link interface.
It can be seen that, in the application terminal of this embodiment, the monitoring unit 20 monitors the push information sent by the application server, and the notification unit 22 reports a piece of push information of the application server as content summary information of a piece of multimedia information, where a certain application server may be an information source system. Therefore, the content abstract information of each piece of multimedia information is monitored and obtained only when each application terminal operates, and an additional server is not needed for monitoring, so that the content abstract information of the multimedia information is conveniently obtained.
Embodiments of the present invention also provide a server, which is schematically shown in fig. 11, and the server may generate a relatively large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 30 (e.g., one or more processors) and a memory 31, and one or more storage media 32 (e.g., one or more mass storage devices) for storing applications 321 or data 322. The memory 31 and the storage medium 32 may be, among other things, transient storage or persistent storage. The program stored on the storage medium 32 may include one or more modules (not shown), each of which may include a series of instruction operations for the server. Still further, the central processor 30 may be configured to communicate with the storage medium 32 to execute a series of instruction operations in the storage medium 32 on a server.
Specifically, the application 321 stored in the storage medium 32 includes an information recommendation application, and the program may include the information acquisition unit 10, the aggregation unit 11, the parameter acquisition unit 12, the calculation unit 13, the recommendation unit 14, and the attribute determination unit 15 in the information recommendation device, which will not be described herein again. Still further, the central processor 30 may be configured to communicate with the storage medium 32, and execute a series of operations corresponding to the application program recommended by the information stored in the storage medium 32 on the server.
The server may also include one or more power supplies 33, one or more wired or wireless network interfaces 34, one or more input-output interfaces 35, and/or one or more operating systems 323, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and so forth.
The steps performed by the information recommendation device described in the above method embodiments may be based on the structure of the server shown in fig. 11.
An embodiment of the present invention further provides a terminal device, which has a structure similar to that of the server shown in fig. 11, and is different from the server shown in fig. 11, in the terminal device of this embodiment:
the application programs stored in the storage medium include information recommendation application programs, and the programs may include the monitoring unit 20, the adding unit 21, the notifying unit 22, the recommendation receiving unit 23, and the display unit 24 in the application terminal, which will not be described herein. Further, the central processor may be configured to communicate with the storage medium, and execute a series of operations corresponding to the application program recommended by the information stored in the storage medium on the terminal device.
The embodiment of the present invention further provides a storage medium, where the storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the information recommendation method executed by the information recommendation apparatus or the application terminal.
The embodiment of the invention also provides a server, which comprises a processor and a storage medium, wherein the processor is used for realizing each instruction;
the storage medium is used for storing a plurality of instructions which are used for being loaded by the processor and executing the information recommendation method executed by the information recommendation device.
The embodiment of the invention also provides terminal equipment, which comprises a processor and a storage medium, wherein the processor is used for realizing each instruction;
the storage medium is used for storing a plurality of instructions which are used for loading and executing the information recommendation method executed by the application terminal by the processor.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
The information recommendation method, apparatus and storage medium provided by the embodiments of the present invention are described in detail above, and the principles and embodiments of the present invention are explained herein by applying specific examples, and the description of the embodiments is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (19)
1. An information recommendation method, comprising:
acquiring content abstract information corresponding to a plurality of pieces of multimedia information respectively;
performing theme aggregation on a plurality of pieces of multimedia information according to content summary information respectively corresponding to the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups;
respectively counting the number of pieces of multimedia information in each multimedia information group;
calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group;
and recommending at least one piece of multimedia information in the plurality of pieces of multimedia information to an application terminal according to the heat value of each multimedia information group.
2. The method according to claim 1, wherein the obtaining content summary information corresponding to the plurality of pieces of multimedia information respectively comprises:
respectively acquiring content abstract information of the multimedia information from the plurality of information source systems; alternatively, the first and second electrodes may be,
and receiving push notification information respectively sent by each application terminal, wherein the push notification information comprises content summary information of a plurality of pieces of multimedia information pushed to the corresponding application terminal by the application server.
3. The method according to claim 1, wherein the content summary information of any piece of multimedia information includes a title of the piece of multimedia information, and the topic aggregation is performed on the pieces of multimedia information according to the content summary information corresponding to the pieces of multimedia information, specifically including:
segmenting the title of each piece of multimedia information to obtain the title segmentation of each piece of multimedia information;
determining the characteristic vector of each piece of multimedia information according to the title word segmentation of each piece of multimedia information;
and if the distance between the feature vectors of the two pieces of multimedia information is smaller than a preset value, aggregating the two pieces of multimedia information into a multimedia information group with the same theme.
