CN110096646A - The generation of category related information and its video pushing method and relevant device - Google Patents
The generation of category related information and its video pushing method and relevant device Download PDFInfo
- Publication number
- CN110096646A CN110096646A CN201910381032.3A CN201910381032A CN110096646A CN 110096646 A CN110096646 A CN 110096646A CN 201910381032 A CN201910381032 A CN 201910381032A CN 110096646 A CN110096646 A CN 110096646A
- Authority
- CN
- China
- Prior art keywords
- category
- user
- attribute
- related information
- video data
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—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
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/7867—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
-
- 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)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of generation of category related information and its video pushing methods and relevant device.This method comprises: determining the attribute of user, the attribute includes objective attribute target attribute；Determine that category, user's watched belong to the video data of the category；For the user with the same objective attribute target attribute, the target category with incidence relation is determined from the category, as category related information；The category related information is associated with the objective attribute target attribute.According to the viewing preference of group of subscribers, the category with incidence relation is excavated, so that when based on incidence relation selection video data, recall rate can be improved, also, the range of category is larger, when selecting video data based on category, good video data can be recalled, improves practicability.
Description
Technical field
The present embodiments relate to the technology of video processing more particularly to generation and its videos of a kind of category related information
Method for pushing and relevant device.
Background technique
With the development of science and technology, especially mobile terminal is universal, user is recorded, the video data of publication increasingly enriches,
Quantity is more huge.
Currently, other users are mostly in video website by keyword search, video website provides multiple video datas, should
User browses the information such as thumbnail, title, brief introduction, finds the video data viewing of oneself hobby.
Website is in order to save the time that user finds the video data of hobby, generally according to the search history data of user,
Relevant video data is searched, and is pushed to user.
But based on the relevant video data of user's existing search habit lookup, recall rate is lower, in addition, due to
The search range at family is smaller, is easy to cause to waste to the good video data in website.
Summary of the invention
The embodiment of the present invention provides generation and its video pushing method and the relevant device of a kind of category related information, with solution
Certainly according to the relevant video data of search history data-pushing of user, recall rate is low, the good video counts in waste website
According to the problem of.
In a first aspect, the embodiment of the invention provides a kind of generation methods of category related information, comprising:
Determine that the attribute of user, the attribute include objective attribute target attribute；
Determine that category, user's watched belong to the video data of the category；
For the user with the same objective attribute target attribute, the target product with incidence relation are determined from the category
Class, as category related information；
The category related information is associated with the objective attribute target attribute.
Second aspect, the embodiment of the invention also provides a kind of video pushing methods based on category related information, comprising:
Determine the attribute of user；
Determine that the first category, user's watched belong to the video data of first category；
The category related information of the Attribute Association is searched, the category in the category related information has incidence relation；
The determining category with first category with incidence relation in the category related information, as the second product
Class；
The video data for belonging to second category is pushed into the user.
The third aspect, the embodiment of the invention also provides a kind of generating means of category related information, comprising:
Attribute determination module, for determining that the attribute of user, the attribute include objective attribute target attribute；
Category determining module, for determining that category, user's watched belong to the video data of the category；
Category related information generation module, for being directed to the user with the same objective attribute target attribute, from the category
The target category with incidence relation is determined, as category related information；
Relating module, for the category related information to be associated with the objective attribute target attribute.
Fourth aspect, the embodiment of the invention also provides a kind of video push devices based on category related information, comprising:
Attribute determination module, for determining the attribute of user；
First category determining module, for determining that the first category, user's watched belong to the view of first category
Frequency evidence；
Category related information searching module, for searching the category related information of the Attribute Association, the category association
Category in information has incidence relation；
Second category determining module is associated with for determining to have with first category in the category related information
The category of system, as the second category；
Video data pushing module, for the video data for belonging to second category to be pushed to the user.
5th aspect, the embodiment of the invention also provides a kind of computer equipment, the computer equipment includes:
One or more processors；
Memory, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes the generation method of category related information as described in relation to the first aspect, alternatively, being closed as described in second aspect based on category
Join the video pushing method of information.
6th aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program realizes the generation method of category related information as described in relation to the first aspect when the program is executed by processor, alternatively, such as the
Video pushing method based on category related information described in two aspects.
In embodiments of the present invention, the attribute of user is determined, which includes objective attribute target attribute, determines category, and user has seen
It sees the video data for belonging to the category, for the user with same objective attribute target attribute, determines that there is incidence relation from category
Target category, as category related information, by category related information associated objects attribute, on the one hand, according to the sight of group of subscribers
It sees preference, excavates the category with incidence relation, so that can be improved and recall when based on incidence relation selection video data
Rate, also, the range of category is larger, when based on category selection video data, can recall good video data, improve practical
Property, on the other hand, different category related informations is generated for the user of different attribute, embodies different attribute user and watch product
Difference between class improves the push accuracy rate of video data.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the generation method for category related information that the embodiment of the present invention one provides；
Fig. 2 is a kind of flow chart of the generation method of category related information provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of flow chart for video pushing method based on category related information that the embodiment of the present invention three provides；
Fig. 4 is a kind of structural schematic diagram of the generating means for category related information that the embodiment of the present invention four provides；
Fig. 5 is a kind of structural representation for video push device based on category related information that the embodiment of the present invention five provides
Figure；
Fig. 6 is a kind of structural schematic diagram for computer equipment that the embodiment of the present invention six provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is a kind of flow chart of the generation method for category related information that the embodiment of the present invention one provides, the present embodiment
It is applicable to that there is the case where video category of incidence relation for the usage mining of different attribute, this method can be by a kind of product
The generating means of class related information execute, which can be configurable on computer equipment by software and or hardware realization
In, such as server, the video data of polymerizable various categories, this method specifically include following step in the computer equipment
It is rapid:
S101, the attribute for determining user.
Under normal circumstances, user can refer to the user in computer equipment (such as server) registration, with User ID, account etc.
Form indicates that the user has recorded its attribute, such as registion time, gender, age, hobby, residence in registration.
Certainly, user can refer to not in the user of computer equipment registration, with device identification (such as IMEI
(International Mobile Equipment Identity, international mobile equipment identification number)), the shapes such as casual user ID
Formula indicates that the present embodiment is without restriction to this.
Since the account of the history of the video data of viewing can be to video data in computer equipment (such as server) by user
Category hobby generate deviation therefore situation can be enlivened as example in computer equipment (such as server) using user
It is illustrated.
In this example, the user data that can inquire user determines the registion time of the user, when determining registration with this
It is long, difference of the registration time length between registion time and current time.
Further, the different periods can be divided in advance, and for the different user class of different period configurations, generation is reflected
Penetrate relationship.
For example, user class is referred to as new user, and 1 year or more, user class can be referred to as old user in 1 year.
Therefore, for the user, it may be determined that the period belonging to registration time length, to determine the period in the mapping relations
Corresponding user class, the attribute as user.
Certainly, above-mentioned attribute and its method of determination are intended only as example, can be according to practical feelings when implementing the present embodiment
Other attributes and its method of determination is arranged in condition, and the present embodiment is without restriction to this.In addition, in addition to above-mentioned attribute and its determination side
Outside formula, those skilled in the art can also use other attributes and its method of determination according to actual needs, and the present embodiment is to this
It is without restriction.
S102, category is determined.
In practical applications, the video data of the categories such as amusement, game, video display, news, animation is the major search of user
Object, this shows that user has video data itself the property of general demand.User often without very strong purpose, and
" non-that ", but certain scalability is had, as long as in the scope (category) that video data is liked in user.
Therefore, in the present embodiment, the video data of group of subscribers watched is summarized, determines the product of its ownership
Class determines category belonging to the video data of user's watched, in contrast, for the category, property can refer to user
The video data of watched ownership category.
It should be noted that, for the category of its record, being single and unordered, i.e. user's repetition sight for a user
The category seen, record is primary, if any repetition, does not record, also, does not record the sequence that user watches category.
In one embodiment of the invention, S102 may include steps of:
S1021, operation data of the user to the video data for belonging to some category is determined.
S1022, judge whether the operation data meets preset effectiveness condition；If so, S1023 is executed, if it is not,
Then execute S1024.
S1023, to category described in the user record.
S1024, the category is ignored to the user.
In the present embodiment, due to user viewing video data it is large number of, but the value of partial video data compared with
It is low, in order to reduce operand, effectiveness condition can be preset, which is used for the video data for indicating effectively to watch
Possessed characteristic.
If user meets effectiveness condition to the operation data of the video data under some category, then it represents that the video counts
According to effective viewing is belonged to, the video data that the user watches in the category is recorded.
If user does not meet effectiveness condition to the operation data of the video data under some category, then it represents that the video
Data belong to invalid viewing, ignore the video data that the user watches in the category.
In one example, which is viewing time, and the viewing time is effective for same category, it is not required that
Belong to the same video data, i.e., can whithin a period of time in S1021, such as 30 days, accumulation user was to belonging to some category
The viewing time of video data.
Correspondingly, in S1022, it can determine whether viewing time is greater than preset time threshold, such as 5 minutes.If so,
Then determination meets effectiveness condition；If not, it is determined that be unsatisfactory for effectiveness condition.
Certainly, aforesaid operations data and its effectiveness condition are intended only as example, can basis when implementing the present embodiment
Other operation datas and its effectiveness condition is arranged in actual conditions, for example, operation data is such as to deliver barrage, thumb up, share
The operation such as link, effectiveness condition be that the frequency of operation is then considered as greater than preset threshold value and meets effectiveness condition, are otherwise considered as
Effectiveness condition, etc. is not met, the present embodiment is without restriction to this.In addition, in addition to aforesaid operations data and its validity
Outside condition, those skilled in the art can also use other operation datas and its effectiveness condition, this implementation according to actual needs
Example is also without restriction to this.
S103, for the user with the same objective attribute target attribute, the mesh with incidence relation is determined from the category
Category is marked, as category related information.
In the present embodiment, the attribute that can traverse group of subscribers successively selects an attribute, as mesh from all properties
Mark attribute.
For the corresponding category of the objective attribute target attribute, to category be associated analysis (association analysis) or
Correlation rule learns (association rule learning), finds category from the data set of large-scale user-category
Between implication relation imply that there may be very strong between two kinds of categories there are correlation rule (association rules)
Relationship.
In this regard, the category with incidence relation can be described as target category, incidence relation can be described as category related information.
It should be noted that due to different attribute user for video data category hobby there are deviation,
For the user of different attribute, different data sets is generated to the category of the video data of its watched respectively, is excavated respectively not
Same category related information.
S104, the category related information is associated with the objective attribute target attribute.
For the category related information excavated, then can mark be associated with the objective attribute target attribute, storage in the database, so as to
The subsequent user for same alike result uses.
For example, generating key-value using the objective attribute target attribute as key (key), category related information as value (value)
(key-value pair), so that storage is in the database.
In another example index file is generated, to be stored in data using the objective attribute target attribute as the index of category related information
In library.
In embodiments of the present invention, the attribute of user is determined, which includes objective attribute target attribute, determines category, and user has seen
It sees the video data for belonging to the category, for the user with same objective attribute target attribute, determines that there is incidence relation from category
Target category, as category related information, by category related information associated objects attribute, on the one hand, according to the sight of group of subscribers
It sees preference, excavates the category with incidence relation, so that can be improved and recall when based on incidence relation selection video data
Rate, also, the range of category is larger, when based on category selection video data, can recall good video data, improve practical
Property, on the other hand, different category related informations is generated for the user of different attribute, embodies different attribute user and watch product
Difference between class improves the push accuracy rate of video data.
Embodiment two
Fig. 2 is a kind of flow chart of the generation method of category related information provided by Embodiment 2 of the present invention, the present embodiment
Based on previous embodiment, the processing operation of Mining Frequent Itemsets Based is further increased.This method specifically comprises the following steps:
S201, the attribute for determining user.
Wherein, which includes objective attribute target attribute.
S202, category is determined.
Wherein, user's watched belongs to the video data of the category.
S203, for the user with the same objective attribute target attribute, frequent item set is determined from the category.
Confidence level in S204, the calculating frequent item set between category.
If S205, the confidence level are greater than preset confidence threshold value, it is determined that the category in the frequent item set is mesh
Mark category.
S206, determine that the incidence relation between the target category is category related information.
In the present embodiment, for the user with same attribute, tool can be determined from category by Apriori algorithm
Relevant target category, as category related information.
For Apriori algorithm, can Mining Frequent Itemsets Based, support, confidence level be all satisfied requirement (frequent item set
Support is greater than confidence threshold value greater than the confidence level between category in support threshold, frequent item set), in the frequent item set
Incidence relation between non-proper subclass (category) is category related information.
Wherein, frequent item set (frequent item sets) is the set of the category occurred together.
Item collection refers to the set of item, and item can be the category of video data, then item collection is exactly the set of category.
Support (support) refers to that ratio shared by the record in data set comprising the item collection, that is, the item collection exist
The frequency of occurrences in data set, to measure the frequent degree of item collection.
For category X and Y, then corresponding support are as follows:
Support (X, Y)=P (XY)=number (XY)/num (AllSamples)
Wherein, P (XY) refers to X and Y in data set while the probability of appearance, number (XY) are the number that X and Y occurs simultaneously
Amount, num (AllSamples) refer to the total quantity of data intensive data, and a data refers to record user-category information.
Confidence level (confidence), also known as confidence level, define for correlation rule, indicate certain item collection specified
Under the conditions of probability of occurrence, to measure the relationship between category.
For category X and Y, then corresponding confidence level are as follows:
Wherein, P (XY) refer to X and Y in data set and meanwhile occur probability, the probability that P (Y) occurs in data set.
For example, the confidence level that X corresponds to Y is 40%, support 1% in data set.Then mean in data set,
A total of 1% user had not only watched the video data of category Y but also had watched the video data of category X, watched the video data of category Y
User in have 40% user watch category X video data.
It should be noted that being typically chosen the frequent item set greater than 1 in frequent item set.
In one embodiment of the invention, S203 includes the following steps:
S2031, for the user with the same objective attribute target attribute, calculate the support of this frequent item set.
Wherein, frequent item set is a category for the first time.
S2032, the frequent item set for being less than preset support threshold from support described in the frequent episode concentration filter.
S2033, judge whether the filtered frequent item set is empty；If so, S2034 is executed, if it is not, then executing
S2035。
S2034, determine that iteration terminates.
S2035, the category is increased to the filtered frequent item set, as frequent item set next time, returned
S2031。
In the present embodiment, the method that Apriori algorithm uses iteration, excavates maximum frequent item set, also, in order to
The calculating time needed for reducing, beta pruning is carried out during iteration.
In the concrete realization, the support of frequent k item collection is calculated, k is positive integer.
The data set that support in frequent k item collection is lower than support threshold is removed, frequent k item collection, the frequency to be output are obtained
Numerous k item collection.
If obtained frequent k item collection is sky, it is determined that the set of frequent k-1 item collection is as maximum frequent item set, repeatedly
In generation, terminates.
If obtained frequent k item collection is not sky, it is based on frequent k item collection, connection generates frequent k+1 item collection, and entrance is next
Secondary iteration.
S207, the category related information is associated with the objective attribute target attribute.
Certainly, above-mentioned Apriori algorithm is intended only as example, when implementing the present embodiment, can set according to the actual situation
It sets other modes and excavates category related information, for example, FP-Tree, GSP (FP-Tree, Generalized Sequential
Pattern mining algorithm, broad sense Sequential Pattern Mining Algorithm), CBA (Classification base of
Association, the algorithm classified based on correlation rule), etc., the present embodiment is without restriction to this.In addition, in addition to
Outside above-mentioned Apriori algorithm, those skilled in the art can also excavate category association letter using other way according to actual needs
Breath, the present embodiment are also without restriction to this.
To make those skilled in the art more fully understand the present embodiment, illustrate the present embodiment below by way of specific example
The generation method of middle category related information.
For the user jointly with some attribute (such as new user), a data set D (user-is represented with following table
Category), wherein support threshold 50%, confidence threshold value 70%.
| User | Item collection (category) |
| User_1 | Game A, game B, game C, game D |
| User_2 | Game A, game B, game C |
| User_3 | Game C, game D |
| User_4 | Game A, game B, game E |
(1) frequent 1 item collection C1={ { game A }, { game B }, { game C }, { game E }, { game D } } is generated.
(2) scan data set D calculates support of each item collection in D in C1.
It can be concluded that the support number of each item collection is respectively 3,3,3,1,2 in C1 from data set D, data in data set D
Total quantity be 4, it therefore follows that the support of each item collection is respectively 75%, 75%, 75%, 25%, 50% in C1.According to
Support threshold (50%) is filtered C1, it can be deduced that frequent 1 item collection L1={ { game A }, { game B }, { trip after filtering
Play C }, { game D } }.
(3) candidate frequent 2 item collection C2={ { game A, game B }, { game A, game C }, { game A, trip are generated according to L1
Play D }, { game B, game C }, { game B, game D }, { game C, game D } }.
(4) scan data set D calculates support of each item collection in D in C2.
It can be concluded that the support number of each item collection of C2 is respectively 3,2,1,2,1,2 from data set D, data in data set D
Total quantity be 4, it therefore follows that the support of each item collection is respectively 75%, 50%, 25%, 50%, 25% in C2,
50%.C2 is filtered according to support threshold (50%), it can be deduced that filtered frequent 2 item collection L2={ { game A, trip
Play B }, { game A, game C }, { game B, game C }, { game C, game D } }.
(5) candidate frequent 3 item collection C3={ { game A, game B, game C }, { game A, game B, game are generated according to L2
D }, { game A, game C, game D }, { game B, game C, game D } }.
Since a subset { game B, game D } in C3 in item collection { game A, game B, game D } is present in L2,
Therefore it can remove.
Similarly, item collection { game A, game C, game D }, { game B, game C, game D } also can remove.
Therefore, C3={ game A, game B, game C }.
(6) scan data set D calculates support of each item collection in D in C3.
It can be concluded that the support number of each item collection is respectively 2 from data set D, the total quantity of data is 4 in data set D,
It therefore follows that the support of each item collection is respectively 50% in C2.Root is filtered C3 according to support threshold (50%),
It can be concluded that filtered frequent 3 item collection L3={ { game A, game B, game C } }.
(7) L=L1 ∪ L2 ∪ L3={ { game A }, { game B }, { game C }, { game D }, { game A, game B }, { trip
Play A, game C }, { game B, game C }, { game C, game D }, { game A, game B, game C } }.
(8) consider the item collection that item collection length is greater than 1.
For example, { game A, game B, game C }, its all non-proper subclass { game A }, { game B }, { game C }, { trip
Play A, game B }, { game A, game C }, { game B, game C } calculates separately correlation rule { game A }-> { game B, game
C }, { game B }-> { game A, game C }, { game C }-> { game A, game B }, { game A, game B }-> { game C },
{ game A, game C }-> { game B }, the confidence level of { game B, game C }-> { game A }, value is respectively 67%, 67%,
67%, 67%, 100%, 100%.Since confidence threshold value is 70%, can obtain, { game A, game C }-> { game B }, { game
B, game C }-> { game A } be category correlation rule, that is to say, that viewing game A and game C video data while meeting
The video data of game B is watched, is also the video counts that can watch game A while watching the video data of game B and game C
According to.
Embodiment three
Fig. 3 is a kind of flow chart for video pushing method based on category related information that the embodiment of the present invention three provides,
The present embodiment is applicable to have the case where video data of incidence relation, this method for user's push category of different attribute
Can be executed by a kind of video push device based on category related information, the device can by software and or hardware realization,
It is configurable in computer equipment, such as server, the video data of polymerizable various categories, the party in the computer equipment
Method specifically comprises the following steps:
S301, the attribute for determining user.
Under normal circumstances, user can refer to the user in computer equipment (such as server) registration, with User ID, account etc.
Form indicates that the user has recorded its attribute, such as registion time, gender, age, hobby, residence in registration.
Certainly, user can refer to not in the user of computer equipment registration, with device identification (such as IMEI
(International Mobile Equipment Identity, international mobile equipment identification number)), the shapes such as casual user ID
Formula indicates that the present embodiment is without restriction to this.
Since the account of the history of the video data of viewing can be to video data in computer equipment (such as server) by user
Category hobby generate deviation therefore situation can be enlivened as example in computer equipment (such as server) using user
It is illustrated.
In this example, the user data that can inquire user determines the registion time of the user, when determining registration with this
It is long, difference of the registration time length between registion time and current time.
Further, the different periods can be divided in advance, and for the different user class of different period configurations, generation is reflected
Penetrate relationship.
For example, user class is referred to as new user, and 1 year or more, user class can be referred to as old user in 1 year.
Therefore, for the user, it may be determined that the period belonging to registration time length, to determine the period in the mapping relations
Corresponding user class, the attribute as user.
Certainly, above-mentioned attribute and its method of determination are intended only as example, can be according to practical feelings when implementing the present embodiment
Other attributes and its method of determination is arranged in condition, and the present embodiment is without restriction to this.In addition, in addition to above-mentioned attribute and its determination side
Outside formula, those skilled in the art can also use other attributes and its method of determination according to actual needs, and the present embodiment is to this
It is without restriction.
S302, the first category is determined.
In practical applications, the video data of the categories such as amusement, game, video display, news, animation is the major search of user
Object, this shows that user has video data itself the property of general demand.User often without very strong purpose, and
" non-that ", but certain scalability is had, as long as in the scope (category) that video data is liked in user.
Therefore, in the present embodiment, the video data of active user's watched is summarized, determines the first of its ownership
Category determines the first category belonging to the video data of user's watched, in contrast, for the category, property can be with
Refer to that user's watched belongs to the video data of the first category.
It should be noted that, for the category of its record, being single and unordered, i.e. user's repetition sight for active user
The category seen, record is primary, if any repetition, does not record, also, does not record the sequence that user watches category.
In one embodiment of the invention, S302 may include steps of:
S3021, operation data of the user to the video data for belonging to some the first category is determined.
S3022, judge whether the operation data meets preset effectiveness condition；If so, S3023 is executed, if it is not,
Then execute S3024.
S3023, to the first category described in the user record.
S3024, first category is ignored to the user.
In the present embodiment, due to user viewing video data it is large number of, but the value of partial video data compared with
It is low, in order to reduce operand, effectiveness condition can be preset, which is used for the video data for indicating effectively to watch
Possessed characteristic.
If user meets effectiveness condition to the operation data of the video data under some first category, then it represents that the view
Frequency records the video data that the user watches in first category according to effective viewing is belonged to.
If user does not meet effectiveness condition to the operation data of the video data under some first category, then it represents that should
Video data belongs to invalid viewing, ignores the video data that the user watches in first category.
In one example, which is viewing time, and the viewing time is effective for same category, it is not required that
Belong to the same video data, i.e., can whithin a period of time in S3021, such as 30 days, accumulation user was to some first product of ownership
The viewing time of the video data of class.
Correspondingly, in S3022, it can determine whether viewing time is greater than preset time threshold, such as 5 minutes.If so,
Then determination meets effectiveness condition；If not, it is determined that be unsatisfactory for effectiveness condition.
Certainly, aforesaid operations data and its effectiveness condition are intended only as example, can basis when implementing the present embodiment
Other operation datas and its effectiveness condition is arranged in actual conditions, for example, operation data is such as to deliver barrage, thumb up, share
The operation such as link, effectiveness condition be that the frequency of operation is then considered as greater than preset threshold value and meets effectiveness condition, are otherwise considered as
Effectiveness condition, etc. is not met, the present embodiment is without restriction to this.In addition, in addition to aforesaid operations data and its validity
Outside condition, those skilled in the art can also use other operation datas and its effectiveness condition, this implementation according to actual needs
Example is also without restriction to this.
S303, the category related information for searching the Attribute Association.
In the present embodiment, the category in category related information has incidence relation, and category related information is shaped like X → Y
Implication expression formula, wherein X and Y is disjoint item collection (category), indicate user viewing category X video data it is same
When, the video data of category Y can be watched.
The category related information can be generated by the method in embodiment one, embodiment two and relating attribute (target category
Property), record in the database, after determining the attribute of active user, searches the attribute (objective attribute target attribute) pass in the database
The category related information of connection.
S304, the determining category with first category with incidence relation in the category related information, as the
Two categories.
In the concrete realization, the category in category related information can be traversed, searching has incidence relation with the first category
Category, as the second category.
In one embodiment of the invention, S304 may include steps of:
The first quantity of S3041, statistics first category.
If S3042, first quantity are less than preset first threshold, lookup and institute in the category related information
The category that the first category has incidence relation is stated, as the second category.
In the present embodiment, first threshold can be preset.
If the first quantity of the first category is less than the first threshold, indicate that the category of user's viewing is less, at this point, can be
The category that there is incidence relation with the first category is searched in category related information, as the second category.
If the first quantity of the first category is greater than or equal to the first threshold, indicate that the category of user's viewing is more, this
When, it is negligible to push other video datas to the user using category related information.
S305, the video data for belonging to second category is pushed into the user.
In the concrete realization, after determining the second category, can store from the computer equipment (such as server) this
In video data under two categories, the video data of the suitable user is selected, the client of user login is pushed to.
In one embodiment of the invention, S305 may include steps of:
S3051, lookup belong to the video data of second category.
The second quantity of S3052, the statistics video data.
If S3053, second quantity are greater than preset second threshold, the video factor is selected to meet preset push away
The video data for sending condition, as target video data.
S3054, the target video data is pushed into the user.
In the present embodiment, video data has the video factor, that is, influences the factor of selection video data.
If counting the second quantity of video data, it is greater than second threshold, indicates that the quantity of video data is more, this
When, postsearch screening can be carried out based on the video factor, select the data of suitable video, push to the use as target video data
The client that family logs in.
In one example, the video factor includes at least one of the confidence level of the second category, the temperature of video data.
For the confidence level of the second category, which can be set to confidence level greater than third threshold value, and/or, it sets
Reliability highest n, etc..
For the temperature of video data, it can be presented as playback volume, comment number, barrage number etc., which can set
Temperature is set to greater than the 4th threshold value, and/or, temperature highest m, etc..
In embodiments of the present invention, the attribute for determining user determines that the first category, user's watched belong to the first category
Video data searches the category related information of Attribute Association, and the category in category related information has incidence relation, closes in category
It is determining in connection information with the first category to there is the category of incidence relation belong to the video of the second category as the second category
Data-pushing is to user, on the one hand, category incidence relation can embody the viewing preference of group of subscribers, be selected based on category incidence relation
Video data is selected, recall rate can be improved, also, the range of category is larger, video data is selected based on category, can be recalled good
Video data improves practicability, on the other hand, uses different category related informations for the user of different attribute, embodies
Different attribute user watches the difference between category, improves the push accuracy rate of video data.
Example IV
Fig. 4 is a kind of structural schematic diagram of the generating means for category related information that the embodiment of the present invention four provides, the dress
It sets and can specifically include following module:
Attribute determination module 401, for determining that the attribute of user, the attribute include objective attribute target attribute；
Category determining module 402, for determining that category, user's watched belong to the video data of the category；
Category related information generation module 403, for being directed to the user with the same objective attribute target attribute, from the category
Middle determination has the target category of incidence relation, as category related information；
Relating module 404, for the category related information to be associated with the objective attribute target attribute.
In one embodiment of the invention, the attribute determination module 401 includes:
Registion time determines submodule, for determining the registion time of the user；
Registration time length determines submodule, and for determining registration time length, the registration time length is for the registion time and currently
Difference between time；
Period determines submodule, for determining the period belonging to the registration time length；
User class determines submodule, the attribute for determining the period corresponding user class, as the user.
In one embodiment of the invention, the category determining module 402 includes:
Operation data determines submodule, for determining operand of the user to the video data for belonging to some category
According to；
Effectiveness condition judging submodule, English judge whether the operation data meets preset effectiveness condition；If
It is then to call record sub module, ignores submodule if it is not, then calling；
Record sub module, for category described in the user record；
Ignore submodule, for ignoring the category to the user.
In an example of the present embodiment, the operation data determines that submodule includes:
Viewing time cumulative unit, for accumulating viewing time of the user to the video data for belonging to some category；
The effectiveness condition judging submodule includes:
Time threshold judging unit, for judging whether the viewing time is greater than preset time threshold；If so, adjusting
With determination unit is met, determination unit is unsatisfactory for if it is not, then calling；
Meet determination unit, meets effectiveness condition for determination；
It is unsatisfactory for determination unit, is unsatisfactory for effectiveness condition for determination.
In one embodiment of the invention, the category related information generation module 403 includes:
Frequent item set determines submodule, for being directed to the user with the same objective attribute target attribute, from the category really
Determine frequent item set, the support of the frequent item set is greater than preset support threshold；
Confidence calculations submodule, for calculating the confidence level in the frequent item set between category；
Target category determines submodule, if being greater than preset confidence threshold value for the confidence level, it is determined that the frequency
Category in numerous item collection is target category；
Category related information determines submodule, for determining that the incidence relation between the target category is category association letter
Breath.
In one embodiment of the invention, the frequent item set determines that submodule includes:
Support computing unit, for calculating this frequent item set for the user with the same objective attribute target attribute
Support, the frequent item set is the category for the first time；
Frequent item set filter element, for being less than preset support threshold from support described in the frequent episode concentration filter
The frequent item set of value；
Frequent item set judging unit, for judging whether the filtered frequent item set is empty；If so, calling terminates
Unit, if it is not, then calling frequent item set adding unit；
End unit, for determining that iteration terminates；
Frequent item set adding unit, for increasing the category to the filtered frequent item set, as next time
Frequent item set returns to the support computing unit.
The generating means of category related information provided by the embodiment of the present invention can be performed any embodiment of that present invention and be mentioned
The generation method of the category related information of confession has the corresponding functional module of execution method and beneficial effect.
Embodiment five
Fig. 5 is a kind of structural representation for video push device based on category related information that the embodiment of the present invention four provides
Figure, the device can specifically include following module:
Attribute determination module 501, for determining the attribute of user；
First category determining module 502, for determining that the first category, user's watched belong to first category
Video data；
Category related information searching module 503, for searching the category related information of the Attribute Association, the category is closed
The category joined in information has incidence relation；
Second category determining module 504 has pass with first category for determining in the category related information
The category of connection relationship, as the second category；
Video data pushing module 505, for the video data for belonging to second category to be pushed to the user.
In one embodiment of the invention, the attribute determination module 501 includes:
Registion time determines submodule, for determining the registion time of the user；
Registration time length determines submodule, and for determining registration time length, the registration time length is for the registion time and currently
Difference between time；
Period determines submodule, for determining the period belonging to the registration time length；
User class determines submodule, the attribute for determining the period corresponding user class, as the user.
In one embodiment of the invention, the first category determining module 502 includes:
Operation data determines submodule, for determining operation of the user to the video data for belonging to some the first category
Data；
Effectiveness condition judging submodule, English judge whether the operation data meets preset effectiveness condition；If
It is then to call record sub module, ignores submodule if it is not, then calling；
Record sub module, for the first category described in the user record；
Ignore submodule, for ignoring first category to the user.
In an example of the present embodiment, the operation data determines that submodule includes:
Viewing time cumulative unit, when for accumulating viewing of the user to the video data for belonging to some the first category
Between；
The effectiveness condition judging submodule includes:
Time threshold judging unit, for judging whether the viewing time is greater than preset time threshold；If so, adjusting
With determination unit is met, determination unit is unsatisfactory for if it is not, then calling；
Meet determination unit, meets effectiveness condition for determination；
It is unsatisfactory for determination unit, is unsatisfactory for effectiveness condition for determination.
In one embodiment of the invention, the second category determining module 504 includes:
First quantity statistics submodule, for counting the first quantity of first category；
Category searches submodule, if being less than preset first threshold for first quantity, is associated in the category
The category that there is incidence relation with first category is searched in information, as the second category.
In one embodiment of the invention, the video data pushing module 505 includes:
Video data searches submodule, for searching the video data for belonging to second category, the video data
With the video factor, the video factor include the confidence level of second category, in the temperature of the video data at least
It is a kind of；
Second quantity statistics submodule, for counting the second quantity of the video data；
Target video data selects submodule, if being greater than preset second threshold for second quantity, selects institute
The video data that the video factor meets preset pushing condition is stated, as target video data；
Target video data pushes submodule, for the target video data to be pushed to the user.
The executable present invention of video push device based on category related information provided by the embodiment of the present invention is any real
The video pushing method based on category related information provided by example is applied, has the corresponding functional module of execution method and beneficial to effect
Fruit.
Embodiment six
Fig. 6 is a kind of structural schematic diagram for computer equipment that the embodiment of the present invention six provides.As shown in fig. 6, the calculating
Machine equipment includes processor 600, memory 601, communication module 602, input unit 603 and output device 604；Computer equipment
The quantity of middle processor 600 can be one or more, in Fig. 6 by taking a processor 600 as an example；Processing in computer equipment
Device 600, memory 601, communication module 602, input unit 603 and output device 604 can be connected by bus or other modes
It connects, in Fig. 6 for being connected by bus.
Memory 601 is used as a kind of computer readable storage medium, can be used for storing software program, journey can be performed in computer
Sequence and module, such as the generation method or a kind of view based on category related information of one of the present embodiment category related information
The corresponding module of frequency method for pushing is (for example, the attribute in a kind of generating means of category related information as shown in Figure 4 determines mould
Block 401, category determining module 402, category related information generation module 403 and relating module 404；Alternatively, as shown in Figure 5 one
The attribute determination module 501 of video push device of the kind based on category related information, the first category determining module 502, category are closed
Join information searching module 503, the second category determining module 504 and video data pushing module 505).Processor 600 passes through operation
Storage software program, instruction and module in the memory 601, thereby executing computer equipment various function application and
Data processing realizes that a kind of generation method of above-mentioned category related information or a kind of video based on category related information push away
Delivery method.
Memory 601 can mainly include storing program area and storage data area, wherein storing program area can store operation system
Application program needed for system, at least one function；Storage data area, which can be stored, uses created number according to computer equipment
According to etc..In addition, memory 601 may include high-speed random access memory, it can also include nonvolatile memory, such as extremely
A few disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, memory 601
It can further comprise the memory remotely located relative to processor 600, these remote memories can be by being connected to the network extremely
Computer equipment.The example of above-mentioned network include but is not limited to internet, intranet, local area network, mobile radio communication and its
Combination.
Communication module 602 for establishing connection with display screen, and realizes the data interaction with display screen.Input unit 603
It can be used for receiving the number or character information of input, and generate related with the user setting of computer equipment and function control
Key signals input.
A kind of category association that any embodiment of the present invention provides can be performed in a kind of computer equipment provided in this embodiment
The generation method of information or a kind of video pushing method based on category related information, specific corresponding function and beneficial effect.
Embodiment seven
The embodiment of the present invention seven also provides a kind of computer readable storage medium, is stored thereon with computer program, one
In embodiment, which realizes a kind of generation method of category related information when being executed by processor, this method comprises:
Determine that the attribute of user, the attribute include objective attribute target attribute；
Determine that category, user's watched belong to the video data of the category；
For the user with the same objective attribute target attribute, the target product with incidence relation are determined from the category
Class, as category related information；
The category related information is associated with the objective attribute target attribute.
In another embodiment, a kind of video push based on category related information is realized when which is executed by processor
Method, this method comprises:
Determine the attribute of user；
Determine that the first category, user's watched belong to the video data of first category；
The category related information of the Attribute Association is searched, the category in the category related information has incidence relation；
The determining category with first category with incidence relation in the category related information, as the second product
Class；
The video data for belonging to second category is pushed into the user.
Certainly, computer readable storage medium provided by the embodiment of the present invention, computer program are not limited to institute as above
The method operation stated, can also be performed a kind of generation method or one of category related information provided by any embodiment of the invention
Relevant operation in video pushing method of the kind based on category related information.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to which by hardware realization, but in many cases, the former is more
Good embodiment.Based on this understanding, technical solution of the present invention substantially in other words contributes to the prior art
Part can be embodied in the form of software products, which can store in computer readable storage medium
In, floppy disk, read-only memory (Read-Only Memory, ROM), random access memory (Random such as computer
Access Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are with so that a computer is set
Standby (can be personal computer, server or the network equipment etc.) executes method described in each embodiment of the present invention.
It is worth noting that, the generating means or a kind of view based on category related information of a kind of above-mentioned category related information
In the embodiment of frequency driving means, included each unit and module are only divided according to the functional logic, but not
It is confined to above-mentioned division, as long as corresponding functions can be realized；In addition, the specific name of each functional unit is also only
Convenient for mutually distinguishing, the protection scope that is not intended to restrict the invention.
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (13)
1. a kind of generation method of category related information characterized by comprising
Determine that the attribute of user, the attribute include objective attribute target attribute；
Determine that category, user's watched belong to the video data of the category；
For the user with the same objective attribute target attribute, the target category with incidence relation is determined from the category, is made
For category related information；
The category related information is associated with the objective attribute target attribute.
2. the method according to claim 1, wherein the attribute of the determining user, comprising:
Determine the registion time of the user；
Determine registration time length, difference of the registration time length between the registion time and current time；
Determine the period belonging to the registration time length；
Determine the period corresponding user class, the attribute as the user.
3. the method according to claim 1, wherein the determining category, comprising:
Determine operation data of the user to the video data for belonging to some category；
Judge whether the operation data meets preset effectiveness condition；
If so, to category described in the user record；
If it is not, then ignoring the category to the user.
4. according to the method described in claim 3, it is characterized in that,
Operation data of the determination user to the video data for belonging to some category, comprising:
Accumulate viewing time of the user to the video data for belonging to some category；
It is described to judge whether the operation data meets preset effectiveness condition, comprising:
Judge whether the viewing time is greater than preset time threshold；
If so, determination meets effectiveness condition；
If not, it is determined that be unsatisfactory for effectiveness condition.
5. method according to claim 1-3, which is characterized in that described for the same objective attribute target attribute
User, from the category determine have incidence relation target category, as category related information, comprising:
For the user with the same objective attribute target attribute, frequent item set, the branch of the frequent item set are determined from the category
Degree of holding is greater than preset support threshold；
Calculate the confidence level in the frequent item set between category；
If the confidence level is greater than preset confidence threshold value, it is determined that the category in the frequent item set is target category；
Determine that the incidence relation between the target category is category related information.
6. according to the method described in claim 5, it is characterized in that, it is described for the same objective attribute target attribute user,
Frequent item set is determined from the category, comprising:
For the user with the same objective attribute target attribute, the support of this frequent item set is calculated, the frequent item set is first
Secondary is the category；
It is less than the frequent item set of preset support threshold from support described in the frequent episode concentration filter；
Judge whether the filtered frequent item set is empty；
If so, determining that iteration terminates；
If it is not, then increasing the category as frequent item set next time to the filtered frequent item set returns to the needle
To the user with the same attribute, the support of this frequent item set is calculated.
7. a kind of video pushing method based on category related information characterized by comprising
Determine the attribute of user；
Determine that the first category, user's watched belong to the video data of first category；
The category related information of the Attribute Association is searched, the category in the category related information has incidence relation；
The determining category with first category with incidence relation in the category related information, as the second category；
The video data for belonging to second category is pushed into the user.
8. the method according to the description of claim 7 is characterized in that described determining with described the in the category related information
One category has the category of incidence relation, as the second category, comprising:
Count the first quantity of first category；
If first quantity is less than preset first threshold, searched and first category in the category related information
Category with incidence relation, as the second category.
9. method according to claim 7 or 8, which is characterized in that the video counts that second category will be belonged to
According to pushing to the user, comprising:
The video data for belonging to second category is searched, the video data has the video factor, and the video is because of attached bag
Include at least one of confidence level, the temperature of the video data of second category；
Count the second quantity of the video data；
If second quantity is greater than preset second threshold, the video factor is selected to meet the view of preset pushing condition
Frequency evidence, as target video data；
The target video data is pushed into the user.
10. a kind of generating means of category related information characterized by comprising
Attribute determination module, for determining that the attribute of user, the attribute include objective attribute target attribute；
Category determining module, for determining that category, user's watched belong to the video data of the category；
Category related information generation module, for being determined from the category for the user with the same objective attribute target attribute
Target category with incidence relation, as category related information；
Relating module, for the category related information to be associated with the objective attribute target attribute.
11. a kind of video push device based on category related information characterized by comprising
Attribute determination module, for determining the attribute of user；
First category determining module, for determining that the first category, user's watched belong to the video counts of first category
According to；
Category related information searching module, for searching the category related information of the Attribute Association, the category related information
In category have incidence relation；
Second category determining module has incidence relation with first category for determining in the category related information
Category, as the second category；
Video data pushing module, for the video data for belonging to second category to be pushed to the user.
12. a kind of computer equipment, which is characterized in that the computer equipment includes:
One or more processors；
Memory, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now such as generation method of category related information as claimed in any one of claims 1 to 6, alternatively, such as institute any in claim 7-9
The video pushing method based on category related information stated.
13. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The generation method such as category related information as claimed in any one of claims 1 to 6 is realized when execution, alternatively, such as claim 7-9
In any video pushing method based on category related information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910381032.3A CN110096646A (en) | 2019-05-08 | 2019-05-08 | The generation of category related information and its video pushing method and relevant device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910381032.3A CN110096646A (en) | 2019-05-08 | 2019-05-08 | The generation of category related information and its video pushing method and relevant device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110096646A true CN110096646A (en) | 2019-08-06 |
Family
ID=67447350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910381032.3A Pending CN110096646A (en) | 2019-05-08 | 2019-05-08 | The generation of category related information and its video pushing method and relevant device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110096646A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110659388A (en) * | 2019-10-10 | 2020-01-07 | 北京奇艺世纪科技有限公司 | To-be-recommended information screening method and device, electronic equipment and storage medium |
| CN113468402A (en) * | 2021-05-25 | 2021-10-01 | 北京达佳互联信息技术有限公司 | Target object determination method, device and storage medium |
| CN114025209A (en) * | 2021-10-28 | 2022-02-08 | 广州纵游互娱网络科技有限公司 | Large data stream processing method |
| WO2022247671A1 (en) * | 2021-05-24 | 2022-12-01 | 百果园技术(新加坡)有限公司 | User recall method and apparatus, and computer device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957949A (en) * | 2012-05-18 | 2013-03-06 | 华东师范大学 | Device and method for recommending video to user |
| CN103700005A (en) * | 2013-12-17 | 2014-04-02 | 南京信息工程大学 | Association-rule recommending method based on self-adaptive multiple minimum supports |
| CN107391687A (en) * | 2017-07-24 | 2017-11-24 | 华中师范大学 | A kind of mixing commending system towards local chronicle website |
| CN109426998A (en) * | 2017-08-29 | 2019-03-05 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
-
2019
- 2019-05-08 CN CN201910381032.3A patent/CN110096646A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957949A (en) * | 2012-05-18 | 2013-03-06 | 华东师范大学 | Device and method for recommending video to user |
| CN103700005A (en) * | 2013-12-17 | 2014-04-02 | 南京信息工程大学 | Association-rule recommending method based on self-adaptive multiple minimum supports |
| CN107391687A (en) * | 2017-07-24 | 2017-11-24 | 华中师范大学 | A kind of mixing commending system towards local chronicle website |
| CN109426998A (en) * | 2017-08-29 | 2019-03-05 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
Non-Patent Citations (1)
| Title |
|---|
| 夏林: ""基于Hadoop的视频推荐***的研究与应用"", 《万方》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110659388A (en) * | 2019-10-10 | 2020-01-07 | 北京奇艺世纪科技有限公司 | To-be-recommended information screening method and device, electronic equipment and storage medium |
| WO2022247671A1 (en) * | 2021-05-24 | 2022-12-01 | 百果园技术(新加坡)有限公司 | User recall method and apparatus, and computer device and storage medium |
| CN113468402A (en) * | 2021-05-25 | 2021-10-01 | 北京达佳互联信息技术有限公司 | Target object determination method, device and storage medium |
| CN113468402B (en) * | 2021-05-25 | 2024-05-17 | 北京达佳互联信息技术有限公司 | Target object determining method, device and storage medium |
| CN114025209A (en) * | 2021-10-28 | 2022-02-08 | 广州纵游互娱网络科技有限公司 | Large data stream processing method |
| CN114025209B (en) * | 2021-10-28 | 2023-11-24 | 广州纵游互娱网络科技有限公司 | Big data stream processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2501078C2 (en) | Ranking search results using edit distance and document information | |
| CN110096646A (en) | The generation of category related information and its video pushing method and relevant device | |
| US20150120717A1 (en) | Systems and methods for determining influencers in a social data network and ranking data objects based on influencers | |
| CN103020212B (en) | Method and device for finding hot videos based on user query logs in real time | |
| US20080201297A1 (en) | Method and System for Determining Relation Between Search Terms in the Internet Search System | |
| US20100241647A1 (en) | Context-Aware Query Recommendations | |
| CN111666448B (en) | Search method, search device, electronic equipment and computer readable storage medium | |
| CN104933100A (en) | Keyword recommendation method and device | |
| EP2488970A1 (en) | System and method for grouping multiple streams of data | |
| US10762122B2 (en) | Method and device for assessing quality of multimedia resource | |
| US9268861B2 (en) | Method and system for recommending relevant web content to second screen application users | |
| Li et al. | Probabilistic frequent itemset mining over uncertain data streams | |
| CN105574030A (en) | Information search method and device | |
| CN105447131B (en) | Internet resources relatedness determines method and apparatus | |
| Nasir et al. | Tiptap: approximate mining of frequent k-subgraph patterns in evolving graphs | |
| CN114238360A (en) | User behavior analysis system | |
| CN102915313B (en) | Error correction relation generation method and system in web search | |
| CN105447004A (en) | Mining device for query suggestion words, related query method and device | |
| CN104156492A (en) | Method and device for prompting search content | |
| Chen et al. | Ldp-fpminer: Fp-tree based frequent itemset mining with local differential privacy | |
| CN110866144B (en) | Song retrieval method and device | |
| CN116662645B (en) | Video event tracing analysis method and system, storage medium and electronic equipment | |
| Cao et al. | On application-unbiased benchmarking of web videos from a social network perspective | |
| CN116701737A (en) | Document searching method and device | |
| Fisichella et al. | Event Detection in Wikipedia Edit History Improved by Documents Web Based Automatic Assessment. Big Data Cogn. Comput. 2021, 5, 34 |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190806 |