CN112445970A - Information recommendation method and device, electronic equipment and storage medium - Google Patents
Information recommendation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112445970A CN112445970A CN201910838698.7A CN201910838698A CN112445970A CN 112445970 A CN112445970 A CN 112445970A CN 201910838698 A CN201910838698 A CN 201910838698A CN 112445970 A CN112445970 A CN 112445970A
- Authority
- CN
- China
- Prior art keywords
- information
- multimedia information
- multimedia
- target account
- target
- 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 69
- 238000012216 screening Methods 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims description 19
- 238000013507 mapping Methods 0.000 claims description 11
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 14
- 238000012549 training Methods 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 239000013598 vector Substances 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000000342 Monte Carlo simulation Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000003997 social interaction Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure relates to an information recommendation method, an information recommendation device, an electronic device and a storage medium, wherein the method comprises the following steps: determining target account characteristics of a target account, wherein the target account is an account to receive information recommendation results; retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information; screening out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information; and recommending the target information recommendation result to the target account. The method and the device can solve the problem that the type of the recommendation information is single in the related art.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to an information recommendation method and apparatus, an electronic device, and a storage medium.
Background
With the development of computer technology, more and more websites or application software can provide information recommendation services for accounts. By providing personalized information recommendation service for the account, on one hand, the account of a website or application software can obtain better use experience, and on the other hand, the information provided by an information provider can obtain more attention or click rate and the like.
Currently, when information is recommended for an account, only a certain type of information is recommended for the account. For example, only videos are recommended to the account, or only pictures are recommended to the account. Although information recommendation is realized, information recommended for the account is not abundant due to the fact that the type of the recommended information is single.
Invention information
The disclosure provides an information recommendation method, an information recommendation device, an electronic device and a storage medium, which are used for at least solving the problem of single recommendation information type in the related art. The technical scheme of the disclosure is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided an information recommendation method, including:
determining target account characteristics of a target account, wherein the target account is an account to receive information recommendation results;
retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information;
screening out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information;
and recommending the target information recommendation result to the target account.
Optionally, the step of retrieving, from a preset information base storing multiple types of information, each type of multimedia information matching the target account feature includes:
determining the target account characteristic as an index value;
aiming at each type of multimedia information in a preset information base, determining a first candidate information set from the type of multimedia information by using the index value and a pre-established reverse index of the type of multimedia information; determining information characteristics of each multimedia information in the first candidate information set; predicting the matching degree of the target account and each piece of multimedia information according to the target account characteristics and the information characteristics of each piece of multimedia information; and determining multimedia information matched with the target account characteristics from the first candidate information set by utilizing the predicted matching degree.
Optionally, the process of establishing the inverted index for each type of multimedia information includes:
determining at least one tag for each of the types of multimedia information;
removing the duplicate of the determined label to obtain a first label set;
for each tag in the first tag set, determining target multimedia information containing the tag in the type of multimedia information, and establishing a mapping relation between the tag and an information ID of the determined target multimedia information to obtain a reverse arrangement table of the tag for the type of multimedia information;
and taking the established inverted list of each label as an inverted index of the type of multimedia information.
Optionally, after the step of determining the target account characteristic as an index value, the method further includes:
determining the index value as a feature tag;
the step of determining a first candidate information set from the type of multimedia information by using the index value and a pre-established reverse index of the type of multimedia information includes:
determining a target label in the first label set that matches the feature label;
determining the information ID recorded in the inverted arrangement table of each target label, and removing the duplication of the determined information ID;
and taking the multimedia information corresponding to the information ID after the duplication removal as a first candidate information set.
Optionally, the step of determining, from the first candidate information set, multimedia information that matches the target account feature by using the predicted matching degree includes:
sequencing all multimedia information in the first candidate information set according to the sequence of the corresponding matching degrees from large to small to obtain an information sequence;
and determining multimedia information matched with the target account characteristics by using the information sequence.
Optionally, the step of determining, by using the information sequence, multimedia information matching the target account feature includes:
and adjusting the sequence of each multimedia information in the information sequence by using the heat of each multimedia information in the information sequence to obtain an adjusted information sequence, wherein the heat of any multimedia information is used for expressing: the frequency of access to the multimedia information within a predetermined time period;
and taking the multimedia information in the adjusted information sequence as the multimedia information matched with the target account characteristics.
Optionally, the step of screening out target information recommendation results including at least two types of multimedia information from the retrieved multimedia information includes:
selecting the same number of multimedia information from each type of multimedia information in the retrieved at least two types of multimedia information, and taking the selected multimedia information as a target information recommendation result; alternatively, the first and second electrodes may be,
determining the selection number corresponding to the type of multimedia information according to the weight of each type of multimedia information in the at least two types of retrieved multimedia information and the total number of preset target information recommendation results, selecting the selected number of multimedia information from the retrieved type of multimedia information, and taking the selected multimedia information as a target information recommendation result; alternatively, the first and second electrodes may be,
and randomly selecting multimedia information from each type of multimedia information in the at least two types of retrieved multimedia information, and taking the selected multimedia information as a target information recommendation result.
Optionally, the multiple types of multimedia information include: at least two of the predetermined types of multimedia information, the predetermined types of multimedia information including at least: video, pictures, text, and animation.
According to a second aspect of the embodiments of the present disclosure, there is provided an information recommendation apparatus including:
the determining module is configured to determine target account characteristics of a target account, wherein the target account is an account to receive the information recommendation result;
the retrieval module is configured to retrieve various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information;
the screening module is configured to screen out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information;
a recommending module configured to recommend the target information recommending result to the target account.
Optionally, the retrieving module includes:
a first determination submodule configured to determine the target account feature as an index value;
the second determining submodule is configured to determine a first candidate information set from each type of multimedia information in a preset information base by using the index value and a pre-established reverse index of the type of multimedia information; determining information characteristics of each multimedia information in the first candidate information set; predicting the matching degree of the target account and each piece of multimedia information according to the target account characteristics and the information characteristics of each piece of multimedia information; and determining multimedia information matched with the target account characteristics from the first candidate information set by utilizing the predicted matching degree.
Optionally, the apparatus further comprises:
a tag determination module configured to determine at least one tag for each of the types of multimedia information;
the tag deduplication module is configured to perform deduplication on the determined tags to obtain a first tag set;
a reverse arrangement table establishing module configured to determine, for each tag in the first tag set, target multimedia information including the tag in the type of multimedia information, and establish a mapping relationship between the tag and an information ID of the determined target multimedia information to obtain a reverse arrangement table of the tag for the type of multimedia information;
and the inverted index establishing module is configured to take the established inverted list of each label as an inverted index of the type of multimedia information.