4. The method according to claim 1, wherein after obtaining the content summary information corresponding to each of the plurality of pieces of multimedia information, the method further comprises:
respectively determining at least one attribute information of each piece of multimedia information according to the content abstract information of each piece of multimedia information;
performing theme aggregation on the multiple pieces of multimedia information according to the content summary information respectively corresponding to the multiple pieces of multimedia information to obtain multiple multimedia information groups, specifically including:
and performing theme aggregation on the plurality of pieces of multimedia information according to the content summary information and the attribute information respectively corresponding to the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups.
5. The method of claim 4,
if the at least one attribute information includes a region attribute, performing topic aggregation on the plurality of pieces of multimedia information according to the content summary information and the attribute information respectively corresponding to the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups, specifically including: aggregating at least one piece of multimedia information which belongs to the same theme and has the same region attribute into the same cluster;
if the at least one attribute information includes a topic classification attribute, performing topic aggregation on the multiple pieces of multimedia information according to the content summary information and the attribute information respectively corresponding to the multiple pieces of multimedia information to obtain multiple multimedia information groups, specifically including: and aggregating at least one piece of multimedia information which belongs to the same theme and has the same theme classification attribute into the same cluster.
6. The method of any of claims 1 to 5, wherein prior to said calculating a heat value for said respective set of multimedia information, the method further comprises:
acquiring parameter values of user interaction parameters respectively corresponding to the multimedia information groups;
calculating the heat value of each multimedia information group according to the number of information corresponding to each multimedia information group, specifically comprising: and calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group and the parameter value of the user interaction parameter.
7. The method of claim 6, wherein calculating the heat value of a certain multimedia message group according to the number of pieces of information corresponding to the certain multimedia message group and the parameter value of the user interaction parameter comprises:
taking a weighted value of the number of information pieces of the certain multimedia information group and an arbitrary function calculation value of a parameter value of a user interaction parameter as a heat value of the certain multimedia information group;
the weighted value of the number of the information pieces of the certain multimedia information group is the product of the number of the information pieces of the certain multimedia information group and the corresponding weighted value; and the user interaction parameter corresponding to a certain multimedia information group is a parameter operated by the terminal equipment after the multimedia information in the certain multimedia information group is displayed by the terminal equipment in the corresponding information source system.
8. An information recommendation method is applied to an application terminal, and comprises the following steps:
monitoring push information sent by an application server to an application terminal, wherein the push information comprises content summary information of multimedia information;
when the application terminal is detected to receive the push information, adding the content abstract information of the multimedia information into a push notification message;
and sending the push notification message so that the information recommendation device can recommend the information according to the content summary information of the multimedia information in the push notification message.
9. The method according to claim 8, wherein the monitoring the push information sent by the application server to the application terminal specifically includes:
intercepting the push information sent by the application server through an interface of a hook program in an application terminal; or intercepting the push information sent by the application server through a program interface of a corresponding application in the application terminal.
10. The method of claim 8 or 9, wherein the method further comprises:
receiving information recommended by the information recommendation device, wherein the recommended information comprises content summary information of at least one piece of multimedia information and information used for indicating that the at least one piece of multimedia information belongs to hot content;
displaying the identification information of the at least one link interface on a user interface according to the recommended information, wherein the identification information of the at least one link interface is the content summary information of the at least one piece of multimedia information;
and displaying the heat mark of the at least one piece of multimedia information on the user interface according to the information indicating that the at least one piece of multimedia information belongs to the hot content in the recommended information.
11. The method as claimed in claim 10, wherein if the recommended information further includes a region attribute corresponding to each of the at least one piece of multimedia information, the method further comprises:
and if the regional attribute of a certain piece of multimedia information is consistent with the region to which the application terminal belongs, displaying the local mark of the certain piece of multimedia information on the user interface.
12. The method according to claim 10, wherein if the recommended information further includes a topic classification attribute corresponding to each of the at least one piece of multimedia information, displaying, on a user interface, identification information of at least one link interface according to the recommended information, where the identification information of at least one link interface is content summary information of the at least one piece of multimedia information, specifically including:
and displaying the content abstract information of the corresponding multimedia information on the user interface corresponding to the theme classification attribute as the identification information of the link interface.
13. An information recommendation apparatus, comprising:
the information acquisition unit is used for acquiring content abstract information corresponding to a plurality of pieces of multimedia information respectively;
the aggregation unit is used for performing theme aggregation on the plurality of pieces of multimedia information according to the content abstract information of the plurality of pieces of multimedia information to obtain a plurality of multimedia information groups;
the parameter acquisition unit is used for respectively counting the number of pieces of multimedia information in each multimedia information group;
the calculating unit is used for calculating the heat value of each multimedia information group according to the number of the information corresponding to each multimedia information group;
and the recommending unit is used for recommending at least one piece of multimedia information in the plurality of pieces of multimedia information to the application terminal according to the heat value of each multimedia information group.
14. The apparatus of claim 13, further comprising:
the attribute determining unit is used for respectively determining at least one piece of attribute information of each piece of multimedia information according to the content abstract information of each piece of multimedia information;
the aggregation unit is specifically configured to perform theme aggregation on the multiple pieces of multimedia information according to the content summary information and the attribute information respectively corresponding to the multiple pieces of multimedia information, so as to obtain multiple multimedia information groups.
15. An application terminal, comprising:
the monitoring unit is used for monitoring push information sent to the information recommendation device by the application server, and the push information comprises content summary information of the multimedia information;
the adding unit is used for adding the content abstract information of the multimedia information into a push notification message when the information recommending device is detected to receive the push information;
and the notification unit is used for sending the push notification message so that the information recommendation device can recommend the information according to the content summary information of the multimedia information in the push notification message.
16. The application terminal of claim 15, further comprising:
the recommendation receiving unit is used for receiving information recommended by the information recommendation device, wherein the recommended information comprises content abstract information of at least one piece of multimedia information and information used for indicating that the at least one piece of multimedia information belongs to hot content;
the display unit is used for displaying the identification information of the at least one link interface on a user interface according to the recommended information, wherein the identification information of the at least one link interface is the content summary information of the at least one piece of multimedia information; and displaying the heat mark of the at least one piece of multimedia information on the user interface according to the information indicating that the at least one piece of multimedia information belongs to the hot content in the recommended information.
17. A storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the information recommendation method according to any one of claims 1 to 12.
18. A server comprising a processor and a storage medium, the processor configured to implement instructions;
the storage medium is used for storing a plurality of instructions which are used for being loaded by a processor and executing the information recommendation method according to any one of claims 1 to 7.
19. A terminal device comprising a processor and a storage medium, the processor configured to implement instructions;
the storage medium is used for storing a plurality of instructions which are used for being loaded by a processor and executing the information recommendation method according to any one of claims 8 to 12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810608262.4A CN110659404B (en) | 2018-06-13 | 2018-06-13 | Information recommendation method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810608262.4A CN110659404B (en) | 2018-06-13 | 2018-06-13 | Information recommendation method, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110659404A true CN110659404A (en) | 2020-01-07 |
| CN110659404B CN110659404B (en) | 2024-03-15 |
Family
ID=69026168
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810608262.4A Active CN110659404B (en) | 2018-06-13 | 2018-06-13 | Information recommendation method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659404B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111324753A (en) * | 2020-01-22 | 2020-06-23 | 天窗智库文化传播(苏州)有限公司 | Media information publishing management method and system |
| CN111475081A (en) * | 2020-03-31 | 2020-07-31 | 维沃移动通信有限公司 | Information display method and electronic equipment |
| CN111966733A (en) * | 2020-08-18 | 2020-11-20 | 中国银行股份有限公司 | Hotspot knowledge generation method and device |
| CN111259259B (en) * | 2020-03-11 | 2021-03-30 | 郑州工程技术学院 | University student news recommendation method, device, equipment and storage medium |
| CN113836355A (en) * | 2021-10-20 | 2021-12-24 | 盐城金堤科技有限公司 | Video recommendation method and device, computer storage medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130212493A1 (en) * | 2012-02-09 | 2013-08-15 | Kishore Adekhandi Krishnamurthy | Efficient multimedia content discovery and navigation based on reason for recommendation |
| CN104598450A (en) * | 2013-10-30 | 2015-05-06 | 北大方正集团有限公司 | Popularity analysis method and system of network public opinion event |
| CN106354867A (en) * | 2016-09-12 | 2017-01-25 | 传线网络科技（上海）有限公司 | Multimedia resource recommendation method and device |
| CN106570006A (en) * | 2015-10-08 | 2017-04-19 | 腾讯科技（深圳）有限公司 | Multimedia file recommendation system and releasing information storage method |
| CN107979820A (en) * | 2016-10-25 | 2018-05-01 | 腾讯科技（北京）有限公司 | A kind of method and device, equipment for testing PUSH message |
-
2018
- 2018-06-13 CN CN201810608262.4A patent/CN110659404B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130212493A1 (en) * | 2012-02-09 | 2013-08-15 | Kishore Adekhandi Krishnamurthy | Efficient multimedia content discovery and navigation based on reason for recommendation |
| CN104598450A (en) * | 2013-10-30 | 2015-05-06 | 北大方正集团有限公司 | Popularity analysis method and system of network public opinion event |
| CN106570006A (en) * | 2015-10-08 | 2017-04-19 | 腾讯科技（深圳）有限公司 | Multimedia file recommendation system and releasing information storage method |
| CN106354867A (en) * | 2016-09-12 | 2017-01-25 | 传线网络科技（上海）有限公司 | Multimedia resource recommendation method and device |
| CN107979820A (en) * | 2016-10-25 | 2018-05-01 | 腾讯科技（北京）有限公司 | A kind of method and device, equipment for testing PUSH message |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111324753A (en) * | 2020-01-22 | 2020-06-23 | 天窗智库文化传播(苏州)有限公司 | Media information publishing management method and system |
| CN111324753B (en) * | 2020-01-22 | 2021-09-03 | 天窗智库文化传播(苏州)有限公司 | Media information publishing management method and system |
| CN111259259B (en) * | 2020-03-11 | 2021-03-30 | 郑州工程技术学院 | University student news recommendation method, device, equipment and storage medium |
| CN111475081A (en) * | 2020-03-31 | 2020-07-31 | 维沃移动通信有限公司 | Information display method and electronic equipment |
| CN111475081B (en) * | 2020-03-31 | 2022-02-15 | 维沃移动通信有限公司 | Information display method and electronic equipment |
| CN111966733A (en) * | 2020-08-18 | 2020-11-20 | 中国银行股份有限公司 | Hotspot knowledge generation method and device |
| CN113836355A (en) * | 2021-10-20 | 2021-12-24 | 盐城金堤科技有限公司 | Video recommendation method and device, computer storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110659404B (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11727048B2 (en) | Methods, systems, and media for presenting recommended content based on social cues | |
| CN107222566B (en) | Information pushing method and device and server | |
| CN110659404B (en) | Information recommendation method, device and storage medium | |
| CN107463641B (en) | System and method for improving access to search results | |
| US9953063B2 (en) | System and method of providing a content discovery platform for optimizing social network engagements | |
| JP5779740B1 (en) | Receiving information about users from third-party applications based on action type | |
| WO2018192437A1 (en) | Media content recommendation method, server, client and storage medium | |
| US20170286539A1 (en) | User profile stitching | |
| US10540660B1 (en) | Keyword analysis using social media data | |
| US8893012B1 (en) | Visual indicator based on relative rating of content item | |
| CN108960975A (en) | Personalized Precision Marketing Method, server and storage medium based on user's portrait | |
| CN107426328B (en) | Information pushing method and device | |
| KR20160055800A (en) | Selecting content items for presentation to a social networking system user in a newsfeed | |
| US20180365709A1 (en) | Personalized creator recommendations | |
| CN111178970B (en) | Advertisement putting method and device, electronic equipment and computer readable storage medium | |
| EP3873065B1 (en) | Content recommendation method, mobile terminal, and server | |
| US20210209624A1 (en) | Online platform for predicting consumer interest level | |
| CN114448921A (en) | Information display method and device based on social relation chain and server | |
| JP2011227721A (en) | Interest extraction device, interest extraction method, and interest extraction program | |
| CN106487655B (en) | Message interaction method and device and processing server | |
| US20210004844A1 (en) | Building topic-oriented audiences | |
| EP3321824A1 (en) | Data retrieval system | |
| US7788249B2 (en) | System and method for automatically generating a result set | |
| CN116208579A (en) | Information pushing method, device, equipment and storage medium | |
| CN115187330A (en) | Product recommendation method, device, equipment and medium based on user label |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40018768 Country of ref document: HK |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221117 Address after: 1402, Floor 14, Block A, Haina Baichuan Headquarters Building, No. 6, Baoxing Road, Haibin Community, Xin'an Street, Bao'an District, Shenzhen, Guangdong 518,101 Applicant after: Shenzhen Yayue Technology Co.,Ltd. Address before: 518057 Tencent Building, No. 1 High-tech Zone, Nanshan District, Shenzhen City, Guangdong Province, 35 floors Applicant before: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |