CN109635153A - Migration path generation method, device and storage medium - Google Patents
Migration path generation method, device and storage medium Download PDFInfo
- Publication number
- CN109635153A CN109635153A CN201811400866.6A CN201811400866A CN109635153A CN 109635153 A CN109635153 A CN 109635153A CN 201811400866 A CN201811400866 A CN 201811400866A CN 109635153 A CN109635153 A CN 109635153A
- Authority
- CN
- China
- Prior art keywords
- classification
- migration path
- classifications
- similarity
- similar categorization
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The embodiment of the invention discloses a kind of migration path generation method, device and storage mediums, belong to field of computer technology.This method comprises: the identical object of information in specified dimension is divided into same category, multiple classifications are obtained, include at least one object in each classification;Feature extraction is carried out to the detail information of the object of each classification respectively, obtains the feature vector of multiple classifications, and then determines similar categorization of each classification in multiple classifications;According to multiple classifications and the similar categorization of each classification, correlation networks are created;According to correlation networks, determines and reach home the migration path that classification passed through from starting point classification, starting point classification and terminal classification are any two classifications in multiple classifications.Information base is provided for subsequent Function Extension, without artificial setting label, lost labor's resource is avoided, reduces human cost, and avoid due to the subjective problem for causing accuracy lower of the label of setting, improve accuracy.
Description
Technical field
The present embodiments relate to field of computer technology, in particular to a kind of migration path generation method, device and deposit
Storage media.
Background technique
Listening to song is a kind of common approach that people enrich the life of the spirit, and user often listens to the singer oneself liked
The song of performance, and over time, the singer that user likes may change.
In the related technology, multiple labels manually can be set for each singer, multiple label can be gender label, year
For label, languages label etc..Category division is carried out according to multiple labels of each singer, multiple classifications are obtained, wherein each class
It Bao Kuo not at least one singer.The broadcasting record of multiple users is obtained, includes that user played in the broadcasting record of each user
Song, the data such as the broadcasting situation of singer and the song of singing the song, according to the broadcasting of each user record and
Classification belonging to the singer in record is played, can determine that the classification of each user, the category refer to the singer institute that user likes
The classification of category, and then the user group of available each classification.
Aforesaid way needs that label manually is arranged for each singer, consumes higher human cost, and set
Label is subjective, causes accuracy lower.
Summary of the invention
The embodiment of the invention provides a kind of migration path generation method, device and storage mediums, can solve related skill
Art there are the problem of.The technical solution is as follows:
On the one hand, a kind of migration path generation method is provided, which comprises
The identical object of information in specified dimension is divided into same category, obtains multiple classifications, in each classification
Including at least one object, and the mark of each classification information of at least one object in the specified dimension
It indicates;
Feature extraction is carried out to the detail information of the object of each classification respectively, obtains the multiple class another characteristic
Vector determines similar categorization of each classification in the multiple classification according to the feature vector of the multiple classification, appoints
Similarity between one classification and similar categorization is greater than the similarity between non-similar categorization;
According to the multiple classification and the similar categorization of each classification, correlation networks, the correlation are created
Network includes the multiple classification, and each classification is associated with corresponding similar categorization;
According to the correlation networks, determines and reach home the migration path that classification passed through from starting point classification, described
Point classification and the terminal classification are any two classifications in the multiple classification.
On the other hand, a kind of migration path generating means are provided, described device includes:
Division module obtains multiple classes for the identical object of information in specified dimension to be divided into same category
It not, include at least one object in each classification, and at least one object described in the mark of each classification is in the finger
The information determined in dimension indicates;
Characteristic extracting module, the detail information for the object respectively to each classification carry out feature extraction, obtain
The feature vector of the multiple classification includes at least one object in each classification;
Similar categorization determining module determines each classification in institute for the feature vector according to the multiple classification
The similar categorization in multiple classifications is stated, the similarity between any classification and similar categorization is greater than the phase between non-similar categorization
Like degree;
Creation module creates correlation net for the similar categorization according to the multiple classification and each classification
Network, the correlation networks include the multiple classification, and each classification is associated with corresponding similar categorization;
Migration path determining module is reached home classification institute for according to the correlation networks, determining from starting point classification
The migration path of process, the starting point classification and the terminal classification are any two classifications in the multiple classification.
On the other hand, the device of migration path generation is provided, described device includes processor and memory, the storage
At least one instruction, at least a Duan Chengxu, code set or instruction set, described instruction, described program, the code are stored in device
Collection or described instruction collection are loaded by the processor and are executed to realize such as possessed behaviour in the migration path generation method
Make.
In another aspect, providing a kind of computer readable storage medium, it is stored in the computer readable storage medium
At least one instruction, at least a Duan Chengxu, code set or instruction set, described instruction, described program, the code set or the finger
Collection is enabled to be loaded by processor and had to realize such as possessed operation in the migration path generation method.
Method, apparatus provided in an embodiment of the present invention and storage medium pass through at least one object to each classification
Detail information carries out feature extraction, obtains the feature vector of multiple classifications, obtains the Similarity Class of each classification according to feature vector
Not, each classification is associated with its similar categorization, creates correlation networks, can be obtained in multiple classifications according to correlation networks
The migration path of any two classifications, and be not only to calculate the migration probability between any two classifications, which can indicate
Other classifications passed through when being migrated between the two classifications, simulate and gradually migrate between two or more classifications
Process, the migration between any two non-similar categorizations provides other classifications for transition, information content improved, after being
Continuous Function Extension provides Information base.Also, without artificial setting label, lost labor's resource is avoided, people is reduced
Power cost, and avoid due to the subjective problem for causing accuracy lower of the label of setting, improve migration path
Accuracy.
Also, during obtaining similar categorization, hyperspace is created according to the multidimensional characteristic vectors of each classification, and
The hyperspace is divided into multiple regions according to position of each classification in the hyperspace, to obtain the phase of each classification
Like classification, characteristic value of each classification in multiple dimensions is used during determining similar categorization, the information of use is more
Comprehensively, the accuracy of similar categorization is improved.Also, hyperspace is created, in institute after progress region division in hyperspace
Belong to and search similar categorization in region and adjacent area, reduces calculation amount, accelerate calculating speed.
Also, the least migration path of classification number for including is chosen from a plurality of migration path, or from a plurality of migration
The smallest migration path of migration distance is chosen in path, finds the most short migration between starting point classification and terminal classification as much as possible
Path is found from starting point class and is clipped to the minimum transition classification passed through when terminal classification migrates, analyzes starting point classification and terminal
Most fast migration pattern between classification.
Also, based on migration path, which can be applied to recommended or to pair of recommendation
As the scene reversely analyzed, reasonable object can be recommended for user, moreover it is possible to it checks out recommend unreasonable situation in time,
It improves and recommends accuracy, expand application range, extend more functions.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the embodiment of the present invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a kind of flow chart of migration path generation method provided in an embodiment of the present invention;
Fig. 2 is a kind of division schematic diagram of hyperspace provided in an embodiment of the present invention;
Fig. 3 is a kind of binary tree schematic diagram provided in an embodiment of the present invention;
Fig. 4 is a kind of schematic diagram of correlation networks provided in an embodiment of the present invention;
Fig. 5 is a kind of structural schematic diagram of adjacency list provided in an embodiment of the present invention;
Fig. 6 is a kind of display interface schematic diagram of migration path provided in an embodiment of the present invention;
Fig. 7 is a kind of structural schematic diagram of migration path generating means provided in an embodiment of the present invention;
Fig. 8 is a kind of structural schematic diagram of terminal provided in an embodiment of the present invention;
Fig. 9 is a kind of structural schematic diagram of server provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Embodiment is described in further detail.
Before the embodiment of the present invention is described in detail, description below is carried out to the concept being related to first:
1, object: can be used as data existing for independent individual, from data format for, object may include audio number
According to the data of, multiple formats such as video data, web data, image data, such as song, film.
Each object have detail information, which includes the information in multiple dimensions, as song include singer, when
The detail informations such as long, performance language, issuing date, film include the detail informations such as director, duration, language, issuing date.
2, classification: whether referring to identical according to the information in specified dimension, to the classification that multiple objects are divided,
It wherein include at least one identical object of information in specified dimension in each classification, and each classification is used in specified dimension
Information on degree indicates.
Such as multiple songs can be divided into multiple classifications, singer's phase of the song of each classification according to the difference of singer
Together, and item name be song name.Alternatively, multiple films can be divided into multiple classifications according to the difference of director,
The director of the film of each classification is identical, and item name is the name directed.
In the related technology, multiple labels manually can be set for each singer, multiple label is used to describe each singer,
Category division is carried out according to multiple labels of each singer, multiple classifications are obtained, wherein each classification includes at least one singer.
The broadcasting record for obtaining multiple users, the classification of each user is determined according to the broadcasting record of each user.According to multiple users
Broadcasting record within multiple periods is counted, and user group of each classification within multiple periods is obtained, will be each
The user group of classification in different time period compares, and obtains the migration probability between any two classifications.Which needs
Label manually is set for each singer, consumes higher human cost, and set label is subjective, is caused really
The accuracy of fixed migration probability is lower.
And the embodiment of the invention provides a kind of migration path generation methods, by least one same category of object
Detail information carry out feature extraction, obtain the feature vector of multiple classifications, the similar of each classification obtained according to feature vector
Each classification is associated with by classification with its similar categorization, creates correlation networks, and then obtain any two classifications in multiple classifications
Migration path, the migration path can indicate other classifications passed through when being migrated between the two classifications, simulate two
The process gradually migrated between a or more than two classifications, improves information content.Also, it without artificial setting label, avoids
Lost labor's resource reduces human cost, and avoid causes accuracy lower since the label of setting is subjective
Problem improves the accuracy of migration path.
The embodiment of the present invention can be applied under the scene for obtaining the migration path between any two classifications.For example, by mutual
A large amount of songs in networking are divided to obtain classification belonging to object, each class according to the singer to give song recitals as object
It is not indicated using singer, then use method provided in an embodiment of the present invention, creation includes the correlation networks of multiple singers, root
The migration path between any two singers can be determined according to the correlation networks, which represents between multiple singers
Style transition process.
Further, according between any two singers migration path and user currently like the singer of song, can be with
Recommend other singers for user.For example, obtaining multiple singers that the user often listens to, sung according to any two in multiple singer
Migration path between hand can recommend the song for other singers for including in the migration path for the user.
Alternatively, being divided to obtain object according to the director of Making Movies using a large amount of films in internet as object
Affiliated classification, each classification are indicated using director, then use method provided in an embodiment of the present invention, and creation is led comprising multiple
The correlation networks drilled can determine the migration path between any two directors, the migration path generation according to the correlation networks
Style route of transition between the multiple directors of table.
Further, according between any two directors migration path and user currently like the director of film, can be with
Recommend other directors for user.For example, obtaining the director for multiple films that the user often watches, appoint according in multiple director
Migration path between two directors can recommend the film for other directors for including in the migration path for the user.
Alternatively, being divided to obtain according to the type of merchandise belonging to commodity using the shiploads of merchandise in internet as object
Classification belonging to object, each classification are indicated using the type of merchandise, then use method provided in an embodiment of the present invention, creation packet
Correlation networks containing multiple type of merchandises can determine the migration road between any two type of merchandises according to the correlation networks
Diameter, the migration path represent the route of transition between multiple type of merchandises.
Further, according between any two type of merchandises migration path and user currently often buy commodity class
Type can recommend other types of merchandise for user.For example, multiple type of merchandises that the user often buys are obtained, according to this
Migration path in multiple type of merchandises between any two type of merchandises, can for the user recommend include in the migration path
The commodity of other type of merchandises.
Fig. 1 is a kind of flow chart of migration path generation method provided in an embodiment of the present invention.The embodiment of the present invention is held
Row main body is generating means, referring to Fig. 1, this method comprises:
101, the identical object of information in specified dimension is divided into same category by generating means, obtains multiple classes
Not, feature extraction is carried out to the detail information of the object of each classification respectively, obtains the feature vector of multiple classifications.
The generating means can be terminal or server.When generating means are servers, server can be collected more
Multiple objects of a terminal access and the detail information of multiple object, or collect multiple objects that multiple publishers upload
And the detail information of multiple object, the object being collected into and its detail information are stored by server, and to storage
Detail information divided and counted, obtain the feature vector of multiple classifications.Alternatively, when generating means are terminals, terminal
The detail information for receiving multiple objects and multiple object that server is sent, by multiple object and multiple object
Detail information is stored, and the detail information of storage is divided and counted, and obtains the feature vector of multiple classifications.
In the embodiment of the present invention, each object has detail information, the detail information packet of each object in multiple objects
The information for including multiple dimensions, for describing corresponding object from multiple dimensions.Detail information according to each object is specified
Information in dimension, multiple object can be divided into different classes of, multiple classifications be obtained, wherein being included in each classification
At least one identical object of information in specified dimension, and the mark of each classification at least one object in specified dimension
On information indicate.
The specified dimension is any dimension in multiple dimension, can be determined according to the demand of user, can also be according to
Popular degree of the information in internet in each dimension, the dimension that user is paid close attention to the most are determined as specified dimension.
For example, the detail information of each song may include singer, duration, sing language, distribution when object is song
The detail informations such as time.For multiple song, if dividing according to singer, available multiple classifications, each classification is used
One singer indicates, at least one song that at least one object characterization of each classification is sung by the singer.
Alternatively, the detail information of each commodity may include the color of the commodity, valence when multiple objects are multiple commodity
The information such as lattice, model, manufacturer and the type of merchandise.For multiple commodity, if being divided according to the type of merchandise, can obtain
To multiple classifications, a kind of each type of merchandise of categorized representation, at least one object characterization of each classification belongs to every kind of commodity class
At least one commodity of type.
Generating means obtain the detail information of each object, and carry out feature extraction to the detail information of each object,
The feature vector of each object is obtained, the feature vector of each object is used to describe the feature of each object.Due to same
Classification includes at least one object, thus generating means respectively to belong to the feature vector of at least one same category of object into
Row statistics, obtains the feature vector of each classification in multiple classifications.
In one possible implementation, the details of generating means also at least one object of available each classification
Information, and at least one phrase for including in the detail information of each object is obtained, for each phrase, can be turned using default
Scaling method carries out word to the conversion of term vector, obtains the corresponding term vector of the phrase, so that it is corresponding to obtain at least one phrase
At least one term vector, by least one term vector combine constitute object feature vector.Wherein, the default transfer algorithm
It can be word2vec (word converting vector) algorithm or other algorithms.
In alternatively possible implementation, generating means get the feature of at least one other object of any sort to
After amount, the feature vector of at least one object is subjected to mean value computation, using the vector obtained after mean value computation as the category
Feature vector.Each classification is calculated respectively according to identical calculation, obtains the feature vector of each classification.
102, generating means calculate the similarity between any two classifications according to the feature vector of multiple classifications.
In the embodiment of the present invention, after generating means get the feature vector of multiple classifications, according to multiple class another characteristics
Vector, the feature vector by the feature vector of each classification respectively with other classifications other than the category compare, can
To obtain the similarity between any two classifications, and then according to obtained similarity, the similar categorization of each classification is determined.
Wherein, the similarity between any two classifications represents the similarity degree between any two classifications, and similarity is higher, this
Two classifications are more similar, and similarity is lower, the two classifications are more dissimilar.
In one possible implementation, after generating means get the feature vector of multiple classifications, by each class
Another characteristic vector carries out dot product calculating with the feature vector of other classifications respectively, obtains between each classification and other classifications
Similarity, and then obtain the similarity in multiple classifications between any two classifications.
In addition to the mode for carrying out dot product calculating, it can also use and calculate Euclidean distance, calculate the modes such as cosine similarity,
Calculate the similarity between any two classifications.
103, for each classification, generating means are according to other classes in the category and multiple classifications other than the category
Similarity between not, determines similar categorization of the category in multiple classifications.
Wherein, the similarity between any classification and similar categorization is greater than the similarity between non-similar categorization.
It, can be according to the category and other classifications for each classification in order to determine the similarity degree between each classification
Between similarity, by these other category divisions be the category similar categorization and non-similar categorization.Wherein, the category and phase
It is greater than the similarity between the category and non-similar categorization like the similarity between classification.
In a kind of possible implementation, the detail information of each object includes information of the object in multiple dimensions,
The feature vector of each classification also includes characteristic value of the category in multiple dimensions, then can comprehensively consider each classification more
Characteristic value in a dimension, to determine the similar categorization of each classification.
For this purpose, as shown in Fig. 2, generating means create hyperspace according to multiple dimension, according to each class another characteristic
Characteristic value of the vector in multiple dimensions in each dimension determines the position of each classification in the hyperspace, according to more
The hyperspace is divided into multiple regions, includes at least one in each region by position of a classification in the hyperspace
Classification.For each classification, using the adjacent area of category affiliated area and affiliated area as the alternative area of the category,
The similarity between other classifications in the category and alternative area in addition to the category is obtained respectively, so that it is determined that the category exists
Similar categorization in alternative area.
For example, generating means can carry out region division to hyperspace using default partitioning algorithm.Wherein, this default stroke
Dividing algorithm can be a kind of KD tree (data structure for dividing k dimension data space) algorithm or other algorithms.As shown in Figures 2 and 3,
KD tree algorithm constructs a binary tree according to hyperspace, the area in a node on behalf hyperspace on binary tree
Domain can be that the region corresponding node on binary tree generates son section when dividing every time to any region of hyperspace
Point, until each leaf node on binary tree can represent a minimum unit when region division of hyperspace is completed
A label is added for each leaf node in region, which can represent the quantity of classification in corresponding region.It is calculated using KD tree
Method may be implemented quickly to divide, and accelerate to divide speed, and then accelerate the speed of determining similar categorization.
In one possible implementation, for each classification, according to the similarity between the category and other classifications,
Using the maximum classification of the similarity in alternative area between classification as the similar categorization of the category.
In alternatively possible implementation, for each classification, according to similar between the category and other classifications
Degree arranges these other classifications according to the sequence from big to small of the similarity between the category, and according to put in order from
The middle classification for choosing preset quantity, as similar categorization.
In alternatively possible implementation, the similarity between the category and other classifications is obtained, other classifications are pressed
After the sequence of the similarity between the category from big to small, the similarity list of the category, the similarity can be created
Other classifications in list comprising arranging in sequence.Then can be according to the similarity list, choosing first classification conduct should
The similar categorization of classification, alternatively, choosing similar categorization of the classification of preset quantity as the category in sequence.
Certainly, generating means can also determine the similar categorization of each classification using other modes.
104, generating means create correlation networks according to multiple classifications and the similar categorization of each classification.
Generating means get the similar categorization of each classification, and the similar categorization of each classification and the category is closed
Connection, create correlation networks, then the correlation networks include multiple classification, and each classification with the similar categorization of the category
Association, which can be as shown in Figure 4.
In one possible implementation, after getting above-mentioned correlation networks, Protobuf (one can be used
The format of kind data exchange) creation adjacency list, which is stored as by graphic network with the structure of adjacency list, according to this
Graphic network can visually see the degree of correlation between each classification.
Wherein, at least one similar categorization of each classification is stored in adjacency list, and at least one similar categorization is pressed
It is arranged according to sequence.For example, the adjacency list can be as shown in figure 5, the similar categorization that classification A is arranged in sequence be respectively classification B
With classification D, the similar categorization that classification B is arranged in sequence is respectively classification A, classification C and classification D, and the similar categorization of classification C is
Classification B, the similar categorization that classification D is arranged in sequence are respectively classification A and classification B.
105, generating means are according to correlation networks, determine and reach home the migration road that classification passed through from starting point classification
Diameter.
Wherein, above-mentioned starting point classification and terminal classification are any two classifications in multiple classifications, are at least wrapped on migration path
Starting point classification and terminal classification the two classifications are included, certainly can also include other classes in addition to starting point classification and terminal classification
Not.
Optionally, generating means can use preset algorithm zequin class after determining starting point classification and terminal classification
The migration path that classification of not reaching home is passed through, wherein the preset algorithm can be Dijkstra (Di Jiesitela) algorithm
Or other algorithms, wherein Dijkstra is a kind of signal source shortest path algorithm, for calculating a node to other all nodes
Shortest path.
In one possible implementation, generating means can be looked into starting point classification using dijkstra's algorithm
It looks for, until finding terminal classification, migration path lookup at this time terminates, and generating means, which obtained, to be clipped to terminal from starting point class
The migration path of classification and preservation.
In alternatively possible implementation, generating means can be looked into terminal classification using dijkstra's algorithm
It looks for, until finding starting point classification, migration path lookup at this time terminates, and generating means, which obtained, to be clipped to terminal from starting point class
The migration path of classification and preservation.
In alternatively possible implementation, generating means can use simultaneously in starting point classification and terminal classification
Dijkstra's algorithm is searched, until the lookup carried out from the starting point classification and the terminal classification is in a certain middle category phase
Meet, at this time migration path lookup terminate, generating means obtain from the starting point class be clipped to the migration path between the middle category with
And the middle category is saved as to the migration path of the terminal classification, and by the migration path that two migration paths collectively form
The migration path of the terminal classification is clipped to from the starting point class.
It reaches home a plurality of migration path that classification passed through if determined in correlation networks from starting point classification,
Any bar migration path is chosen from a plurality of migration path.
In one possible implementation, choose that the classification number for including is least to move from above-mentioned a plurality of migration path
Move path.Alternatively, choosing the smallest migration path of migration distance from above-mentioned a plurality of migration path.Wherein, any two classifications it
Between migration distance and any two classifications between similarity negative correlation, i.e. similarity is higher, and migration distance is shorter.
For example, being multiple songs from the starting point classification migration path that classification is passed through of reaching home when classification is singer
Migration path in hand between any two singers, as shown in fig. 6, being shown on migration path display interface for search starting point class
Not and the search column of terminal classification, and " search " button, user can respectively in two search columns input starting point classification with
Terminal classification, the migration path being clipped between terminal classification from starting point class can be shown by then clicking " search " button.If starting point
Classification is singer A, and terminal classification is singer F, then runs dijkstra's algorithm simultaneously in starting point classification and terminal classification and looked into
It looks for, has obtained the migration path of " singer A → singer B → singer C → singer D → singer E → singer F ", then the migration path is
For from singer A to the migration path of singer F.
The embodiment of the present invention can be applied under the scene for showing migration path for user, and user can be in generating means
Starting point classification and terminal classification are set, is generated by generating means from starting point classification and is reached home the migration path that classification passed through
Afterwards, user is showed, user is it can be learnt that involved in transition process and transition process between starting point classification and terminal classification
Other classifications arrived provide a kind of novel exhibition method, enhance interest.
For example, when object be song, classification be singer when, user be arbitrarily arranged the very big singer A of two stylistic differences and
The migration path between singer A and singer B then can be generated in B, user you can learn that between singer A and singer B it is existing other
Singer understands the style evolution process of singer A to singer B.
In addition, the embodiment of the present invention can also be applied under the scene for user's recommended, determining from starting point classification
To terminal classification by migration path after, the list of object to be recommended can also be established according to the migration path, and should
User is recommended in list.Wherein, which also includes other than including the object in starting point classification and terminal classification
There is at least one object in other classifications on the migration path between starting point classification and terminal classification.
For example, when classification is singer, the multiple singers currently liked according to user can be user when object is song
Recommend in multiple singer, the song that other singers on the migration path of any two singers are sung.Alternatively, when object is electricity
Shadow, when classification is director, the multiple directors currently liked according to user can recommend in multiple director for user, any two
Film made by other directors on the migration path of director.Alternatively, when object is commodity, when classification is the type of merchandise, root
The type of merchandise currently often bought according to user can recommend in multiple type of merchandise for user, any two type of merchandises
The commodity that other type of merchandises on migration path are included.
In addition, the embodiment of the present invention can also be applied under the scene reversely analyzed the object of recommendation, using one
A little proposed algorithms are that can use method provided in an embodiment of the present invention after user recommends object, obtain pair from recommendation
The migration path being clipped between a certain classification that user likes as affiliated class can be with according to the migration distance of the migration path
Whether reasonable analyze recommended object, if meet the hobby of user, the situation for recommending mistake can be checked out in time, from
And realize reversed analysis.
For example, when object is song, it is current according to user when recommending a certain song for user when classification is singer
Migration path between the singer liked and the singer for singing the recommendation song, it can be determined that whether the singer of recommendation is reasonable.
Method provided in an embodiment of the present invention carries out feature by the detail information of at least one object to each classification
It extracts, obtains the feature vector of multiple classifications, the similar categorization of each classification is obtained according to feature vector, by each classification and its
Similar categorization is associated, and is created correlation networks, can be obtained any two classifications in multiple classifications according to correlation networks
Migration path, and be not only to calculate the migration probability between any two classifications, which can indicate in the two classifications
Between other classifications for passing through when being migrated, the process gradually migrated between two or more classifications is simulated, to appoint
Migration between two non-similar categorizations provides other classifications for transition, improves information content, expands for subsequent function
Exhibition provides Information base.Also, without artificial setting label, lost labor's resource is avoided, reduces human cost, and keep away
Exempt to improve the accuracy of migration path due to the subjective problem for causing accuracy lower of the label of setting.
Also, during obtaining similar categorization, hyperspace is created according to the multidimensional characteristic vectors of each classification, and
The hyperspace is divided into multiple regions according to position of each classification in the hyperspace, to obtain the phase of each classification
Like classification, characteristic value of each classification in multiple dimensions is used during determining similar categorization, the information of use is more
Comprehensively, the accuracy of similar categorization is improved.Also, hyperspace is created, in institute after progress region division in hyperspace
Belong to and search similar categorization in region and adjacent area, reduces calculation amount, accelerate calculating speed.
Also, the least migration path of classification number for including is chosen from a plurality of migration path, or from a plurality of migration
The smallest migration path of migration distance is chosen in path, finds the most short migration between starting point classification and terminal classification as much as possible
Path is found from starting point class and is clipped to the minimum transition classification passed through when terminal classification migrates, analyzes starting point classification and terminal
Most fast migration pattern between classification.
Also, based on migration path, which can be applied to recommended or to pair of recommendation
As the scene reversely analyzed, reasonable object can be recommended for user, moreover it is possible to it checks out recommend unreasonable situation in time,
It improves and recommends accuracy, expand application range, extend more functions.
Fig. 7 is a kind of structural schematic diagram of migration path generating means provided in an embodiment of the present invention, referring to Fig. 7, the dress
It sets and includes:
The identical object of information in specified dimension is divided by division module 701 for executing in above-described embodiment
Same category, the step of obtaining multiple classifications;
Characteristic extracting module 702, for execute in above-described embodiment respectively to the detail information of the object of each classification into
Row feature extraction, the step of obtaining the feature vector of multiple classifications;
Similar categorization determining module 703 determines every for executing in above-described embodiment according to the feature vector of multiple classifications
The step of similar categorization of a classification in multiple classifications;
Creation module 704, for executing in above-described embodiment according to multiple classifications and the similar categorization of each classification, wound
The step of building correlation networks;
Migration path determining module 705 is determined for executing in above-described embodiment according to correlation networks from starting point classification
The step of migration path that classification of reaching home is passed through.
Optionally, characteristic extracting module 702, comprising:
Feature extraction unit carries out feature extraction to the detail information of each object for executing in above-described embodiment, obtains
To each object feature vector the step of;
Statistic unit, for execute in above-described embodiment respectively to the feature vector of at least one object of each classification into
The step of row counts, and obtains the feature vector of multiple classifications.
Optionally, similar categorization determining module 703, comprising:
Computing unit, for executing in above-described embodiment according to the feature vector of multiple classifications, calculate any two classifications it
Between similarity the step of;
Similar categorization determination unit, for executing for each classification in above-described embodiment, according to classification and multiple classifications
In similarity between other classifications in addition to classification, the step of determining similar categorization of the classification in multiple classifications.
Optionally, the detail information of object includes information of the object in multiple dimensions, then the feature vector of classification includes
Characteristic value of the classification in multiple dimensions;
Similar categorization determination unit creates hyperspace according to multiple dimensions for executing in above-described embodiment, more based on this
Dimension space determines the step of similar categorization of the classification in alternative area.
Optionally, similar categorization determination unit is also used to execute in above-described embodiment according to each in classification and alternative area
Similarity between classification, using the maximum classification of the similarity in alternative area between classification as similar categorization;Alternatively, pressing
According to the sequence of the similarity between classification from big to small, the classification of preset quantity is chosen from alternative area, as Similarity Class
Other step;
Optionally, migration path determining module 705, comprising:
Migration path determination unit, for executing according to correlation networks in above-described embodiment, determination is clipped to from starting point class
The step of a plurality of migration path passed through up to terminal classification;
Selection unit, it is least for executing the classification number chosen from a plurality of migration path in above-described embodiment and include
Migration path;Alternatively, the smallest migration path of migration distance is chosen from a plurality of migration path, the migration between any two classifications
The step of similarity negative correlation between distance and any two classifications.
It should be understood that migration path generating means provided by the above embodiment are when generating migration path, only more than
The division progress of each functional module is stated for example, can according to need and in practical application by above-mentioned function distribution by difference
Functional module complete, i.e., the internal structure of generating means is divided into different functional modules, it is described above complete to complete
Portion or partial function.In addition, migration path generating means provided by the above embodiment and migration path generation method embodiment
Belong to same design, specific implementation process is detailed in embodiment of the method, and which is not described herein again.
Fig. 8 shows the structural schematic diagram of the terminal 800 of an illustrative embodiment of the invention offer.The terminal 800 can
To be portable mobile termianl, such as: smart phone, tablet computer, MP3 player (Moving Picture Experts
Group Audio Layer III, dynamic image expert's compression standard audio level 3), MP4 (Moving Picture
Experts Group Audio Layer IV, dynamic image expert's compression standard audio level 4) player, laptop,
Desktop computer, headset equipment or any other intelligent terminal.Terminal 800 is also possible to referred to as user equipment, portable terminal
Other titles such as end, laptop terminal, terminal console.
In general, terminal 800 includes: processor 801 and memory 802.
Processor 801 may include one or more processing cores, such as 4 core processors, 8 core processors etc..Place
Reason device 801 can use DSP (Digital Signal Processing, Digital Signal Processing), FPGA (Field-
Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array, may be programmed
Logic array) at least one of example, in hardware realize.Processor 801 also may include primary processor and coprocessor, master
Processor is the processor for being handled data in the awake state, also referred to as CPU (Central Processing
Unit, central processing unit);Coprocessor is the low power processor for being handled data in the standby state.?
In some embodiments, processor 801 can be integrated with GPU (Graphics Processing Unit, image processor),
GPU is used to be responsible for the rendering and drafting of content to be shown needed for display screen.In some embodiments, processor 801 can also be wrapped
AI (Artificial Intelligence, artificial intelligence) processor is included, the AI processor is for handling related machine learning
Calculating operation.
Memory 802 may include one or more computer readable storage mediums, which can
To be non-transient.Memory 802 may also include high-speed random access memory and nonvolatile memory, such as one
Or multiple disk storage equipments, flash memory device.In some embodiments, the non-transient computer in memory 802 can
Storage medium is read for storing at least one instruction, at least one instruction by processor 801 for being had to realize this Shen
Please in embodiment of the method provide migration path generation method.
In some embodiments, terminal 800 is also optional includes: peripheral device interface 803 and at least one peripheral equipment.
It can be connected by bus or signal wire between processor 801, memory 802 and peripheral device interface 803.Each peripheral equipment
It can be connected by bus, signal wire or circuit board with peripheral device interface 803.Specifically, peripheral equipment includes: radio circuit
804, at least one of touch display screen 805, camera 806, voicefrequency circuit 807, positioning component 808 and power supply 809.
Peripheral device interface 803 can be used for I/O (Input/Output, input/output) is relevant outside at least one
Peripheral equipment is connected to processor 801 and memory 802.In some embodiments, processor 801, memory 802 and peripheral equipment
Interface 803 is integrated on same chip or circuit board;In some other embodiments, processor 801, memory 802 and outer
Any one or two in peripheral equipment interface 803 can realize on individual chip or circuit board, the present embodiment to this not
It is limited.
Radio circuit 804 is for receiving and emitting RF (Radio Frequency, radio frequency) signal, also referred to as electromagnetic signal.It penetrates
Frequency circuit 804 is communicated by electromagnetic signal with communication network and other communication equipments.Radio circuit 804 turns electric signal
It is changed to electromagnetic signal to be sent, alternatively, the electromagnetic signal received is converted to electric signal.Optionally, radio circuit 804 wraps
It includes: antenna system, RF transceiver, one or more amplifiers, tuner, oscillator, digital signal processor, codec chip
Group, user identity module card etc..Radio circuit 804 can be carried out by least one wireless communication protocol with other terminals
Communication.The wireless communication protocol includes but is not limited to: Metropolitan Area Network (MAN), each third generation mobile communication network (2G, 3G, 4G and 8G), wireless office
Domain net and/or WiFi (Wireless Fidelity, Wireless Fidelity) network.In some embodiments, radio circuit 804 may be used also
To include the related circuit of NFC (Near Field Communication, wireless near field communication), the application is not subject to this
It limits.
Display screen 805 is for showing UI (User Interface, user interface).The UI may include figure, text, figure
Mark, video and its their any combination.When display screen 805 is touch display screen, display screen 805 also there is acquisition to show
The ability of the touch signal on the surface or surface of screen 805.The touch signal can be used as control signal and be input to processor
801 are handled.At this point, display screen 805 can be also used for providing virtual push button and/or dummy keyboard, also referred to as soft button and/or
Soft keyboard.In some embodiments, display screen 805 can be one, and the front panel of terminal 800 is arranged;In other embodiments
In, display screen 805 can be at least two, be separately positioned on the different surfaces of terminal 800 or in foldover design;In still other reality
It applies in example, display screen 805 can be flexible display screen, be arranged on the curved surface of terminal 800 or on fold plane.Even, it shows
Display screen 805 can also be arranged to non-rectangle irregular figure, namely abnormity screen.Display screen 805 can use LCD (Liquid
Crystal Display, liquid crystal display), OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode)
Etc. materials preparation.
CCD camera assembly 806 is for acquiring image or video.Optionally, CCD camera assembly 806 include front camera and
Rear camera.In general, the front panel of terminal is arranged in front camera, the back side of terminal is arranged in rear camera.One
In a little embodiments, rear camera at least two is main camera, depth of field camera, wide-angle camera, focal length camera shooting respectively
Any one in head, to realize that main camera and the fusion of depth of field camera realize background blurring function, main camera and wide-angle
Camera fusion realizes that pan-shot and VR (Virtual Reality, virtual reality) shooting function or other fusions are clapped
Camera shooting function.In some embodiments, CCD camera assembly 806 can also include flash lamp.Flash lamp can be monochromatic warm flash lamp,
It is also possible to double-colored temperature flash lamp.Double-colored temperature flash lamp refers to the combination of warm light flash lamp and cold light flash lamp, can be used for not
With the light compensation under colour temperature.
Voicefrequency circuit 807 may include microphone and loudspeaker.Microphone is used to acquire the sound wave of user and environment, and will
Sound wave, which is converted to electric signal and is input to processor 801, to be handled, or is input to radio circuit 804 to realize voice communication.
For stereo acquisition or the purpose of noise reduction, microphone can be separately positioned on the different parts of terminal 800 to be multiple.Mike
Wind can also be array microphone or omnidirectional's acquisition type microphone.Loudspeaker is then used to that processor 801 or radio circuit will to be come from
804 electric signal is converted to sound wave.Loudspeaker can be traditional wafer speaker, be also possible to piezoelectric ceramic loudspeaker.When
When loudspeaker is piezoelectric ceramic loudspeaker, the audible sound wave of the mankind can be not only converted electrical signals to, it can also be by telecommunications
Number the sound wave that the mankind do not hear is converted to carry out the purposes such as ranging.In some embodiments, voicefrequency circuit 807 can also include
Earphone jack.
Positioning component 808 is used for the current geographic position of positioning terminal 800, to realize navigation or LBS (Location
Based Service, location based service).Positioning component 808 can be the GPS (Global based on the U.S.
Positioning System, global positioning system), the dipper system of China, Russia Gray receive this system or European Union
The positioning component of Galileo system.
Power supply 809 is used to be powered for the various components in terminal 800.Power supply 809 can be alternating current, direct current,
Disposable battery or rechargeable battery.When power supply 809 includes rechargeable battery, which can support wired charging
Or wireless charging.The rechargeable battery can be also used for supporting fast charge technology.
In some embodiments, terminal 800 further includes having one or more sensors 810.The one or more sensors
810 include but is not limited to: acceleration transducer 811, gyro sensor 812, pressure sensor 813, fingerprint sensor 814,
Optical sensor 815 and proximity sensor 816.
The acceleration that acceleration transducer 811 can detecte in three reference axis of the coordinate system established with terminal 800 is big
It is small.For example, acceleration transducer 811 can be used for detecting component of the acceleration of gravity in three reference axis.Processor 801 can
With the acceleration of gravity signal acquired according to acceleration transducer 811, touch display screen 805 is controlled with transverse views or longitudinal view
Figure carries out the display of user interface.Acceleration transducer 811 can be also used for the acquisition of game or the exercise data of user.
Gyro sensor 812 can detecte body direction and the rotational angle of terminal 800, and gyro sensor 812 can
To cooperate with acquisition user to act the 3D of terminal 800 with acceleration transducer 811.Processor 801 is according to gyro sensor 812
Following function may be implemented in the data of acquisition: when action induction (for example changing UI according to the tilt operation of user), shooting
Image stabilization, game control and inertial navigation.
The lower layer of side frame and/or touch display screen 805 in terminal 800 can be set in pressure sensor 813.Work as pressure
When the side frame of terminal 800 is arranged in sensor 813, user can detecte to the gripping signal of terminal 800, by processor 801
Right-hand man's identification or prompt operation are carried out according to the gripping signal that pressure sensor 813 acquires.When the setting of pressure sensor 813 exists
When the lower layer of touch display screen 805, the pressure operation of touch display screen 805 is realized to UI circle according to user by processor 801
Operability control on face is controlled.Operability control includes button control, scroll bar control, icon control, menu
At least one of control.
Fingerprint sensor 814 is used to acquire the fingerprint of user, collected according to fingerprint sensor 814 by processor 801
The identity of fingerprint recognition user, alternatively, by fingerprint sensor 814 according to the identity of collected fingerprint recognition user.It is identifying
When the identity of user is trusted identity out, authorize the user that there is relevant sensitive operation, the sensitive operation packet by processor 801
Include solution lock screen, check encryption information, downloading software, payment and change setting etc..Terminal can be set in fingerprint sensor 814
800 front, the back side or side.When being provided with physical button or manufacturer Logo in terminal 800, fingerprint sensor 814 can be with
It is integrated with physical button or manufacturer's mark.
Optical sensor 815 is for acquiring ambient light intensity.In one embodiment, processor 801 can be according to optics
The ambient light intensity that sensor 815 acquires controls the display brightness of touch display screen 805.Specifically, when ambient light intensity is higher
When, the display brightness of touch display screen 805 is turned up;When ambient light intensity is lower, the display for turning down touch display screen 805 is bright
Degree.In another embodiment, the ambient light intensity that processor 801 can also be acquired according to optical sensor 815, dynamic adjust
The acquisition parameters of CCD camera assembly 806.
Proximity sensor 816, also referred to as range sensor are generally arranged at the front panel of terminal 800.Proximity sensor 816
For acquiring the distance between the front of user Yu terminal 800.In one embodiment, when proximity sensor 816 detects use
When family and the distance between the front of terminal 800 gradually become smaller, touch display screen 805 is controlled from bright screen state by processor 801
It is switched to breath screen state;When proximity sensor 816 detects user and the distance between the front of terminal 800 becomes larger,
Touch display screen 805 is controlled by processor 801 and is switched to bright screen state from breath screen state.
It will be understood by those skilled in the art that the restriction of the not structure paired terminal 800 of structure shown in Fig. 8, can wrap
It includes than illustrating more or fewer components, perhaps combine certain components or is arranged using different components.
Fig. 9 is a kind of structural schematic diagram of server provided in an embodiment of the present invention, which can be because of configuration or property
Energy is different and generates bigger difference, may include one or more processors (central processing
Units, CPU) 901 and one or more memory 902, wherein at least one finger is stored in the memory 902
It enables, at least one instruction is loaded by the processor 901 and executed the side to realize above-mentioned each embodiment of the method offer
Method.Certainly, which can also have the components such as wired or wireless network interface, keyboard and input/output interface, so as to
Input and output are carried out, which can also include other for realizing the component of functions of the equipments, and this will not be repeated here.
Server 900 can be used for executing step performed by generating means in above-mentioned migration path generation method.
The embodiment of the invention also provides a kind of migration path generating means, which includes processor and memory, is deposited
At least one instruction, at least a Duan Chengxu, code set or instruction set, instruction, program, code set or instruction set are stored in reservoir
Loaded by processor and had possessed operation in the migration path generation method to realize above-described embodiment.
The embodiment of the invention also provides a kind of computer readable storage medium, stored in the computer readable storage medium
Have at least one instruction, at least a Duan Chengxu, code set or instruction set, the instruction, the program, the code set or the instruction set by
Processor loads and has possessed operation in the migration path generation method to realize above-described embodiment.
Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware
It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The foregoing is merely the preferred embodiments of the embodiment of the present invention, are not intended to limit the invention embodiment, all at this
Within the spirit and principle of inventive embodiments, any modification, equivalent replacement, improvement and so on be should be included in of the invention
Within protection scope.
Claims (14)
1. a kind of migration path generation method, which is characterized in that the described method includes:
The identical object of information in specified dimension is divided into same category, multiple classifications is obtained, includes in each classification
At least one object, and the mark of each classification information table of at least one object in the specified dimension
Show;
Feature extraction is carried out to the detail information of the object of each classification respectively, obtain the multiple class another characteristic to
Amount, according to the feature vector of the multiple classification, determines similar categorization of each classification in the multiple classification, any
Similarity between classification and similar categorization is greater than the similarity between non-similar categorization;
According to the multiple classification and the similar categorization of each classification, correlation networks, the correlation networks are created
Comprising the multiple classification, and each classification is associated with corresponding similar categorization;
According to the correlation networks, determines and reach home the migration path that classification passed through from starting point classification, the starting point class
Not and the terminal classification is any two classifications in the multiple classification.
2. the method according to claim 1, wherein described respectively believe the details of the object of each classification
Breath carries out feature extraction, obtains the feature vector of the multiple classification, comprising:
Feature extraction is carried out to the detail information of each object, obtains the feature vector of each object;
The feature vector of at least one object of each classification is counted respectively, obtains the multiple class another characteristic
Vector.
3. the method according to claim 1, wherein the feature vector according to the multiple classification, determines
Similar categorization of each classification in the multiple classification, comprising:
According to the feature vector of the multiple classification, the similarity between any two classifications is calculated;
For each classification, according between other classifications in the classification and the multiple classification in addition to the classification
Similarity, determine similar categorization of the classification in the multiple classification.
4. according to the method described in claim 3, it is characterized in that, the detail information of the object includes the object multiple
Information in dimension, then the feature vector of the classification includes characteristic value of the classification in the multiple dimension;
Similarity between described other classifications according in the classification and the multiple classification in addition to the classification determines
Similar categorization of the classification in the multiple classification, comprising:
Hyperspace is created according to the multiple dimension, according to the feature vector of each classification in the multiple dimension
Characteristic value determines position of each classification in the hyperspace;
According to position of the multiple classification in the hyperspace, the hyperspace is divided into multiple regions;
For each classification, by the classification affiliated area and the adjacent area in the region alternately region, root
According to the similarity between each classification in the classification and the alternative area, determine the classification in the alternative area
Similar categorization.
5. according to the method described in claim 4, it is characterized in that, described according to each in the classification and the alternative area
Similarity between classification determines similar categorization of the classification in the alternative area, comprising:
According to the similarity in the classification and the alternative area between each classification, by the alternative area with the class
The maximum classification of similarity between not is as the similar categorization;Alternatively,
According to the sequence of the similarity between the classification from big to small, the class of preset quantity is chosen from the alternative area
Not, as the similar categorization.
6. determining from starting point class the method according to claim 1, wherein described according to the correlation networks
The migration path that classification of not reaching home is passed through, comprising:
According to the correlation networks, determines and reach home a plurality of migration path that classification passed through from starting point classification;
The least migration path of classification number for including is chosen from a plurality of migration path;Alternatively,
The smallest migration path of migration distance is chosen from a plurality of migration path, the migration distance between any two classifications with
Similarity negative correlation between any two described classifications.
7. a kind of migration path generating means, which is characterized in that described device includes:
Division module obtains multiple classifications, often for the identical object of information in specified dimension to be divided into same category
It include at least one object in a classification, and at least one object described in the mark of each classification is in the specified dimension
On information indicate;
Characteristic extracting module, the detail information for the object respectively to each classification carry out feature extraction, obtain described
The feature vector of multiple classifications includes at least one object in each classification;
Similar categorization determining module determines each classification described more for the feature vector according to the multiple classification
Similar categorization in a classification, the similarity between any classification and similar categorization are greater than similar between non-similar categorization
Degree;
Creation module creates correlation networks, institute for the similar categorization according to the multiple classification and each classification
Stating correlation networks includes the multiple classification, and each classification is associated with corresponding similar categorization;
Migration path determining module is passed through for according to the correlation networks, determining from starting point classification classification of reaching home
Migration path, the starting point classification and the terminal classification are any two classifications in the multiple classification.
8. device according to claim 7, which is characterized in that the characteristic extracting module, comprising:
Feature extraction unit carries out feature extraction for the detail information to each object, obtains each object
Feature vector;
Statistic unit, the feature vector at least one object respectively to each classification count, and obtain described
The feature vector of multiple classifications.
9. device according to claim 7, which is characterized in that the similar categorization determining module, comprising:
Computing unit calculates the similarity between any two classifications for the feature vector according to the multiple classification;
Similar categorization determination unit is used for for each classification, according to removing in the classification and the multiple classification
The similarity between other classifications except classification, determines similar categorization of the classification in the multiple classification.
10. device according to claim 9, which is characterized in that the detail information of the object includes the object more
Information in a dimension, then the feature vector of the classification includes characteristic value of the classification in the multiple dimension;
The similar categorization determination unit is also used to create hyperspace according to the multiple dimension, according to each classification
Characteristic value of the feature vector in the multiple dimension, determine position of each classification in the hyperspace;Root
According to position of the multiple classification in the hyperspace, the hyperspace is divided into multiple regions;For described every
A classification, by the classification affiliated area and the adjacent area in the region alternately region, according to the classification and institute
The similarity in alternative area between each classification is stated, determines similar categorization of the classification in the alternative area.
11. device according to claim 10, which is characterized in that the similar categorization determination unit is also used to according to institute
The similarity in classification and the alternative area between each classification is stated, by the phase in the alternative area between the classification
Like the maximum classification of degree as the similar categorization;Alternatively, the sequence according to the similarity between the classification from big to small,
The classification that preset quantity is chosen from the alternative area, as the similar categorization.
12. device according to claim 7, which is characterized in that the migration path determining module, comprising:
Migration path determination unit is passed through for according to the correlation networks, determining from starting point classification classification of reaching home
A plurality of migration path;
Selection unit, for choosing the least migration path of classification number for including from a plurality of migration path;Alternatively,
The selection unit, for choosing the smallest migration path of migration distance, any two classes from a plurality of migration path
The similarity negative correlation between migration distance and any two described classifications between not.
13. the device that a kind of migration path generates, which is characterized in that described device includes processor and memory, the storage
At least one instruction, at least a Duan Chengxu, code set or instruction set, described instruction, described program, the code are stored in device
Collection or described instruction collection are loaded as the processor and are executed to realize moving as described in claim 1 to 6 any claim
Move operation performed in path generating method.
14. a kind of computer readable storage medium, which is characterized in that be stored at least one in the computer readable storage medium
Item instruction, at least a Duan Chengxu, code set or instruction set, described instruction, described program, the code set or described instruction collection by
Processor is loaded and is executed to realize in the migration path generation method as described in claim 1 to 6 any claim and hold
Capable operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811400866.6A CN109635153B (en) | 2018-11-22 | 2018-11-22 | Migration path generation method, device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811400866.6A CN109635153B (en) | 2018-11-22 | 2018-11-22 | Migration path generation method, device and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109635153A true CN109635153A (en) | 2019-04-16 |
CN109635153B CN109635153B (en) | 2023-01-13 |
Family
ID=66068804
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811400866.6A Active CN109635153B (en) | 2018-11-22 | 2018-11-22 | Migration path generation method, device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109635153B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116049096A (en) * | 2022-05-05 | 2023-05-02 | 荣耀终端有限公司 | Data migration method, electronic equipment and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103617540A (en) * | 2013-10-17 | 2014-03-05 | 浙江大学 | E-commerce recommendation method of tracking user interest changes |
CN106372961A (en) * | 2016-08-23 | 2017-02-01 | 北京小米移动软件有限公司 | Commodity recommendation method and device |
CN106445995A (en) * | 2016-07-18 | 2017-02-22 | 腾讯科技(深圳)有限公司 | Picture classification method and apparatus |
WO2018049900A1 (en) * | 2016-09-14 | 2018-03-22 | 广东欧珀移动通信有限公司 | Data migration method and related devices |
CN107977370A (en) * | 2016-10-21 | 2018-05-01 | 北京酷我科技有限公司 | A kind of singer recommends method and system |
CN108334640A (en) * | 2018-03-21 | 2018-07-27 | 北京奇艺世纪科技有限公司 | A kind of video recommendation method and device |
WO2018174908A1 (en) * | 2017-03-24 | 2018-09-27 | Hitachi, Ltd. | Method to enhance reliability of monitoring data for edge-core distributed analytics systems |
-
2018
- 2018-11-22 CN CN201811400866.6A patent/CN109635153B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103617540A (en) * | 2013-10-17 | 2014-03-05 | 浙江大学 | E-commerce recommendation method of tracking user interest changes |
CN106445995A (en) * | 2016-07-18 | 2017-02-22 | 腾讯科技(深圳)有限公司 | Picture classification method and apparatus |
CN106372961A (en) * | 2016-08-23 | 2017-02-01 | 北京小米移动软件有限公司 | Commodity recommendation method and device |
WO2018049900A1 (en) * | 2016-09-14 | 2018-03-22 | 广东欧珀移动通信有限公司 | Data migration method and related devices |
CN107977370A (en) * | 2016-10-21 | 2018-05-01 | 北京酷我科技有限公司 | A kind of singer recommends method and system |
WO2018174908A1 (en) * | 2017-03-24 | 2018-09-27 | Hitachi, Ltd. | Method to enhance reliability of monitoring data for edge-core distributed analytics systems |
CN108334640A (en) * | 2018-03-21 | 2018-07-27 | 北京奇艺世纪科技有限公司 | A kind of video recommendation method and device |
Non-Patent Citations (1)
Title |
---|
李世旭: "音乐推荐系统的设计和实现", 《中国优秀硕士学位论文全文数据库 电子期刊网》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116049096A (en) * | 2022-05-05 | 2023-05-02 | 荣耀终端有限公司 | Data migration method, electronic equipment and storage medium |
CN116049096B (en) * | 2022-05-05 | 2024-04-16 | 荣耀终端有限公司 | Data migration method, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109635153B (en) | 2023-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111652678B (en) | Method, device, terminal, server and readable storage medium for displaying article information | |
CN108304441A (en) | Network resource recommended method, device, electronic equipment, server and storage medium | |
CN109740068B (en) | Media data recommendation method, device and storage medium | |
CN108197327B (en) | Song recommendation method, device and storage medium | |
WO2020020156A1 (en) | Video processing method and apparatus, terminal device, server, and storage medium | |
CN110110203A (en) | Resource information method for pushing and server, resource information methods of exhibiting and terminal | |
CN109918669A (en) | Entity determines method, apparatus and storage medium | |
CN108491534B (en) | Information displaying method, device and computer equipment in virtual environment | |
CN110134804A (en) | Image search method, device and storage medium | |
CN108270794B (en) | Content distribution method, device and readable medium | |
CN109040297A (en) | User's portrait generation method and device | |
CN110147805A (en) | Image processing method, device, terminal and storage medium | |
CN111897996B (en) | Topic label recommendation method, device, equipment and storage medium | |
CN109640125A (en) | Video content processing method, device, server and storage medium | |
CN107463876A (en) | Information processing method and device, computer installation and storage medium | |
CN108717432A (en) | resource query method and device | |
CN109168073A (en) | The method and apparatus that direct broadcasting room cover is shown | |
CN109784351A (en) | Data classification method, disaggregated model training method and device | |
CN111611490A (en) | Resource searching method, device, equipment and storage medium | |
CN109068160A (en) | The methods, devices and systems of inking video | |
CN110139143A (en) | Virtual objects display methods, device, computer equipment and storage medium | |
CN110096525A (en) | Calibrate method, apparatus, equipment and the storage medium of interest point information | |
CN110246110A (en) | Image evaluation method, device and storage medium | |
CN109917988A (en) | Choose content display method, device, terminal and computer readable storage medium | |
CN113032587A (en) | Multimedia information recommendation method, system, device, terminal and server |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |