CN116610858A - Information distribution method, device, electronic equipment and storage medium - Google Patents
Information distribution method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116610858A CN116610858A CN202310559073.3A CN202310559073A CN116610858A CN 116610858 A CN116610858 A CN 116610858A CN 202310559073 A CN202310559073 A CN 202310559073A CN 116610858 A CN116610858 A CN 116610858A
- Authority
- CN
- China
- Prior art keywords
- search
- target
- information
- multimedia content
- historical
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013475 authorization Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims description 19
- 238000012216 screening Methods 0.000 claims description 15
- 230000032683 aging Effects 0.000 claims description 10
- 230000003993 interaction Effects 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 description 6
- 230000001960 triggered effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000002452 interceptive effect Effects 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241000282376 Panthera tigris Species 0.000 description 1
- 238000010411 cooking Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 235000019633 pungent taste Nutrition 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/435—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure provides an information distribution method, an information distribution device, an electronic device and a storage medium, wherein the method comprises the following steps: determining each search term meeting target consumption requirements based on historical consumption data obtained through user authorization; the historical consumption data comprises historical search data and/or historical browsing data; acquiring target multimedia contents respectively matched with each search word from a multimedia content library, and determining target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition; and distributing the information of the target recommendation information on an information flow recommendation page.
Description
Technical Field
The disclosure relates to the technical field of internet, and in particular relates to an information distribution method, an information distribution device, electronic equipment and a storage medium.
Background
With the development of internet technology, other multimedia contents which are possibly interested can be recommended to the user while the user browses the multimedia contents, so that not only the browsing interest of the user can be improved, but also the time for the user to search for the other multimedia contents which are interested can be shortened.
However, the existing recommendation method mainly recommends other related and high-heat content according to the multimedia content currently consumed by the user, the method is easy to enable the multimedia content recommended to the user to be converged more and more, and the multimedia content finally pushed to each user is some high-heat content which is currently, so that the method does not meet the personalized requirements of some users.
Disclosure of Invention
The embodiment of the disclosure at least provides an information distribution method, an information distribution device, electronic equipment and a storage medium.
In a first aspect, an embodiment of the present disclosure provides an information distribution method, including:
determining each search term meeting target consumption requirements based on historical consumption data obtained through user authorization; the historical consumption data comprises historical search data and/or historical browsing data;
acquiring target multimedia contents respectively matched with each search word from a multimedia content library, and determining target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition;
and distributing the information of the target recommendation information on an information flow recommendation page.
In an alternative embodiment, the determining, based on the historical consumption data obtained through authorization of the user, each search term meeting the target consumption requirement includes:
determining each candidate search word matched with a preset consumption frequency threshold value based on the historical consumption data obtained through user authorization;
and screening each search word from the candidate search words based on the search type corresponding to each candidate search word.
In an optional implementation manner, the step of screening each search word from the candidate search words based on the search type corresponding to each candidate search word includes:
screening each search word matching the target search type from the candidate search words based on the search type corresponding to each candidate search word and at least one target search type; and the search demand aging corresponding to the target search type is larger than a set aging threshold.
In an alternative embodiment, the obtaining, from a multimedia content library, the target multimedia content that matches with each of the search terms respectively includes:
determining each candidate multimedia content according to the first matching degree between the tag information of each multimedia content in the multimedia content library and the search word;
And determining the target multimedia content from the candidate multimedia contents according to the interaction data corresponding to the candidate multimedia contents.
In an alternative embodiment, the obtaining, from a multimedia content library, the target multimedia content that matches with each of the search terms respectively includes:
after searching is initiated based on the search word, each search result is obtained;
and acquiring the multimedia content with the second matching degree larger than a set threshold value from the multimedia content library as the target multimedia content.
In an optional implementation manner, the information distribution of the target recommendation information on the information flow recommendation page includes:
displaying the target recommendation information at a target display position in the information flow recommendation page; the target display positions belong to a preset number of information display positions which are ranked in front.
In an optional implementation manner, the information distribution of the target recommendation information on the information flow recommendation page includes:
and based on a pre-trained ranking model, carrying out mixed ranking on the determined target recommendation information and other contents to be recommended, and carrying out information distribution on the information flow recommendation page based on a ranking result.
In an alternative embodiment, the ranking model is trained according to the following steps:
based on the search attribute characteristics corresponding to each training sample, positive samples and negative samples in the training samples and recommended weights of the positive samples are determined; the positive sample comprises multimedia content searched by a user, the negative sample comprises multimedia content not searched by the user, and the recommendation weight of the positive sample is related to the searching willingness indicated by the searching attribute characteristics of the positive sample;
the ranking model is trained based on the positive and negative samples, and a recommended weight for each positive sample.
In a second aspect, an embodiment of the present disclosure further provides an information distribution apparatus, including:
the determining unit is used for determining each search word meeting the target consumption requirement based on the historical consumption data authorized by the user; the historical consumption data comprises historical search data and/or historical browsing data;
the acquisition unit is used for acquiring target multimedia contents respectively matched with the search words from the multimedia content library and determining target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition;
And the distribution unit is used for distributing the information of the target recommendation information on the information flow recommendation page.
In a third aspect, embodiments of the present disclosure further provide an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication via the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the first aspect, or any of the possible implementations of the first aspect.
In a fourth aspect, the presently disclosed embodiments also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the first aspect, or any of the possible implementations of the first aspect.
In the embodiment of the disclosure, each search term meeting the target consumption requirement can be determined based on the historical consumption data obtained through the authorization of the user; wherein the historical consumption data comprises historical search data and/or historical browsing data. According to the embodiment, the more effective and reasonable search words meeting the target consumption requirement can be determined through the historical consumption data of the user, so that the suitability between the search words and the current user is higher. Then, target multimedia contents respectively matched with each search word can be obtained from the multimedia content library, target recommendation information corresponding to the target multimedia contents is determined, and then information distribution is carried out on the target recommendation information on the information flow recommendation page. The embodiment not only widens the information sources of the information distributed in the information flow recommendation page and improves the diversity and richness of information distribution, but also enables the information distributed in the information flow recommendation page to more accord with the consumption requirements of the current user, further enables the distributed information to be more effective and improves the consumption experience of the user.
The foregoing objects, features and advantages of the disclosure will be more readily apparent from the following detailed description of the preferred embodiments taken in conjunction with the accompanying drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for the embodiments are briefly described below, which are incorporated in and constitute a part of the specification, these drawings showing embodiments consistent with the present disclosure and together with the description serve to illustrate the technical solutions of the present disclosure. It is to be understood that the following drawings illustrate only certain embodiments of the present disclosure and are therefore not to be considered limiting of its scope, for the person of ordinary skill in the art may admit to other equally relevant drawings without inventive effort.
FIG. 1 shows a flow chart of an information distribution method provided by an embodiment of the present disclosure;
FIG. 2 is a flow chart of a method for obtaining target multimedia content from a multimedia content library that matches each search term separately, provided by an embodiment of the present disclosure;
FIG. 3 is a flow chart of another method for obtaining target multimedia content from a multimedia content library that is matched with each search term separately, provided by an embodiment of the present disclosure;
Fig. 4 shows a schematic diagram of an information distribution apparatus provided by an embodiment of the present disclosure;
fig. 5 shows a schematic diagram of an electronic device provided by an embodiment of the disclosure.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are only some embodiments of the present disclosure, but not all embodiments. The components of the embodiments of the present disclosure, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure provided in the accompanying drawings is not intended to limit the scope of the disclosure, as claimed, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be made by those skilled in the art based on the embodiments of this disclosure without making any inventive effort, are intended to be within the scope of this disclosure.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
The term "and/or" is used herein to describe only one relationship, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.
It will be appreciated that prior to using the technical solutions disclosed in the embodiments of the present disclosure, the user should be informed and authorized of the type, usage range, usage scenario, etc. of the personal information related to the present disclosure in an appropriate manner according to the relevant legal regulations.
According to research, with the development of internet technology, other multimedia contents which are possibly interested can be recommended to the user while the user browses the multimedia contents, so that not only can the browsing interests of the user be improved, but also the time for the user to search for the other multimedia contents which are interested can be shortened.
However, the existing recommendation method mainly recommends other related and high-heat content according to the multimedia content currently consumed by the user, the method is easy to enable the multimedia content recommended to the user to be converged more and more, and the multimedia content finally pushed to each user is some high-heat content which is currently, so that the method does not meet the personalized requirements of some users.
Based on the above study, the present disclosure provides an information distribution method, an apparatus, an electronic device, and a storage medium. In the embodiment of the disclosure, each search term meeting the target consumption requirement can be determined based on the historical consumption data obtained through the authorization of the user; wherein the historical consumption data comprises historical search data and/or historical browsing data. According to the embodiment, the more effective and reasonable search words meeting the target consumption requirement can be determined through the historical consumption data of the user, so that the suitability between the search words and the current user is higher. Then, target multimedia contents respectively matched with each search word can be obtained from the multimedia content library, target recommendation information corresponding to the target multimedia contents is determined, and then information distribution is carried out on the target recommendation information on the information flow recommendation page. The embodiment not only widens the information sources of the information distributed in the information flow recommendation page and improves the diversity and richness of information distribution, but also enables the information distributed in the information flow recommendation page to more accord with the consumption requirements of the current user, further enables the distributed information to be more effective and improves the consumption experience of the user.
For the convenience of understanding the present embodiment, first, a detailed description will be given of an information distribution method disclosed in an embodiment of the present disclosure, and an execution body of the information distribution method provided in the embodiment of the present disclosure is generally an electronic device with a certain computing capability.
Referring to fig. 1, a flowchart of an information distribution method according to an embodiment of the present disclosure is shown, where the method includes steps S101 to S105, where:
s101: determining each search term meeting target consumption requirements based on historical consumption data obtained through user authorization; the historical consumption data includes historical search data and/or historical browsing data.
Here, the historical search data may include search information entered in a search engine by a user, which is authorized to be obtained by the user, or a search initiated by triggering certain search information (recommending search terms or links, etc.). Specifically, the search information may include search information of the user within a latest preset period of time.
In the embodiment of the disclosure, the search information input by the user or the triggered search information obtained by the authorization of the user can be directly determined as the historical search data of the user, or the search keywords obtained by extracting the keywords of the search information triggered or input by the user can be determined as the historical search data of the user.
The historical browsing data may include multimedia content browsed by the user, which is obtained through authorization of the user, and at this time, the multimedia content browsed by the user may include multimedia content browsed by the user and meeting browsing requirements.
The browsing request may indicate that the user views a part of the multimedia content when browsing the multimedia content, for example, the browsing request may be 50%, 30%, 100%, or the like.
For example, multimedia content that has been browsed by a user and that has been viewed by the user that is greater than 50% of the total multimedia content may be determined as historical browsing data for the user. Alternatively, the multimedia content that has been browsed by the user and has been completely read (in this case, the percentage of the part of the multimedia content that has been browsed by the user to the whole multimedia content is 100%) may be determined as the historical browsing data of the user.
In an embodiment of the present disclosure, determining each search term that meets the target consumption requirement may include, for example: determining that the historical search data and/or the historical browsing data meet each search word with preset requirements, for example, the corresponding search times of the search words in the historical search data meet the preset requirements, and/or the corresponding browsing times of the search words in the historical browsing data meet the preset requirements. Or, the consumption scores corresponding to the search words can be calculated by combining various historical consumption data and the weight of each historical consumption data, and the search words with the corresponding consumption scores larger than the set threshold value are selected as the search words meeting the target consumption requirement.
In another alternative embodiment, based on the historical consumption data obtained through the authorization of the user, determining each search term meeting the target consumption requirement may specifically include the following procedures:
step one, determining each candidate search word matched with a preset consumption frequency threshold value based on the historical consumption data authorized by the user.
Here, the number of consumption may be understood as the number of historical searches and/or historical browses counted after the historical consumption data is obtained by the user authorization.
For example, for historical search data, the number of times a search is initiated for the historical search data may be counted, resulting in a first number of consumption of the historical search data. At this time, the history search data with similar semantics can be determined as the same history search data, and the consumption times of the history search data with similar semantics can be counted to be used as the first consumption times of the history search data with similar semantics.
For another example, for historical browsing data, the number of times of triggering the browsing operation performed on the multimedia content may be counted, and the number of times may be determined as the second consumption number corresponding to the multimedia content.
Finally, the combination of the first consumption times and the second consumption times may be used as the consumption times corresponding to the historical consumption data, or the first consumption times and the second consumption times may be multiplied by weights corresponding to the consumption data types and added to be used as the consumption times corresponding to the historical consumption data.
At this time, the preset consumption time threshold may be a positive integer set arbitrarily, for example, the preset consumption time threshold may be 10, 20, 50, or the like, and the preset consumption time threshold is not specifically limited in the present disclosure so as to meet the actual needs.
For example, assuming that the preset consumption number threshold is 20, it may be determined that the historical consumption data having the search number or the browsing number greater than 20 among the historical consumption data authorized to be acquired by the user is determined as the historical consumption data matching the preset consumption number threshold.
Thereafter, each candidate search term may be determined based on the historical consumption data matching a preset number of consumption thresholds. For example, search information (or search keywords in the search information) input or triggered by the user in the historical consumption data matching the preset consumption number threshold may be determined as each candidate search word, and/or related information (e.g., content tags/topic tags corresponding to the multimedia content, title information of the multimedia content, keywords in the title information of the multimedia content, etc.) of the multimedia content consumed by the user in the historical consumption data matching the preset consumption number threshold may be determined as each candidate search word.
And step two, screening each search word from the candidate search words based on the search type corresponding to each candidate search word.
Here, the search type corresponding to each candidate search term may indicate a category of multimedia content corresponding to the candidate term, and for example, the search type may include fashion categories, delicacies, fun categories, knowledge science categories, travel categories, store categories, movie categories, education categories, events, news information, and the like.
For example, assuming that the candidate search term is "how a tiger is drawn," then it may be determined that the search type to which the candidate search term corresponds is "teaching class". As another example, assuming that the candidate search term is "football match," then the search type to which the candidate search term corresponds may be determined to be "event.
It should be noted here that the categories and names of the search types described above are merely exemplary, and in actual implementation, the search types may be divided according to requirements.
In one possible implementation manner, when screening search words from each candidate search word according to a search type, in an initial recommendation stage, equalization screening may be performed for each search type, for example, for each search type, a first preset number of search words with high consumption heat (for example, may be determined by using the consumption number or the latest comment number or the like) are selected from the candidate search words, so that each search word screened finally covers more search types as much as possible. Then, the search type for filtering can be updated in combination with the consumption condition of the user on the target recommendation information (see the description about S103 for details) under multiple search types, and further the subsequent target recommendation information can be updated.
In another possible implementation, since the search requirement of the user for information under different search types is different in time, for example, for news information, the user searches for a certain hot news several times a day, but may have the search requirement only for a short time when the hot news is generated, and after a long time, the user does not search for news information type contents any more. For another example, users often search for videos of some cooking categories, indicating that users may have a long-term search requirement for videos of food items.
Based on this, a scheme is proposed in this embodiment: some search types of the user with a bias of long-term search requirements, such as the delicacies, can be mined, search words meeting the time-lapse of the long-term search requirements can be screened out, and recommendation of related contents can be performed.
Specifically, the search terms matching the target search types can be selected from the candidate search terms based on the search types corresponding to the candidate search terms and at least one target search type; and the search demand aging corresponding to the target search type is larger than a set aging threshold.
In the embodiment of the present disclosure, the search requirement aging is used to indicate the effective duration corresponding to the search requirement, for example, the search requirement aging may be one month, one year, or the like.
The target search types may be determined based on historical search data of authorized users, for example, by counting some search types that are in the set age threshold, such as partial continuous searches, according to the set age threshold. For example, if a user searches for a food video from time to time within a month, it may be determined that the food belongs to the target search type of the user.
In addition, the target search type may be a search type that is determined by performing statistical analysis on global historical search data (historical search data of different users) and meets long-term requirement characteristics of general users.
In the embodiment of the disclosure, the search type corresponding to each candidate search word may be matched with at least one preset target search type, and in case of successful matching, the candidate search word corresponding to the search type that is successfully matched may be determined as each search word.
In the above embodiment, each candidate search term matching the preset consumption frequency threshold may be determined based on the historical consumption data obtained through the authorization of the user, so that the determined search recommended term may be ensured to be a search term that is not triggered by mistake, and thus the validity of the determined candidate recommended term may be ensured. And then, based on the search type corresponding to each candidate search term and at least one target search type with the preset search requirement time greater than the set time threshold, each search term matching the target search type is screened from each candidate search term, so that the determined each search term is ensured to be in a long-term requirement, and the related content corresponding to the search term is also in a long-term requirement, and the continuous consumption requirement of a user can be better met by recommending the related content.
S103: acquiring target multimedia contents respectively matched with each search word from a multimedia content library, and determining target recommendation information corresponding to the target multimedia contents; and the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition.
In an embodiment of the present disclosure, the foregoing preset condition may include that the matching degree is greater than a preset matching degree threshold, or may include that the matching degree is ranked before a preset ranking (from big to small ranking). That is, under the first preset condition, the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word is greater than or equal to a preset matching degree threshold; that is, from the multimedia content library, target multimedia content having a matching degree between search results corresponding to the search term and/or the search term greater than the preset matching degree threshold is screened. Under the second preset condition, the number threshold value N of the target multimedia contents can be considered to be preset, and the corresponding target multimedia contents with the previous N matching degrees are screened from the multimedia content library.
In embodiments of the present disclosure, each search term may match one or more targeted multimedia content, resulting in a plurality of targeted multimedia content.
In the embodiment of the disclosure, in the process of acquiring the target multimedia content respectively matched with each search word from the multimedia content library, one way is to determine the target multimedia content respectively matched with each search word from the multimedia content library, for example, the multimedia content of which the content tag of each multimedia content in the multimedia content library is matched with each search word may be determined as the target multimedia content, or the multimedia content of which the keyword of the keyword/text description information of the title information corresponding to each multimedia content in the multimedia content library is matched with each search word may also be determined as the target multimedia content.
Alternatively, the target multimedia content may be determined by determining multimedia content corresponding to the search results of the respective search terms from a multimedia content library. For example, multimedia content corresponding to search results corresponding to respective search terms in a multimedia content library may be determined as target multimedia content. Alternatively, the multimedia content in the multimedia content library, in which the search results corresponding to the respective search terms satisfy the association relationship, may be determined as the target multimedia content.
In the embodiment of the disclosure, after the multimedia content respectively matched with each search term is acquired, the target recommendation information corresponding to the target multimedia content may be determined, where the target recommendation information may be understood as one of content profile information, title information, text description information, recommendation terms related to the target multimedia content, and the like corresponding to the target multimedia content.
S105: and distributing the information of the target recommendation information on an information flow recommendation page.
In the embodiment of the disclosure, an information display style corresponding to each target multimedia content can be determined, and information distribution is performed on target recommendation information corresponding to the target multimedia content on an information flow recommendation page according to the information display style.
The information display style can indicate at least one of display position, display shape, display color, display font and display quantity of target recommendation information corresponding to the target multimedia content on the information flow recommendation page.
In the embodiment of the disclosure, each search term meeting the target consumption requirement can be determined based on the historical consumption data obtained through the authorization of the user; wherein the historical consumption data comprises historical search data and/or historical browsing data. According to the embodiment, the more effective and reasonable search words meeting the target consumption requirement can be determined through the historical consumption data of the user, so that the suitability between the search words and the current user is higher. Then, target multimedia contents respectively matched with each search word can be obtained from the multimedia content library, target recommendation information corresponding to the target multimedia contents is determined, and then information distribution is carried out on the target recommendation information on the information flow recommendation page. The embodiment not only widens the information sources of the information distributed in the information flow recommendation page and improves the diversity and richness of information distribution, but also enables the information distributed in the information flow recommendation page to more accord with the consumption requirements of the current user, further enables the distributed information to be more effective and improves the consumption experience of the user.
In an alternative embodiment, as shown in fig. 2, for S103 described above: the method comprises the following steps of obtaining target multimedia contents respectively matched with each search word from a multimedia content library:
step S21: determining each candidate multimedia content according to the first matching degree between the tag information of each multimedia content in the multimedia content library and the search word;
step S22: and determining the target multimedia content from the candidate multimedia contents according to the interaction data corresponding to the candidate multimedia contents.
Here, the tag information (i.e., the content tag) of each multimedia content in the multimedia content library may be tag information set in advance for the multimedia content, tag information determined based on text information extracted from the multimedia content, tag information determined based on text description information of the multimedia content (where the text description information may indicate title information or content profile information of the multimedia content, etc.).
In the embodiments of the present disclosure, the first matching degree between the tag information of the multimedia content and the search term may indicate a text similarity and/or a semantic similarity between the tag information of the multimedia content and the search term.
In the embodiment of the disclosure, the multimedia content, in which the first matching degree between the tag information of each multimedia content in the multimedia content library and the search word is greater than the preset matching degree threshold, may be determined as the candidate multimedia content.
And then determining target multimedia content from the candidate multimedia contents according to the interaction data corresponding to the candidate multimedia contents. Wherein the interactive data corresponding to the multimedia content indicates praise number, comment number, play completion rate (i.e. percentage of browsed multimedia content to total multimedia content), effective play rate (i.e. active trigger play) and the like for the multimedia content.
For example, the multimedia content having the highest praise among the candidate multimedia contents may be determined as the target multimedia content, or the multimedia content having the highest completion rate among the candidate multimedia contents may be determined as the target multimedia content.
In the above embodiment, each candidate multimedia content may be determined according to the first matching degree between the tag information of each multimedia content and each search term in the multimedia content library, so that each determined candidate multimedia content may better meet the target search requirement. Then, according to the interaction data corresponding to each candidate multimedia content, determining the target multimedia content from each candidate multimedia content.
In an alternative embodiment, as shown in fig. 3, for S103 described above: the method comprises the following steps of obtaining target multimedia contents respectively matched with each search word from a multimedia content library:
step S31: after searching is initiated based on the search word, each search result is obtained;
step S32: and acquiring the multimedia content with the second matching degree larger than a set threshold value from the multimedia content library as the target multimedia content.
In the embodiment of the disclosure, a search may be initiated based on the search word to obtain a search result corresponding to each search word, and at this time, each search result may be used as seed content to obtain, from the multimedia content library, multimedia content having a second matching degree with each search result greater than a set threshold as a target multimedia content. Or, after obtaining the search results corresponding to each search word, determining the consumption times of the corresponding search results and/or performing screening by using the completion rate to obtain initial search results, and then obtaining the multimedia content with the second matching degree larger than the set threshold value from the multimedia content library based on the initial search results as the target multimedia content.
Here, the second degree of matching between the multimedia content in the multimedia content library and each search result may indicate a degree of similarity between the multimedia content and the search result.
In the embodiment of the present disclosure, the similarity between each multimedia content and each search result may be determined through a word vector model, the similarity may also be calculated by calculating a distance between the multimedia content and the search result, and the like, and the method for calculating the similarity is not limited in this disclosure, so that implementation can be performed.
In the above embodiment, the search may be initiated based on the search word to obtain each search result, and then, the multimedia content with the second matching degree with each search result being greater than the set threshold value is obtained from the multimedia content library as the target multimedia content, so that the range of the obtained target multimedia content is wider, and the diversity and accuracy of the obtained target multimedia content are further improved.
After determining each target recommendation information, each target recommendation information needs to be distributed, and during distribution, the target recommendation information can be distributed in a mode of directly inserting information streams or in a mode of weakly sensing that the target recommendation information and other recommendation information are mixed and arranged, which is described in detail below.
For the manner of direct insertion of information streams, in practice, for S105: in the information flow recommendation page, the information distribution is carried out on the target recommendation information, and the method specifically comprises the following steps:
displaying the target recommendation information at a target display position in the information flow recommendation page; the target display positions belong to a preset number of information display positions which are ranked in front.
In the embodiment of the present disclosure, the preset number of information display bits may be associated with the number of target recommended information, for example, in the case where the number of target recommended information is n, the target display position is n information display bits ordered in front, that is, each target recommended information is sequentially displayed in the first few bits of the information stream, and after the target recommended information, other recommended information of the information stream is displayed (other recommended information is recommended information under the existing recommendation policy of the information stream, and is not determined based on the recommendation information determined in the embodiment of the present disclosure). Or, a preset number of information display bits can be reserved for the target recommendation information in the information display bits which are ranked at the front in the information flow recommendation page no matter the number of the target recommendation information; for example, the first m information display bits among the information display bits may be preset, and may be selectively occupied by the target recommendation information. That is, one or more target recommendation information may be inserted into one or more of the first m information display bits, respectively. n and m are positive integers.
In the case where the number of the target display positions is greater than the number of the target recommendation information, the target recommendation information may occupy a position that is forward of the target display positions, or may randomly occupy a part of the target display positions, and the present disclosure is not particularly limited, and may be implemented.
In the embodiment, the target recommendation information can be displayed through the preset number of information display positions which are in the front of the information flow recommendation page, so that the target recommendation information can be directly inserted into the information flow recommendation page, a user can timely view the target recommendation information, and the time for the user to search the target recommendation information is saved. Meanwhile, the perception degree of the recommended information of the user is relatively strong, and the experience is good for the user who has a demand for the target recommended information.
In the implementation, for S105, a weak perception manner of shuffling the target recommendation information with other recommendation information is adopted: in the information flow recommendation page, the information distribution of the target recommendation information is carried out, and the method specifically comprises the following steps:
and based on a pre-trained ranking model, carrying out mixed ranking on the determined target recommendation information and other contents to be recommended, and carrying out information distribution on the information flow recommendation page based on a ranking result.
Here, the other recommended content is each recommended information obtained according to the existing recommended policy of the information flow, for example, may be recommended information obtained by screening according to a preset screening feature, where the preset screening feature may be at least one of the following: browsing hotness features, search times features, real-time features, completion rate features, interactive data features, etc.
In the embodiment of the disclosure, the determined target recommendation information and other contents to be recommended can be mixed and arranged through a pre-trained sorting model, so that the target recommendation information and the other contents to be recommended can be naturally mixed together for display, on one hand, the mirror-out opportunity of the target recommendation information is increased, and on the other hand, the perception degree of the recommended information by a user is relatively weaker in the mode, and the appearance of the target recommendation information is more natural.
In an embodiment of the present disclosure, the ranking model may be trained according to the following steps:
firstly, positive samples and negative samples in the training samples and recommended weights of the positive samples are determined based on search attribute features corresponding to the training samples; the positive sample comprises multimedia content searched by a user, the negative sample comprises multimedia content not searched by the user, and the recommendation weight of the positive sample is related to the searching willingness indicated by the searching attribute characteristics of the positive sample.
Here, the search attribute feature corresponding to each training sample may indicate the initiative of the search of each training sample, for example, multimedia content searched by the user may be used as a positive sample in the training samples, multimedia content not searched by the user may be used as a negative sample in the training samples, and so on.
For example, a training sample that the user triggers to open (i.e., multimedia content) may be taken as a positive sample, and a training sample that automatically opens after recommendation (i.e., multimedia content) may be taken as a negative sample. Alternatively, training samples that are manually searched by the user (i.e., multimedia content) may be used as positive samples, and training samples that are passively triggered to search (i.e., multimedia content) may be used as negative samples.
In the embodiment of the disclosure, the recommendation weight of the positive sample may be determined based on the search willingness indicated by the search attribute feature of the positive sample, where the search willingness may indicate a subjective degree, for example, in the case that the positive sample is manually searched by a user, the search willingness indicated by the search attribute feature of the positive sample may be greater than the search willingness indicated by the search attribute feature of the positive sample that the positive sample triggers to open for the user. At this time, the search willingness degree may be determined by any one of the following behavior data: whether to manually search (actively input search information search, search not directly triggering search words or links), interactive behavior (e.g., praise, attention, comments, etc.) data, etc
The ranking model is then trained based on the positive and negative samples, and the recommended weights for each positive sample.
During specific training, the training targets are as follows: the positive sample is before, the negative sample is after, and in the positive sample before, the front with large recommendation weight and the rear with small recommendation weight are recommended.
According to the embodiment, the sorting model can be trained through the positive and negative samples and the recommended weight of each positive sample, so that the sorting result of the sorting model after training is more accurate and meets the searching wish of a user, the time for the user to initiate searching again is saved, and the efficiency is improved.
It will be appreciated by those skilled in the art that in the above-described method of the specific embodiments, the written order of steps is not meant to imply a strict order of execution but rather should be construed according to the function and possibly inherent logic of the steps.
Based on the same inventive concept, the embodiments of the present disclosure further provide an information distribution device corresponding to the information distribution method, and since the principle of solving the problem by the device in the embodiments of the present disclosure is similar to that of the information distribution method in the embodiments of the present disclosure, the implementation of the device may refer to the implementation of the method, and the repetition is omitted.
Referring to fig. 4, a schematic diagram of an information distribution apparatus according to an embodiment of the disclosure is provided, where the apparatus includes: a determination unit 41, an acquisition unit 42, a distribution unit 43; wherein, the liquid crystal display device comprises a liquid crystal display device,
a determining unit 41, configured to determine each search term meeting the target consumption requirement based on the historical consumption data authorized to be acquired by the user; the historical consumption data comprises historical search data and/or historical browsing data;
an obtaining unit 42, configured to obtain target multimedia contents respectively matched with the search terms from a multimedia content library, and determine target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition;
and the distributing unit 43 is configured to distribute information on the information stream recommendation page.
In the embodiment of the disclosure, each search term meeting the target consumption requirement can be determined based on the historical consumption data obtained through the authorization of the user; wherein the historical consumption data comprises historical search data and/or historical browsing data. According to the embodiment, the more effective and reasonable search words meeting the target consumption requirement can be determined through the historical consumption data of the user, so that the suitability between the search words and the current user is higher. Then, target multimedia contents respectively matched with each search word can be obtained from the multimedia content library, target recommendation information corresponding to the target multimedia contents is determined, and then information distribution is carried out on the target recommendation information on the information flow recommendation page. The embodiment not only widens the information sources of the information distributed in the information flow recommendation page and improves the diversity and richness of information distribution, but also enables the information distributed in the information flow recommendation page to more accord with the consumption requirements of the current user, further enables the distributed information to be more effective and improves the consumption experience of the user.
In a possible implementation, the determining unit 41 is further configured to:
determining each candidate search word matched with a preset consumption frequency threshold value based on the historical consumption data obtained through user authorization;
and screening each search word from the candidate search words based on the search type corresponding to each candidate search word.
In a possible implementation, the determining unit 41 is further configured to:
screening each search word matching the target search type from the candidate search words based on the search type corresponding to each candidate search word and at least one target search type; and the search demand aging corresponding to the target search type is larger than a set aging threshold.
In a possible implementation, the obtaining unit 42 is further configured to:
determining each candidate multimedia content according to the first matching degree between the tag information of each multimedia content in the multimedia content library and the search word;
and determining the target multimedia content from the candidate multimedia contents according to the interaction data corresponding to the candidate multimedia contents.
In a possible implementation, the obtaining unit 42 is further configured to:
After searching is initiated based on the search word, each search result is obtained;
and acquiring the multimedia content with the second matching degree larger than a set threshold value from the multimedia content library as the target multimedia content.
In a possible embodiment, the distribution unit 43 is further configured to:
displaying the target recommendation information at a target display position in the information flow recommendation page; the target display positions belong to a preset number of information display positions which are ranked in front.
In a possible embodiment, the distribution unit 43 is further configured to:
and based on a pre-trained ranking model, carrying out mixed ranking on the determined target recommendation information and other contents to be recommended, and carrying out information distribution on the information flow recommendation page based on a ranking result.
In a possible embodiment, the device is further configured to train to obtain the ranking model according to the following steps:
based on the search attribute characteristics corresponding to each training sample, positive samples and negative samples in the training samples and recommended weights of the positive samples are determined; the positive sample comprises multimedia content searched by a user, the negative sample comprises multimedia content not searched by the user, and the recommendation weight of the positive sample is related to the searching willingness indicated by the searching attribute characteristics of the positive sample;
The ranking model is trained based on the positive and negative samples, and a recommended weight for each positive sample.
The process flow of each module in the apparatus and the interaction flow between the modules may be described with reference to the related descriptions in the above method embodiments, which are not described in detail herein.
Corresponding to the information distribution method in fig. 1, the embodiment of the present disclosure further provides an electronic device 500, as shown in fig. 5, which is a schematic structural diagram of the electronic device 500 provided in the embodiment of the present disclosure, including:
a processor 51, a memory 52, and a bus 53; memory 52 is used to store execution instructions, including memory 521 and external storage 522; the memory 521 is also referred to as an internal memory, and is used for temporarily storing operation data in the processor 51 and data exchanged with the external memory 522 such as a hard disk, and the processor 51 exchanges data with the external memory 522 through the memory 521, and when the electronic device 500 is operated, the processor 51 and the memory 52 communicate with each other through the bus 53, so that the processor 51 executes the following instructions:
determining each search term meeting target consumption requirements based on historical consumption data obtained through user authorization; the historical consumption data comprises historical search data and/or historical browsing data;
Acquiring target multimedia contents respectively matched with each search word from a multimedia content library, and determining target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition;
and distributing the information of the target recommendation information on an information flow recommendation page.
The disclosed embodiments also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the information distribution method described in the method embodiments described above. Wherein the storage medium may be a volatile or nonvolatile computer readable storage medium.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described system and apparatus may refer to corresponding procedures in the foregoing method embodiments, which are not described herein again. In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present disclosure may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on such understanding, the technical solution of the present disclosure may be embodied in essence or a part contributing to the prior art or a part of the technical solution, or in the form of a software product stored in a storage medium, including several instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present disclosure. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Finally, it should be noted that: the foregoing examples are merely specific embodiments of the present disclosure, and are not intended to limit the scope of the disclosure, but the present disclosure is not limited thereto, and those skilled in the art will appreciate that while the foregoing examples are described in detail, it is not limited to the disclosure: any person skilled in the art, within the technical scope of the disclosure of the present disclosure, may modify or easily conceive changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features thereof; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the disclosure, and are intended to be included within the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (11)
1. An information distribution method, comprising:
determining each search term meeting target consumption requirements based on historical consumption data obtained through user authorization; the historical consumption data comprises historical search data and/or historical browsing data;
Acquiring target multimedia contents respectively matched with each search word from a multimedia content library, and determining target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition;
and distributing the information of the target recommendation information on an information flow recommendation page.
2. The method of claim 1, wherein determining each search term that meets the target consumption requirement based on historical consumption data authorized by the user comprises:
determining each candidate search word matched with a preset consumption frequency threshold value based on the historical consumption data obtained through user authorization;
and screening each search word from the candidate search words based on the search type corresponding to each candidate search word.
3. The method of claim 2, wherein the screening the respective search term from the respective candidate search term based on the search type corresponding to the respective candidate search term comprises:
screening each search word matching the target search type from the candidate search words based on the search type corresponding to each candidate search word and at least one target search type; and the search demand aging corresponding to the target search type is larger than a set aging threshold.
4. The method of claim 1, wherein the obtaining target multimedia content from the multimedia content library that matches the respective search term comprises:
determining each candidate multimedia content according to the first matching degree between the tag information of each multimedia content in the multimedia content library and the search word;
and determining the target multimedia content from the candidate multimedia contents according to the interaction data corresponding to the candidate multimedia contents.
5. The method of claim 1, wherein the obtaining target multimedia content from the multimedia content library that matches the respective search term comprises:
after searching is initiated based on the search word, each search result is obtained;
and acquiring the multimedia content with the second matching degree larger than a set threshold value from the multimedia content library as the target multimedia content.
6. The method according to claim 1, wherein the information distribution of the target recommendation information on the information flow recommendation page includes:
displaying the target recommendation information at a target display position in the information flow recommendation page; the target display positions belong to a preset number of information display positions which are ranked in front.
7. The method according to claim 1, wherein the information distribution of the target recommendation information on the information flow recommendation page includes:
and based on a pre-trained ranking model, carrying out mixed ranking on the determined target recommendation information and other contents to be recommended, and carrying out information distribution on the information flow recommendation page based on a ranking result.
8. The method of claim 7, wherein the ranking model is trained in accordance with the steps of:
based on the search attribute characteristics corresponding to each training sample, positive samples and negative samples in the training samples and recommended weights of the positive samples are determined; the positive sample comprises multimedia content searched by a user, the negative sample comprises multimedia content not searched by the user, and the recommendation weight of the positive sample is related to the searching willingness indicated by the searching attribute characteristics of the positive sample;
the ranking model is trained based on the positive and negative samples, and a recommended weight for each positive sample.
9. An information distribution apparatus, comprising:
the determining unit is used for determining each search word meeting the target consumption requirement based on the historical consumption data authorized by the user; the historical consumption data comprises historical search data and/or historical browsing data;
The acquisition unit is used for acquiring target multimedia contents respectively matched with the search words from the multimedia content library and determining target recommendation information corresponding to the target multimedia contents; the matching degree between the target multimedia content and the search word and/or the search result corresponding to the search word meets the preset condition;
and the distribution unit is used for distributing the information of the target recommendation information on the information flow recommendation page.
10. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication over the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the information distribution method according to any of claims 1 to 8.
11. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the information distribution method according to any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310559073.3A CN116610858A (en) | 2023-05-17 | 2023-05-17 | Information distribution method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310559073.3A CN116610858A (en) | 2023-05-17 | 2023-05-17 | Information distribution method, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116610858A true CN116610858A (en) | 2023-08-18 |
Family
ID=87679341
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310559073.3A Pending CN116610858A (en) | 2023-05-17 | 2023-05-17 | Information distribution method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116610858A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117391824A (en) * | 2023-12-11 | 2024-01-12 | 深圳须弥云图空间科技有限公司 | Method and device for recommending articles based on large language model and search engine |
-
2023
- 2023-05-17 CN CN202310559073.3A patent/CN116610858A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117391824A (en) * | 2023-12-11 | 2024-01-12 | 深圳须弥云图空间科技有限公司 | Method and device for recommending articles based on large language model and search engine |
| CN117391824B (en) * | 2023-12-11 | 2024-04-12 | 深圳须弥云图空间科技有限公司 | Method and device for recommending articles based on large language model and search engine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10366119B2 (en) | Customized content stream utilizing dwelltime-based machine learning | |
| CN106802915B (en) | Academic resource recommendation method based on user behaviors | |
| Wang et al. | Mining term association patterns from search logs for effective query reformulation | |
| US11176142B2 (en) | Method of data query based on evaluation and device | |
| CN109690529B (en) | Compiling documents into a timeline by event | |
| CN107832437B (en) | Audio/video pushing method, device, equipment and storage medium | |
| Szpektor et al. | When relevance is not enough: Promoting diversity and freshness in personalized question recommendation | |
| US10032081B2 (en) | Content-based video representation | |
| US10515130B2 (en) | Prioritizing items from different categories in a news stream | |
| CN109511015B (en) | Multimedia resource recommendation method, device, storage medium and equipment | |
| CA3153598A1 (en) | Method of and device for predicting video playback integrity | |
| CN110334356B (en) | Article quality determining method, article screening method and corresponding device | |
| Tsurel et al. | Fun facts: Automatic trivia fact extraction from wikipedia | |
| Jin et al. | MySpace video recommendation with map-reduce on qizmt | |
| CN111400546B (en) | Video recall method and video recommendation method and device | |
| Arantes et al. | Understanding video-ad consumption on YouTube: a measurement study on user behavior, popularity, and content properties | |
| US20180189291A1 (en) | Method and apparatus for sorting related searches | |
| Wang et al. | Bilateral correspondence model for words-and-pictures association in multimedia-rich microblogs | |
| CN116610858A (en) | Information distribution method, device, electronic equipment and storage medium | |
| CN114417157B (en) | Data pushing method and device, computer equipment and computer medium | |
| Said | Evaluating the accuracy and utility of recommender systems | |
| CN111597469B (en) | Display position determining method and device, electronic equipment and storage medium | |
| Bogers | Recommender systems for social bookmarking | |
| WO2008032037A1 (en) | Method and system for filtering and searching data using word frequencies | |
| CN113282789A (en) | Content display method and device, electronic equipment and readable 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 |