Optionally, the apparatus further comprises:
a feature tag determination module configured to determine the index value as a feature tag;
the second determination submodule includes:
a first determination unit configured to determine a target tag matching the feature tag in the first tag set;
a second determination unit configured to determine the information ID recorded in the posting list of each target tag and perform deduplication on the determined information ID;
and the third determining unit is configured to take the multimedia information corresponding to the information ID after the duplication removal as the first candidate information set.
Optionally, the second determining sub-module includes:
the sorting unit is configured to sort the multimedia information in the first candidate information set according to the sequence of the corresponding matching degrees from large to small to obtain an information sequence;
and determining multimedia information matched with the target account characteristics by using the information sequence.
Optionally, the second determining sub-module includes:
an adjusting unit, configured to adjust the order of each multimedia information in the information sequence by using the heat of each multimedia information in the information sequence, to obtain an adjusted information sequence, where the heat of any multimedia information is used to represent: the frequency of access to the multimedia information within a predetermined time period;
a fourth determining unit configured to use the multimedia information in the adjusted information sequence as the multimedia information matched with the target account feature.
Optionally, the screening module includes:
the first selection submodule is configured to select the same number of multimedia information from each type of the retrieved at least two types of multimedia information, and the selected multimedia information is used as a target information recommendation result; alternatively, the first and second electrodes may be,
the second selection submodule is configured to determine the selection number corresponding to the type of the multimedia information according to the weight of each type of the retrieved at least two types of multimedia information and the total number of preset target information recommendation results, select the selected number of multimedia information from the retrieved type of the multimedia information, and take the selected multimedia information as a target information recommendation result; alternatively, the first and second electrodes may be,
and the third selection submodule is configured to randomly select multimedia information from each type of the retrieved at least two types of multimedia information, and take the selected multimedia information as a target information recommendation result.
Optionally, the multiple types of multimedia information include: at least two of the predetermined types of multimedia information, the predetermined types of multimedia information including at least: video, pictures, text, and animation.
According to a third aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the information recommendation method of the first aspect.
According to a fourth aspect of embodiments of the present disclosure, there is provided a storage medium, wherein instructions that, when executed by a processor of an electronic device, enable the electronic device to perform the information recommendation method according to the first aspect.
According to a fifth aspect of embodiments of the present disclosure, there is provided a computer program product, wherein instructions of the computer program product, when executed by a processor of an electronic device, enable the electronic device to perform the information recommendation method according to the first aspect.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects: when information is recommended, firstly, the target account characteristics of a target account to be received with an information recommendation result are determined; then, retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information; further screening out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information; and finally recommending the target information recommendation result to the target account. According to the method and the device, the various types of multimedia information retrieved from the preset information base are screened to obtain the target information recommendation result containing at least two types of multimedia information, and the target information recommendation result is recommended to the target account, so that the problem that the recommendation information type is single in the related technology can be solved. In addition, the target account characteristics are retrieved from the preset information base, so that the information recommended to the target account is the information according with the target account characteristics, the probability that the target account browses the target information recommendation result is higher, and the information provided by the information provider can obtain more attention or click rate.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
Fig. 1 is a flow chart illustrating an information recommendation method according to an example embodiment.
Fig. 2 is another flow chart illustrating a method of information recommendation according to an example embodiment.
FIG. 3 is another flow chart illustrating a method of information recommendation, according to an example embodiment.
FIG. 4 is another flow chart illustrating a method of information recommendation, according to an example embodiment.
FIG. 5 is another flow chart illustrating a method of information recommendation, according to an example embodiment.
Fig. 6 is a block diagram illustrating an information recommendation apparatus according to an example embodiment.
FIG. 7 is a block diagram illustrating an electronic device in accordance with an example embodiment.
FIG. 8 is a block diagram illustrating an apparatus for information recommendation, according to an example embodiment.
FIG. 9 is a block diagram illustrating another apparatus for information recommendation, according to an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings 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 disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
With the development of computer technology, more and more websites or application software can provide information recommendation services for accounts. By providing personalized information recommendation service for the account, on one hand, the account of a website or application software can obtain better use experience, and on the other hand, the information provided by an information provider can obtain more attention or click rate and the like.
Currently, when information is recommended for an account, only a certain type of information is recommended for the account. For example, only videos are recommended to the account, or only pictures are recommended to the account. Although information recommendation is realized, information recommended for the account is not abundant due to the fact that the type of the recommended information is single.
In order to solve the problems of the related art, embodiments of the present disclosure provide an information recommendation method and apparatus, an electronic device, and a storage medium.
First, an information recommendation method provided by the embodiment of the present disclosure is described below.
It should be noted that an execution subject of the information recommendation method provided by the embodiment of the present disclosure may be an electronic device. In a specific application, the electronic device may be a terminal device or a server. Illustratively, the terminal device may be: smart phones, tablet computers, desktop computers, and the like.
Fig. 1 is a flowchart illustrating an information recommendation method according to an exemplary embodiment, and as shown in fig. 1, an information recommendation method may include the steps of:
s101, determining target account characteristics of the target account.
When the electronic device needs to recommend information to a certain account, the account can be used as a target account. Namely, the target account is the account to receive the information recommendation result. In order to recommend information to the target account, the target account characteristics of the target account may be determined, and information recommendation may be further performed based on the target account characteristics.
The target account characteristics may be characteristics determined based on account information and/or behavior information of the target account. The account information for the target account may be: the gender, age, geographical area, etc. of the user registered with the target account, and the behavior information of the target account may be: the number and type of times the target account browses information within a predetermined period of time, the number and type of times the information is praised, etc. For example, the account information of the target account is: the age of the user who registers the target account is 22 years, and the geographic area where the user who registers the target account is located is beijing, then determining the target account characteristics of the target account based on the account information includes: 90, Beijing. For another example, the behavior information of the target account is: and if the target account browses sports videos and news videos in the past week, determining the target account characteristics of the target account as follows based on the behavior information: sports, news.
In practical applications, the number of target account features of the target account may be one or more.
And S102, retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored.
The preset information base may be provided by a recommender of the information. The predetermined information base is provided, for example, by a video website. As another example, the predetermined library of information is provided by a shopping website. The plurality of types of information stored in the preset information base may include a plurality of types of multimedia information. When information is recommended to the target account, various types of multimedia information matched with the characteristics of the target account can be retrieved from the preset information base, so that a target information recommendation result is further determined based on the retrieved various types of multimedia information and recommended to the target account. In practical applications, the preset information base may be updated periodically, and the updating period may be one day or one week.
Wherein, the multi-type multimedia information comprises: at least two of the predetermined types of multimedia information, the predetermined types of multimedia information including at least: video, pictures, text, and animation. For example, the preset information base stores pictures and videos, and for example, the preset information base stores videos and texts.
It can be understood that there are various specific implementation manners for retrieving various types of multimedia information matched with the target account characteristics from the preset information base storing various types of information. For clarity of the scheme and clarity of layout, the following description will be given by way of example of a process of retrieving various types of multimedia information matching the target account characteristics from a preset information base storing various types of information.
S103, target information recommendation results containing at least two types of multimedia information are screened from the retrieved various types of multimedia information.
After each type of multimedia information is retrieved, because the data volume of each type of retrieved multimedia information is usually large, in order to ensure effective reading of the target account, target information recommendation results containing at least two types of multimedia information can be further screened from each type of retrieved multimedia information.
Optionally, the method for screening out the target information recommendation result including at least two types of multimedia information from the retrieved multimedia information may include any one of the following three methods:
the first mode is as follows: and selecting the same number of multimedia information from each type of multimedia information in the retrieved at least two types of multimedia information, and taking the selected multimedia information as a target information recommendation result.
For example, if the retrieved multimedia information includes videos and pictures, 10 videos are selected from the videos, and 10 pictures are selected from the pictures as the target information recommendation result.
The second mode is as follows: determining the selection number corresponding to the type of multimedia information according to the weight of each type of the retrieved at least two types of multimedia information and the total number of the preset target information recommendation results, selecting the selected number of multimedia information from the retrieved type of multimedia information, and taking the selected multimedia information as the target information recommendation result.
For example, the retrieved multimedia information includes videos and pictures, the total number of the preset target information recommendation results is 20, the weight of the videos is 70%, and the weight of the pictures is 30%, then 14 videos are selected from the videos, and 6 pictures are selected from the pictures as the target information recommendation results.
The third mode is as follows: and randomly selecting multimedia information from each type of multimedia information in the at least two types of retrieved multimedia information, and taking the selected multimedia information as a target information recommendation result.
Illustratively, in one mode, a certain amount of multimedia information may be randomly selected from each type of the retrieved at least two types of multimedia information by using a monte carlo method, and the selected multimedia information may be used as the target information recommendation result.
For example, if the retrieved multimedia information includes videos and pictures, 12 videos are randomly selected from the videos by using the monte carlo method, and 8 pictures are randomly selected from the pictures as the target information recommendation result.
In the filtering, at least two types of multimedia information may be filtered from the retrieved types of multimedia information, and the filtering is not limited to: each type of multimedia information must be screened out.
And S104, recommending the target information recommendation result to a target account.
After the target information recommendation result is determined, the target information recommendation result can be recommended to the target account. For example, if the target information recommendation result is a video recommendation result, the video recommendation result may be displayed in a display interface of a terminal used by the target account, so that the target account can conveniently browse videos in the video recommendation result.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects: when information is recommended, firstly, the target account characteristics of a target account to be received with an information recommendation result are determined; then, retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information; further screening out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information; and finally recommending the target information recommendation result to the target account. According to the method and the device, the various types of multimedia information retrieved from the preset information base are screened to obtain the target information recommendation result containing at least two types of multimedia information, and the target information recommendation result is recommended to the target account, so that the problem that the recommendation information type is single in the related technology can be solved. In addition, the target account characteristics are retrieved from the preset information base, so that the information recommended to the target account is the information according with the target account characteristics, the probability that the target account browses the target information recommendation result is higher, and the information provided by the information provider can obtain more attention or click rate.
Referring to the process of retrieving multimedia information of each type matching the target account feature from the preset information base storing information of a plurality of types in S102, as shown in fig. 2, the process may include the following steps:
s201, determining the target account characteristics as an index value;
s202, aiming at each type of multimedia information in a preset information base, determining a first candidate information set from the type of multimedia information by using an index value and a pre-established reverse index of the type of multimedia information; determining the information characteristics of each multimedia information in the first candidate information set; predicting the matching degree of the target account and each piece of multimedia information according to the characteristics of the target account and the information characteristics of each piece of multimedia information; and determining multimedia information matched with the target account characteristics from the first candidate information set by using the predicted matching degree.
Since the number of information in the preset information base is large, in order to reduce the complexity of recommendation, in this embodiment, an inverted index of each type of multimedia information in the preset information base may be established in advance for each type of multimedia information in the preset information base, so that the established inverted index is used to perform screening from the type of multimedia information.
Specifically, the target account feature may be determined as an index value, and the index value and the established inverted index are used to perform screening in a preset information base to obtain a first candidate information set, so that the multimedia information in the first candidate information set obtained by screening is information according with the target account feature.
It is understood that there are many specific implementations of the established inverted index for each type of multimedia information. For clarity of the scheme and clarity of layout, the process of creating an inverted index for each type of multimedia information is described below as an example. In addition, the specific implementation manner of the first candidate information set is determined from the type of multimedia information by using the index value and the pre-established reverse index of the type of multimedia information, and the introduction is performed after the introduction of the process of establishing the reverse index of each type of multimedia information in relation to the construction process of the reverse index.
Further, after the first candidate information set is determined, in order to recommend information more accurately, the information characteristics of each multimedia information in the first candidate information set may be determined, and the multimedia information matching the target account characteristics may be determined from the first candidate information set by further filtering in the first candidate information set based on the determined information characteristics. The information characteristics of any multimedia information in the first candidate information set can be characterized by the following parameters: the category of the multimedia information, the subject of the multimedia information, the region of the account where the multimedia information is issued, and the like. For example, the categories of a certain multimedia information are: recording sheets; the theme of the multimedia information is as follows: traveling; the region of the account for issuing the multimedia information is as follows: shanghai, etc. The electronic device may analyze each multimedia message in the first candidate message set to obtain the message characteristic of the multimedia message. For example, the electronic device may analyze each multimedia message using a machine learning model to obtain information characteristics of the multimedia message. The electronic device may also read the information characteristic of each multimedia information in the manually labeled first candidate information set. The manual labeling may be performed by a multimedia information recommender, such as a manager of a website, or may be performed by a multimedia information provider, such as an account for uploading the multimedia information.
Specifically, the matching degree between the target account and each piece of multimedia information may be predicted according to the target account feature and the information feature of each piece of multimedia information in the first candidate information set, and then the multimedia information matching the target account feature may be determined from the first candidate information set by using the predicted matching degree.
For example, the matching degree of the target account and each multimedia information can be used for characterizing: the interest degree value of the target account for each multimedia information.
In one implementation, the prediction model may be trained in advance by using sample account characteristics of the sample account, sample information characteristics of the sample information, and a matching degree of the sample account and the sample information. Furthermore, after the prediction model is trained, the target account characteristics and the information characteristics of each multimedia information can be input into the trained prediction model, so that the matching degree between the target account and each multimedia information in the first candidate information set is obtained. Further, multimedia information matching the target account feature is determined from the first candidate information set using the matching degree.
For clarity of the scheme and clarity of layout, the process of training the predictive model is described below.
Further, after the matching degree between the target account and each multimedia information in the first candidate information set is obtained, the information recommendation result may be determined from the first candidate information set by using the obtained matching degree. There are various specific implementation manners for determining the multimedia information matched with the target account feature from the first candidate information set by using the obtained matching degree. For clarity and layout, the following description will be made by way of example of a process for determining multimedia information matching the target account feature from the first candidate information set by using the matching degree.
The following describes a process for creating an inverted index for each type of multimedia information, which may include the following steps, as shown in fig. 3:
s301, at least one label of each multimedia message in the type of multimedia message is determined.
Each of the multimedia information of the type has at least one tag. For example, the labels of the multimedia information 1 are: travel, Beijing; the labels of the multimedia information 2 are: gourmet food, shanghai, etc. The label of each multimedia message reflects the main characteristics of the multimedia message. Also, for a tag, there may be one or more multimedia messages having the tag.
When constructing the reverse index of the type of multimedia information, the type of multimedia information can be regarded as an information set, and the tag of the type of multimedia information is regarded as a tag set, so that the reverse index to be established is: a set of mappings from tags to information. It will be appreciated that the tags in the labelset are not repeated, i.e., the tags in the labelset are distinct.
To determine the set of tags described above, at least one tag of the multimedia information may be determined for each multimedia information of the type of multimedia information, and the set of tags may be derived further based on the determined tags. For example, there are 5 pieces of multimedia information in the type of multimedia information, which are multimedia information 1 to multimedia information 5. The labels of the multimedia information 1 are: label a, label b; the labels of the multimedia information 2 are: a label a; the labels of the multimedia information 3 are: label c, label d; the labels of the multimedia information 4 are: a label a; the labels of the multimedia information 5 are: label b, label d. Then, in this example, 5 of the multimedia information of this type have a total of 4 tags, 8 tags.
S302, the determined labels are subjected to de-duplication to obtain a first label set.
As described above, in the tab set used for establishing the inverted index, each tab is different, and therefore, the determined tab needs to be deduplicated to obtain the first tab set.
Continuing with the above example, after the determined tag is deduplicated, a first tag set is obtained, which includes: tag a, tag b, tag c, and tag d.
S303, for each tag in the first tag set, determining target multimedia information including the tag in the type of multimedia information, and establishing a mapping relationship between the tag and an information ID of the determined target multimedia information to obtain a reverse arrangement table of the tag for the type of multimedia information.
After the first tag set is obtained, for each tag in the first tag set, the target multimedia information including the tag is determined in the type of multimedia information, a mapping relationship between the tag and the information ID of the determined target multimedia information is established, and a posting list of the tag for the type of multimedia information is obtained. Continuing with the above-described example,
the inverted list established for tag a is: label a: multimedia information 1, multimedia information 2, multimedia information 4;
the inverted list established for label b is: and b, label b: multimedia information 1, multimedia information 5;
the inverted list built for label c is: and c, labeling: multimedia information 3;
the inverted list built for label d is: and (3) a label d: multimedia information 3, multimedia information 5.
And S304, taking the established inverted list of each label as an inverted index of the type of multimedia information.
That is, the inverted index of the type of multimedia information includes a set of mapping relationships, specifically: and mapping relation from each label in the first label set to the information ID of the target multimedia information containing the label. Based on the inverted index, the following can be directly indexed through the tags: the information ID of the multimedia information of the tag is included, so that it is possible to easily screen information from the type of multimedia information.
It can be understood that, since information needs to be recommended for the target account, the target account characteristic may be determined as an index value, and for each type of multimedia information in the preset information base, a first candidate information set may be determined from the type of multimedia information by using the index value and a pre-established inverted index of the type of multimedia information. In particular, the method comprises the following steps of,
after the step of determining the target account feature as the index value in S201, the method may further include: the index value is determined as a feature tag.
For example, the index value is: sports, beijing, then generates the corresponding feature labels as: sports, Beijing. That is, the index value is expressed in the form of a tag, and the feature tag is used as an index value, so that the index value is further used to index in the inverted index of the preset information base.
Accordingly, as shown in fig. 4, determining the first candidate information set from the type of multimedia information by using the index value and the pre-established inverted index of the type of multimedia information in S202 may include the following steps:
s401, determining a target label matched with the characteristic label in the first label set.
Wherein, matching with the feature tag may be: the similarity of the characteristic label and the characteristic label is the same as or greater than a set value.
Since the inverted index of this type of multimedia information contains: and the mapping relation of each label in the first label set to the information ID of the target multimedia information containing the label, so that the target label matched with the characteristic label can be determined in the first label set.
Taking the above example of constructing the inverted index of the type of multimedia information as an example, the first tag set includes: tag a, tag b, tag c, and tag d. The generated feature labels are: label a and label e. Wherein, the label e and the label b in the first label set meet the predetermined similar condition, and thus, the target label matching with the feature label in the first label set is determined as follows: label a and label b.
The above set value may be set as desired, for example, the set value is 90%. For example, the similarity between two words, namely beijing and capital is 95%, and then beijing and capital are labels with the similarity larger than a set value.
S402, determining the information ID recorded in the inverted arrangement table of each target label, and performing deduplication on the determined information ID.
After the target tags are determined, the information ID recorded in the inverted list of each target tag may be determined for each target tag, that is, the ID of the multimedia information of the type including the target tag is determined.
Continuing with the above example, the target tag is: label a and label b. Determining the information ID recorded in the inverted list of the label a as: multimedia information 1, multimedia information 2 and multimedia information 4. Determining the information ID recorded in the inverted list of the label b as: multimedia information 1, multimedia information 5. Further, the determined information ID is deduplicated, and the obtained information ID is: multimedia information 1, multimedia information 2, multimedia information 4, multimedia information 5.
And S403, taking the multimedia information corresponding to the information ID after duplication elimination as a first candidate information set.
After the determined information ID is deduplicated, the multimedia information corresponding to the deduplicated information ID may be directly used as the first candidate information set. This process of determining the first set of candidate information from the type of multimedia information may also be referred to as a recall process, i.e. a process of preliminarily screening some multimedia information from the type of multimedia information.
The specific process is that for each type of multimedia information in the preset information base, the target account characteristics of the target account and the established inverted index of the preset information base are utilized to screen the type of multimedia information to obtain the first candidate information set. Therefore, because the target account features are indexed from the inverted index, that is, the characteristics of the target account are considered to be screened from the multimedia information of the type, the obtained multimedia information in the first candidate information set is the multimedia information which accords with the characteristics of the target account, and the possibility that the target account browses the multimedia information is high.
The following describes a process for training a prediction model, which may include the following steps, as shown in fig. 5:
s501, obtaining sample account characteristics of the sample account and obtaining sample information characteristics of the sample information.
And the sample information is the information browsed by the sample account in the information browsing record.
For example, for a recommender of information, such as a video website, the sample account may be an account for browsing videos in the video website within a predetermined time period, and the sample information may be videos browsed by the sample account in a video browsing record within the predetermined time period. The predetermined period of time may be determined as desired, such as the past day or the past week, etc. In practical applications, the sample information may be various types of information, and is not limited to video, pictures, text, links, and the like.
Sample account characteristics for a sample account may be characteristics determined based on account information and/or behavioral information for the sample account, among other things. The sample information characteristics of the sample information may characterize: the type of the sample information, the subject of the sample information, the region of the account where the sample information is issued, etc. The electronic device can analyze each sample information to obtain the sample information characteristics of the sample information. For example, the electronic device may analyze each sample information using a machine learning model to obtain sample information characteristics of the sample information. The electronic device can also read the manually marked sample information features of each sample information.
In a specific application, the sample information characteristics may further include: CTR (Click Through Rate), LTR (Like Through Rate), CMTR (Comment Through Rate), and the Like of the sample information. Wherein CTR represents: the ratio of the number of clicks on the sample information to the number of presentations of the sample information, LTR, represents: the ratio of the number of praise to the number of show of the sample information, CMTR represents: the ratio of the number of reviews for the sample information to the number of presentations of the sample information.
And the obtained sample account characteristics of the sample account and the sample information characteristics of the sample information are used as training samples for training the prediction model. The number of training samples may be determined according to actual needs or experience.
And S502, constructing a cross feature by using the obtained sample account feature and the sample information feature.
Specifically, during training, the obtained sample account features and sample information features can be used for constructing cross features, and the constructed cross features are input into a prediction model.
In this embodiment, both the sample account characteristics and the sample information characteristics may be converted into a vector form. Illustratively, the vector form of the sample account features is: y1 ═ 0101], the vector form of the sample information features is: y2 ═ 1110, then the vector form of the cross feature is to combine the two vectors as: y is [01011110 ]. That is, y1 and y2 are concatenated into a vector form of the cross feature.
The sample account characteristics and the sample information characteristics are converted into a vector form, and a method such as word2vec can be adopted, which is not limited by the present disclosure.
And S503, inputting the cross features and the matching degree of the sample account and the sample information into an initial prediction model for training to obtain a trained prediction model.
After the cross features are constructed, the cross features and the matching degree of the sample account and the sample information can be input into an initial prediction model for training. The matching degree between the sample account and the sample information can be understood as follows: true values referenced when training the predictive model. For example, the matching degree of the sample account and the sample information can be used for characterizing: the degree of interest value of the sample account for the sample information. The goal of training the predictive model is then to make the predictive model learn: the combination of the sample account and the sample information and the mapping relation between the matching degree of the sample account and the sample information. For example, for the combination of sample account 1 and sample information 1, the predictive model learns: the matching degree of the combination is 90%, and the matching degree of the sample account 1 and the sample information 1 is 90%. The specific representation form of the combination of the sample account and the sample information is the cross feature.
Specifically, in the training process, multiple iterations may be performed to train parameters of the prediction model, and the prediction model training is completed until the prediction model reaches the set convergence condition.
Alternatively, the prediction model may be a predictserver model, and the specific model type may be any neural network model existing in the related art, which is not limited herein. In this model, CTR, LTR, CMTR, and the like of the sample information may be trained as sample information features.
The above is the process of training the predictive model. And obtaining the matching degree of the target account to each multimedia information in the first candidate information set by using the trained prediction model. The matching degree can be a real number between 0 and 1, and can be specifically expressed in a percentage form.
The following describes an example of determining multimedia information matching the target account feature from the first candidate information set by using the predicted matching degree in S202.
For example, in the first manner, the multimedia information corresponding to the matching degree greater than or equal to the preset threshold in the first candidate information set may be used as the multimedia information matched with the target account feature.
It is assumed that the first candidate information set includes: the matching degrees of the multimedia information 1, the multimedia information 2, the multimedia information 4 and the multimedia information 5 to the target account are 85%, 92%, 90% and 95% respectively. If the preset threshold is 90%, then multimedia information 2, multimedia information 4, and multimedia information 5 are taken as the multimedia information matching the target account feature.
In an exemplary, second manner, the process of determining the multimedia information matching the target account feature from the first candidate information set by the predicted matching degree in S202 may include the following steps:
and step 1, sequencing all multimedia information in the first candidate information set according to the sequence of the corresponding matching degrees from large to small to obtain an information sequence.
Following the example in the first implementation manner, after sorting, the obtained information sequence is: multimedia information 5, information 2, multimedia information 4, multimedia information 1.
And 2, determining multimedia information matched with the target account characteristics by using the information sequence.
After obtaining the sequence of information, information matching the target account characteristics may be determined by a variety of implementations. Illustratively, three implementations are given below:
implementation mode 1: and taking the information sequence as multimedia information matched with the target account characteristics.
For example, the information sequence: and the multimedia information 5, the multimedia information 2, the multimedia information 4 and the multimedia information 1 are used as multimedia information matched with the target account characteristics. Illustratively, if the matching degree of the target account with each multimedia information in the first candidate information set is used for characterizing: the interest degree value of each multimedia information in the first candidate information set by the target account is sorted according to the corresponding interest degree value of each multimedia information matched with the characteristics of the target account, so that the target account can conveniently browse the multimedia information from the information recommendation result according to the interest degree of the target account when recommending to the target account.
Implementation mode 2: and taking the multimedia information with the preset number in the information sequence as the information matched with the target account characteristics.
For example, if the preset number is 3, then the multimedia information 5, the multimedia information 2, and the multimedia information 4 are used as the multimedia information matching the target account feature. And the multimedia information matched with the target account characteristics can be in the sequence: multimedia information 5, multimedia information 2, multimedia information 4.
Implementation mode 3: adjusting the sequence of each multimedia information in the information sequence by using the heat of each multimedia information in the information sequence to obtain an adjusted information sequence; and taking the multimedia information in the adjusted information sequence as the multimedia information matched with the target account characteristics.
In implementation 3, the heat of any multimedia information in the information sequence is used to represent: a frequency at which the multimedia information is accessed within a predetermined time period. Illustratively, it is reasonable that the accessed frequency can be a private letter reply rate of the multimedia information, or can be calculated by the number of praise of the multimedia information or the number of comments of the multimedia information.
The private message reply rate of the multimedia information is as follows: the probability that the account issuing the multimedia information replies to the received private letter. If the private letter reply rate of the multimedia information is high, the activity of the account issuing the multimedia information is higher, and the user prefers to perform social interaction with other accounts. Then, for the multimedia information with high private message reply rate in the information sequence, the score of the multimedia information can be increased; for multimedia information with low private message reply rate in the information sequence, the score of the multimedia information can be reduced; for multimedia information with a private message reply rate between high and low in the information sequence, the score of the multimedia information can not be changed. Specifically, the matching degree corresponding to each multimedia information in the information sequence may be converted into a percentile score. For example, the scores of multimedia information 5, multimedia information 2, multimedia information 4 and multimedia information 1 in the information sequence are 95,92,90 and 85 respectively. Wherein, the private letter reply rate of the multimedia information 2 and the multimedia information 1 is high, and 3 points are added to the two multimedia information; if the private message reply rate of the multimedia message 5 is between high and low, the score of the multimedia message is not changed; if the private message reply rate of the multimedia message 4 is low, the multimedia message is reduced by 3 points. Then, the order of the multimedia information in the information sequence after the score is adjusted is: multimedia information 5, multimedia information 2, multimedia information 1, multimedia information 4. That is, the information recommendation result is the sequence: multimedia information 5, multimedia information 2, multimedia information 1, multimedia information 4.
Similarly, for multimedia information with more, less or between more and less praise/comment numbers in the information sequence, the score can be correspondingly added, subtracted or not changed, so that the sequence of each multimedia information in the information sequence is adjusted to obtain an adjusted information sequence, and the multimedia information in the adjusted information sequence is used as the multimedia information matched with the target account characteristics.
The heat of each multimedia message in the sequence of messages reflects the level of interest in the multimedia message by most accounts. If the popularity of one multimedia message is high, it indicates that most accounts are interested in the multimedia message, and if the popularity of one multimedia message is low, it indicates that most accounts are not interested in the multimedia message. Wherein, the higher heat of a multimedia message may be: the multimedia information has high private letter reply rate and more praise or comment number, and the lower popularity of one multimedia information can be: the multimedia information has low private letter reply rate and few praise numbers or comment numbers. Therefore, the sequence of each multimedia information in the information sequence is adjusted to serve as an information recommendation result by utilizing the heat of each multimedia information in the information sequence, and the information recommendation result not only accords with the interest of a target account, but also accords with the interest of most accounts, so that the information recommendation effect is better.
The above three implementation manners of determining the multimedia information matching with the target account feature by using the information sequence are only examples, and in practical applications, there may be a plurality of implementation manners of determining the multimedia information matching with the target account feature by using the information sequence.
Fig. 6 is a block diagram illustrating an information recommendation apparatus according to an example embodiment. Referring to fig. 6, the apparatus may include:
the determining module 601 is configured to determine a target account characteristic of a target account, where the target account is an account to receive an information recommendation result;
a retrieving module 602 configured to retrieve, from a preset information base storing multiple types of information, multiple types of multimedia information matched with the target account characteristics, where the multiple types of information include multiple types of multimedia information;
a screening module 603 configured to screen out target information recommendation results including at least two types of multimedia information from the retrieved various types of multimedia information;
a recommending module 604 configured to recommend the target information recommending result to the target account.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects: when information is recommended, firstly, the target account characteristics of a target account to be received with an information recommendation result are determined; then, retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information; further screening out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information; and finally recommending the target information recommendation result to the target account. According to the method and the device, the various types of multimedia information retrieved from the preset information base are screened to obtain the target information recommendation result containing at least two types of multimedia information, and the target information recommendation result is recommended to the target account, so that the problem that the recommendation information type is single in the related technology can be solved. In addition, the target account characteristics are retrieved from the preset information base, so that the information recommended to the target account is the information according with the target account characteristics, the probability that the target account browses the target information recommendation result is higher, and the information provided by the information provider can obtain more attention or click rate.
Optionally, the retrieving module 602 includes:
a first determination sub-module (not shown in the figure) configured to determine the target account feature as an index value;
a second determining sub-module (not shown in the figure) configured to determine, for each type of multimedia information in a preset information base, a first candidate information set from the type of multimedia information by using the index value and a pre-established inverted index of the type of multimedia information; determining information characteristics of each multimedia information in the first candidate information set; predicting the matching degree of the target account and each piece of multimedia information according to the target account characteristics and the information characteristics of each piece of multimedia information; and determining multimedia information matched with the target account characteristics from the first candidate information set by utilizing the predicted matching degree.
Optionally, the apparatus further comprises:
a tag determination module (not shown in the figures) configured to determine at least one tag for each of the types of multimedia information;
a tag deduplication module (not shown in the figure) configured to deduplicate the determined tags, resulting in a first tag set;
a reverse arrangement table establishing module (not shown in the figure), configured to determine, for each tag in the first tag set, target multimedia information including the tag in the type of multimedia information, and establish a mapping relationship between the tag and an information ID of the determined target multimedia information to obtain a reverse arrangement table of the tag for the type of multimedia information;
and an inverted index establishing module (not shown in the figure) configured to use the established inverted list of each label as an inverted index of the type of multimedia information.
Optionally, the apparatus further comprises:
a feature tag determination module (not shown in the figures) configured to determine the index value as a feature tag;
the second determination submodule includes:
a first determination unit configured to determine a target tag matching the feature tag in the first tag set;
a second determination unit configured to determine the information ID recorded in the posting list of each target tag and perform deduplication on the determined information ID;
and the third determining unit is configured to take the multimedia information corresponding to the information ID after the duplication removal as the first candidate information set.
Optionally, the second determining sub-module includes:
the sorting unit is configured to sort the multimedia information in the first candidate information set according to the sequence of the corresponding matching degrees from large to small to obtain an information sequence;
and determining multimedia information matched with the target account characteristics by using the information sequence.
Optionally, the second determining sub-module includes:
an adjusting unit, configured to adjust the order of each multimedia information in the information sequence by using the heat of each multimedia information in the information sequence, to obtain an adjusted information sequence, where the heat of any multimedia information is used to represent: the frequency of access to the multimedia information within a predetermined time period;
a fourth determining unit configured to use the multimedia information in the adjusted information sequence as the multimedia information matched with the target account feature.
Optionally, the screening module 603 includes:
a first selecting sub-module (not shown in the figure) configured to select the same number of multimedia information from each type of the retrieved at least two types of multimedia information, and take the selected multimedia information as a target information recommendation result; alternatively, the first and second electrodes may be,
a second selecting sub-module (not shown in the figure), configured to determine a selecting number corresponding to each type of the retrieved at least two types of multimedia information according to the weight of the type of multimedia information and the total number of the preset target information recommendation results, select the selected number of multimedia information from the retrieved type of multimedia information, and use the selected multimedia information as the target information recommendation result; alternatively, the first and second electrodes may be,
and a third selecting sub-module (not shown in the figure) configured to randomly select multimedia information from each of the retrieved at least two types of multimedia information, and use the selected multimedia information as a target information recommendation result.
Optionally, the multiple types of multimedia information include: at least two of the predetermined types of multimedia information, the predetermined types of multimedia information including at least: video, pictures, text, and animation.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
In addition, corresponding to the information recommendation method provided in the foregoing embodiment, an embodiment of the present disclosure further provides an electronic device, and as shown in fig. 7, the electronic device may include:
a processor 710;
a memory 720 for storing the processor-executable instructions;
wherein the processor 710 is configured to execute the instructions to implement the information recommendation method provided by the above embodiments to obtain the same technical effect.
It is understood that the electronic device may be a server or a terminal device, and in particular applications, the terminal device may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.
The Memory 720 may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory 720 may also be at least one memory device located remotely from the aforementioned processor.
The Processor 710 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
Fig. 8 is a block diagram illustrating an apparatus 800 for information recommendation, according to an example embodiment. For example, the apparatus 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.
Referring to fig. 8, the apparatus 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.
The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing components 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.
The memory 804 is configured to store various types of data to support operations at the apparatus 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The multimedia component 808 includes a screen that provides an output interface between the device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.
The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.
The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed status of the device 800, the relative positioning of components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in the position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration/deceleration of the device 800, and a change in the temperature of the device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 816 is configured to facilitate communications between the apparatus 800 and other devices in a wired or wireless manner. The apparatus 800 may access a wireless network based on a communication standard, such as WiFi, an operator network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.
In an exemplary embodiment, a storage medium comprising instructions, such as the memory 804 comprising instructions, executable by the processor 820 of the apparatus 800 to perform the method described above is also provided. Alternatively, the storage medium may be a non-transitory computer readable storage medium, which may be, for example, a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
Fig. 9 is a block diagram illustrating an apparatus 900 for information recommendation, according to an example embodiment. For example, the apparatus 900 may be provided as a server. Referring to fig. 9, the apparatus 900 includes a processing component 922, which further includes one or more processors, and memory resources, represented by memory 932, for storing instructions, such as applications, that are executable by the processing component 922. The application programs stored in memory 932 may include one or more modules that each correspond to a set of instructions. Further, the processing component 922 is configured to execute instructions to perform any of the information recommendation methods described above.
The device 900 may also include a power component 926 configured to perform power management of the device 900, a wired or wireless network interface 950 configured to connect the device 900 to a network, and an input output (I/O) interface 958. The apparatus 900 may operate based on an operating system, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, stored in the memory 932.
In addition, the embodiment of the present disclosure also provides a storage medium, and when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the information recommendation method provided in the foregoing embodiment to obtain the same technical effect.
In addition, the embodiment of the present disclosure further provides a computer program product, and when instructions in the computer program product are executed by a processor of an electronic device, the electronic device is enabled to execute the information recommendation method provided in the foregoing embodiment, so as to obtain the same technical effect.
It should be noted that the account information, the user information, or the device information related to the present application are all acquired after being authorized by the user and then are subjected to subsequent processing.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. An information recommendation method, comprising:
determining target account characteristics of a target account, wherein the target account is an account to receive information recommendation results;
retrieving various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information;
screening out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information;
and recommending the target information recommendation result to the target account.
2. The information recommendation method according to claim 1, wherein the step of retrieving each type of multimedia information matching the target account feature from a preset information base storing a plurality of types of information comprises:
determining the target account characteristic as an index value;
aiming at each type of multimedia information in a preset information base, determining a first candidate information set from the type of multimedia information by using the index value and a pre-established reverse index of the type of multimedia information; determining information characteristics of each multimedia information in the first candidate information set; predicting the matching degree of the target account and each piece of multimedia information according to the target account characteristics and the information characteristics of each piece of multimedia information; and determining multimedia information matched with the target account characteristics from the first candidate information set by utilizing the predicted matching degree.
3. The information recommendation method according to claim 2, wherein the process of establishing the inverted index for each type of multimedia information comprises:
determining at least one tag for each of the types of multimedia information;
removing the duplicate of the determined label to obtain a first label set;
for each tag in the first tag set, determining target multimedia information containing the tag in the type of multimedia information, and establishing a mapping relation between the tag and an information ID of the determined target multimedia information to obtain a reverse arrangement table of the tag for the type of multimedia information;
and taking the established inverted list of each label as an inverted index of the type of multimedia information.
4. The information recommendation method according to claim 3, wherein the step of determining the target account feature as an index value is followed by further comprising:
determining the index value as a feature tag;
the step of determining a first candidate information set from the type of multimedia information by using the index value and a pre-established reverse index of the type of multimedia information includes:
determining a target label in the first label set that matches the feature label;
determining the information ID recorded in the inverted arrangement table of each target label, and removing the duplication of the determined information ID;
and taking the multimedia information corresponding to the information ID after the duplication removal as a first candidate information set.
5. The information recommendation method according to claim 2, wherein the step of determining multimedia information matching the target account feature from the first candidate information set by using the predicted matching degree comprises:
sequencing all multimedia information in the first candidate information set according to the sequence of the corresponding matching degrees from large to small to obtain an information sequence;
and determining multimedia information matched with the target account characteristics by using the information sequence.
6. The information recommendation method according to claim 5, wherein the step of determining multimedia information matching the target account feature using the information sequence comprises:
and adjusting the sequence of each multimedia information in the information sequence by using the heat of each multimedia information in the information sequence to obtain an adjusted information sequence, wherein the heat of any multimedia information is used for expressing: the frequency of access to the multimedia information within a predetermined time period;
and taking the multimedia information in the adjusted information sequence as the multimedia information matched with the target account characteristics.
7. The information recommendation method according to claim 1, wherein the plurality of types of multimedia information include: at least two of the predetermined types of multimedia information, the predetermined types of multimedia information including at least: video, pictures, text, and animation.
8. An information recommendation apparatus, comprising:
the determining module is configured to determine target account characteristics of a target account, wherein the target account is an account to receive the information recommendation result;
the retrieval module is configured to retrieve various types of multimedia information matched with the target account characteristics from a preset information base in which various types of information are stored, wherein the various types of information comprise various types of multimedia information;
the screening module is configured to screen out target information recommendation results containing at least two types of multimedia information from the retrieved various types of multimedia information;
a recommending module configured to recommend the target information recommending result to the target account.
9. An electronic device, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the information recommendation method of any one of claims 1 to 7.
10. A storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the information recommendation method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910838698.7A CN112445970B (en) | 2019-09-05 | 2019-09-05 | Information recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910838698.7A CN112445970B (en) | 2019-09-05 | 2019-09-05 | Information recommendation method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112445970A true CN112445970A (en) | 2021-03-05 |
CN112445970B CN112445970B (en) | 2024-03-26 |
Family
ID=74733387
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910838698.7A Active CN112445970B (en) | 2019-09-05 | 2019-09-05 | Information recommendation method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112445970B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113362109A (en) * | 2021-05-31 | 2021-09-07 | 北京达佳互联信息技术有限公司 | Recommendation method and device for media resources, electronic equipment and storage medium |
CN113407754A (en) * | 2021-07-13 | 2021-09-17 | 北京达佳互联信息技术有限公司 | Album generating method, apparatus, electronic device and storage medium |
CN113553509A (en) * | 2021-07-29 | 2021-10-26 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and storage medium |
CN113554062A (en) * | 2021-06-25 | 2021-10-26 | 北京百度网讯科技有限公司 | Training method, device and storage medium of multi-classification model |
CN115640370A (en) * | 2022-12-08 | 2023-01-24 | 深圳市智多兴投控科技有限公司 | Data analysis method and related equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015106347A (en) * | 2013-12-02 | 2015-06-08 | 株式会社Nttドコモ | Recommendation device and recommendation method |
US20150339748A1 (en) * | 2014-05-23 | 2015-11-26 | Huawei Technologies Co., Ltd. | Radio channel control method, traffic package trading and recommending methods, and related device |
CN107657004A (en) * | 2017-09-21 | 2018-02-02 | 广州华多网络科技有限公司 | Video recommendation method, system and equipment |
CN109063143A (en) * | 2018-08-07 | 2018-12-21 | 北京奇艺世纪科技有限公司 | A kind of information recommendation method and device |
WO2019128930A1 (en) * | 2017-12-29 | 2019-07-04 | 中兴通讯股份有限公司 | Operation processing method, account information processing method, device, and storage medium |
-
2019
- 2019-09-05 CN CN201910838698.7A patent/CN112445970B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015106347A (en) * | 2013-12-02 | 2015-06-08 | 株式会社Nttドコモ | Recommendation device and recommendation method |
US20150339748A1 (en) * | 2014-05-23 | 2015-11-26 | Huawei Technologies Co., Ltd. | Radio channel control method, traffic package trading and recommending methods, and related device |
CN107657004A (en) * | 2017-09-21 | 2018-02-02 | 广州华多网络科技有限公司 | Video recommendation method, system and equipment |
WO2019128930A1 (en) * | 2017-12-29 | 2019-07-04 | 中兴通讯股份有限公司 | Operation processing method, account information processing method, device, and storage medium |
CN109063143A (en) * | 2018-08-07 | 2018-12-21 | 北京奇艺世纪科技有限公司 | A kind of information recommendation method and device |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113362109A (en) * | 2021-05-31 | 2021-09-07 | 北京达佳互联信息技术有限公司 | Recommendation method and device for media resources, electronic equipment and storage medium |
CN113362109B (en) * | 2021-05-31 | 2023-09-26 | 北京达佳互联信息技术有限公司 | Recommendation method and device for media resources, electronic equipment and storage medium |
CN113554062A (en) * | 2021-06-25 | 2021-10-26 | 北京百度网讯科技有限公司 | Training method, device and storage medium of multi-classification model |
CN113554062B (en) * | 2021-06-25 | 2023-08-01 | 北京百度网讯科技有限公司 | Training method, device and storage medium for multi-classification model |
CN113407754A (en) * | 2021-07-13 | 2021-09-17 | 北京达佳互联信息技术有限公司 | Album generating method, apparatus, electronic device and storage medium |
CN113407754B (en) * | 2021-07-13 | 2023-12-15 | 北京达佳互联信息技术有限公司 | Album generating method, apparatus, electronic device and storage medium |
CN113553509A (en) * | 2021-07-29 | 2021-10-26 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and storage medium |
CN113553509B (en) * | 2021-07-29 | 2024-03-01 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and storage medium |
CN115640370A (en) * | 2022-12-08 | 2023-01-24 | 深圳市智多兴投控科技有限公司 | Data analysis method and related equipment |
Also Published As
Publication number | Publication date |
---|---|
CN112445970B (en) | 2024-03-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109800325B (en) | Video recommendation method and device and computer-readable storage medium | |
CN112445970B (en) | Information recommendation method and device, electronic equipment and storage medium | |
CN107992604B (en) | Task item distribution method and related device | |
CN110688527A (en) | Video recommendation method and device, storage medium and electronic equipment | |
CN107463643B (en) | Barrage data display method and device and storage medium | |
CN111859020A (en) | Recommendation method and device, electronic equipment and computer-readable storage medium | |
CN112508612B (en) | Method for training advertisement creative generation model and generating advertisement creative and related device | |
CN110955821A (en) | Recommendation method and device and readable medium | |
CN113254135A (en) | Interface processing method and device and electronic equipment | |
US11546663B2 (en) | Video recommendation method and apparatus | |
CN112131466A (en) | Group display method, device, system and storage medium | |
CN112148923A (en) | Search result sorting method, sorting model generation method, device and equipment | |
CN111127053B (en) | Page content recommendation method and device and electronic equipment | |
CN112685641B (en) | Information processing method and device | |
CN109542297B (en) | Method and device for providing operation guide information and electronic equipment | |
CN113343028A (en) | Method and device for training intention determination model | |
CN110650364B (en) | Video attitude tag extraction method and video-based interaction method | |
CN117453933A (en) | Multimedia data recommendation method and device, electronic equipment and storage medium | |
CN112612949A (en) | Establishment method and device of recommended data set | |
CN111752436A (en) | Recommendation method and device and recommendation device | |
CN112000266A (en) | Page display method and device, electronic equipment and storage medium | |
CN112784151A (en) | Method and related device for determining recommendation information | |
CN110730382B (en) | Video interaction method, device, terminal and storage medium | |
CN112241486A (en) | Multimedia information acquisition method and device | |
CN111898028B (en) | Entity object recommendation method, device and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |