CN106682190A - Construction method and device of label knowledge base, application search method and server - Google Patents
Construction method and device of label knowledge base, application search method and server Download PDFInfo
- Publication number
- CN106682190A CN106682190A CN201611248542.6A CN201611248542A CN106682190A CN 106682190 A CN106682190 A CN 106682190A CN 201611248542 A CN201611248542 A CN 201611248542A CN 106682190 A CN106682190 A CN 106682190A
- Authority
- CN
- China
- Prior art keywords
- labels
- list
- label
- application
- rule
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
- G06F16/9562—Bookmark management
-
- 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
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The invention discloses a construction method and device of a label knowledge base, an application search method and a server. The construction method includes: acquiring label lists of multiple search words related to applications; acquiring label lists of multiple applications; constructing the label knowledge base according to the label lists of the search words, the label lists of the applications and a preset strategy to match the search words with the applications during application searching, wherein each label list includes one or multiple labels. The label knowledge base constructed by the method solves the technical problem of calculation of relevance between the search words and the applications, applications correlated to inquiry intentions can be searched, layout demonstration of search results is optimized, and search plus recommendation effect is realized. User intentions and applications are mapped into a same semantic space, so that the problem of semantic matching is solved, and functional search technology is realized effectively.
Description
Technical field
The present invention relates to Data Mining, and in particular to a kind of construction method of label repository, device, application searches
Method and server.
Background technology
A mobile terminal software application search engine service during application searches engine, it is possible to provide the app applications on mobile phone are searched
Rope and download, such as 360 mobile phone assistant, Tengxun are using treasured, GooglePlay, Appastore.With the hair of mobile Internet
Exhibition, mobile terminal app number of applications is continuously increased so that user can search in magnanimity application the application for meeting oneself intention
Difficulty is increased.
In traditional application retrieval technique based on keyword match, searched when user is expressed with semantic similar search word
During Suo Yitu, usually intention app applications cannot be obtained because keyword is mismatched.Existing general solution is to excavate synonymous
Word, expects to make up the unmatched defect of word.But in app application vertical searches field, synonym it is sparse so that excavating synonymous
The difficulty of word is very big.Even if in relatively easy web page search field, synonym mining algorithm also occurs producing effects poor, synonymous
The low problem of accuracy rate of word.So, so far also without good method, it can be applied with app across user's application searches
Between semantic gap.
The content of the invention
In view of the above problems, it is proposed that the present invention so as to provide one kind overcome above mentioned problem or at least in part solve on
State construction method, device, application searches method and the server of a kind of label repository of problem.
According to one aspect of the present invention, there is provided a kind of construction method of label repository, the method includes：
Obtain the list of labels of multiple search words on applying；
Obtain the list of labels of multiple applications；
The list of labels and preset strategy of list of labels, the application according to the search word, build label knowledge
Storehouse, for scanning for the matching of word and application in application searches；
Wherein, each list of labels includes one or more labels.
Alternatively, list of labels, the list of labels and preset strategy of the application, structure according to the search word
Building label repository includes：
Collect the list of labels of multiple search words and the list of labels of multiple applications, the set of list of labels is obtained, by institute
The set of list of labels is stated as training data；
Rule digging is associated to the training data, label repository is built according to the correlation rule excavated.
Alternatively, it is described rule digging is associated to the training data to include：
The association rule mining that N takes turns iteration is carried out to the training data using Apriori algorithm, every wheel iteration is obtained and is dug
The correlation rule for excavating；
In every wheel iteration, obtain a plurality of including preceding paragraph and consequent rule, if the preceding paragraph of a rule with it is consequent
Support is not less than the correlation rule of the wheel not less than the minimum support and preceding paragraph of the frequent episode of the wheel with consequent confidence level
Min confidence, it is determined that the rule is correlation rule and is excavated.
Alternatively,
In every wheel iteration, the preceding paragraph in the every rule for obtaining includes one or more labels, consequent including a mark
Sign.
Alternatively,
The minimum support of the frequent episode of the 1st wheel is the first predetermined threshold value, the frequent episode often taken turns in the 2nd wheel to N-1 wheels
Minimum support successively decrease the second predetermined threshold value, the minimum support of the frequent episode of N wheels is the 3rd predetermined threshold value；The pass of each wheel
The regular min confidence of connection is the 4th predetermined threshold value.
Alternatively, the correlation rule that the basis is excavated builds label repository to be included：
Treatment is merged to the correlation rule that each wheel iteration is excavated, the corresponding tree construction of each wheel iteration is obtained；
Merger is carried out to the corresponding tree construction of each wheel iteration, one or more tree constructions after merger are obtained；
Using one or more tree constructions after merger as constructed label repository；Wherein, each tree construction is every
One label of individual node correspondence, the topological structure of tree construction interior joint is used to represent the incidence relation between label.
Alternatively, it is described that treatment is merged to the correlation rule that each wheel iteration is excavated, obtain each wheel iteration corresponding
Tree construction includes：
In the correlation rule that every wheel iteration is excavated, when multiple correlation rules have identical consequent, to described many
The preceding paragraph of individual correlation rule is merged and obtains preceding paragraph set；
Using described consequent as root node, using the preceding paragraph set as the set of leaf node, the wheel iteration pair is obtained
The tree construction answered.
Alternatively, it is described merger is carried out to each wheel corresponding tree construction of iteration to include：
Iteration is taken turns from the 2nd wheel iteration to N, the corresponding tree construction of the i-th wheel iteration and the i-1 wheels before the wheel iteration are changed
Merger is carried out for corresponding tree construction, the corresponding tree construction of preceding i wheels iteration is obtained；Wherein, i be more than 1 and less than or equal to N just
Integer；
Preceding N takes turns the corresponding tree construction of iteration as one or more tree constructions after merger.
Alternatively, using the corresponding tree construction of the i-th wheel iteration as the first tree construction, the i-1 wheels before the wheel iteration are changed
For corresponding tree construction as the second tree construction；
Described i-th takes turns the corresponding tree construction of iteration tree construction corresponding with the i-1 wheel iteration before the wheel iteration is returned
And including：
Horizontal merger is carried out to the first tree construction and the second tree construction；Or, the first tree construction and the second tree construction are entered
The vertical merger of row.
Alternatively, it is described horizontal merger is carried out to the first tree construction and the second tree construction to include：
Calculate the similarity of the first tree construction and the second tree construction；
When the similarity is higher than five predetermined threshold value, determine that the first tree construction is similar tree knot to the second tree construction
Structure；
Similar the first tree construction and the second tree construction are merged in the horizontal direction of tree construction.
Alternatively, first tree construction and the similarity of the second tree construction of calculating includes：
When the root node of the first tree construction and the second tree construction corresponds to identical label, the leaf of the first tree construction is calculated
The Jaccard similarities that child node set is combined with the leaf node of the second tree construction, tie as the first tree construction and the second tree
The similarity of structure；
It is described by similar the first tree construction and the second tree construction the horizontal direction of tree construction merge including：Will be same
The leaf node of one layer of the first tree construction is merged with the leaf node of the second tree construction.
Alternatively, it is described vertical merger is carried out to the first tree construction and the second tree construction to include：
When the root node of the first tree construction, and a leaf node of the second tree construction is identical and the leaf node does not divide
Branch, by the leaf node of the tree construction of replacement second of the first tree construction, as a branch of the tree construction after merger.
Alternatively, the correlation rule that the basis is excavated builds label repository also to be included：
Tree construction after merger is modified, including following one or more：
Optimize the position of tree construction interior joint,
The mount point of branch in adjustment tree construction,
Addition label corresponding for each node adds one or more synonyms so that each node correspondence one is synonymous
Set of words.
According to another aspect of the present invention, there is provided a kind of application searches method, the method includes
Maintenance application database, preserves the list of labels of each application in the application database；
The search word of client upload is received, the list of labels of the search word is obtained；
Based on label repository, between the list of labels respectively applied in the list of labels and database that calculate the search word
Correlation degree；
When the correlation degree between the list of labels of the search word and a list of labels applied is higher than predetermined threshold value
When, the relevant information of the application is back to client and is shown；
The label repository is built by the method any one of present invention one side.
According to another aspect of the present invention, there is provided a kind of application searches device, the device includes：
Search word label acquiring unit, is suitable to obtain the list of labels of multiple search words on applying；
Using label acquiring unit, it is suitable to obtain the list of labels of multiple applications；
Construction of knowledge base unit, is suitable to list of labels according to the search word, the list of labels of the application and pre-
If tactful, label repository is built, for scanning for the matching of word and application in application searches；Wherein, each label
List includes one or more labels.
Alternatively,
The construction of knowledge base unit, is suitable to collect the list of labels of the list of labels of multiple search words and multiple applications,
The set of list of labels is obtained, using the set of the list of labels as training data；Rule are associated to the training data
Then excavate, label repository is built according to the correlation rule excavated.
Alternatively,
The construction of knowledge base unit, is suitable to carry out the training data association that N takes turns iteration using Apriori algorithm
Rule digging, obtains the correlation rule that every wheel iteration is excavated；In every wheel iteration, obtain a plurality of including preceding paragraph and consequent rule
Then, if the preceding paragraph of a rule and consequent support not less than the frequent episode of the wheel minimum support and preceding paragraph with it is rear
Correlation rule min confidence of the confidence level of item not less than the wheel, it is determined that the rule is correlation rule and is excavated.
Alternatively,
In every wheel iteration, the preceding paragraph in the every rule for obtaining includes one or more labels, consequent including a mark
Sign.
Alternatively,
The minimum support of the frequent episode of the 1st wheel is the first predetermined threshold value, the frequent episode often taken turns in the 2nd wheel to N-1 wheels
Minimum support successively decrease the second predetermined threshold value, the minimum support of the frequent episode of N wheels is the 3rd predetermined threshold value；The pass of each wheel
The regular min confidence of connection is the 4th predetermined threshold value.
Alternatively,
The construction of knowledge base unit, is suitable to merge treatment to the correlation rule that each wheel iteration is excavated, and obtains each
The corresponding tree construction of wheel iteration；Merger is carried out to the corresponding tree construction of each wheel iteration, one or more the tree knots after merger are obtained
Structure；Using one or more tree constructions after merger as constructed label repository；Wherein, each node of each tree construction
One label of correspondence, the topological structure of tree construction interior joint is used to represent the incidence relation between label.
Alternatively,
The construction of knowledge base unit, is suitable in the correlation rule that every wheel iteration is excavated, when multiple correlation rules tool
When having identical consequent, the preceding paragraph of the multiple correlation rule is merged and obtains preceding paragraph set；Using described consequent as root
Node, using the preceding paragraph set as the set of leaf node, obtains the corresponding tree construction of wheel iteration.
Alternatively,
The construction of knowledge base unit, is suitable to take turns iteration from the 2nd wheel iteration to N, by the corresponding tree construction of the i-th wheel iteration
Tree construction corresponding with the i-1 wheel iteration before the wheel iteration carries out merger, obtains the corresponding tree construction of preceding i wheels iteration；Wherein,
I is the positive integer more than 1 and less than or equal to N；Preceding N takes turns the corresponding tree construction of iteration as one or more the tree knots after merger
Structure.
Alternatively, using the corresponding tree construction of the i-th wheel iteration as the first tree construction, the i-1 wheels before the wheel iteration are changed
For corresponding tree construction as the second tree construction；
The construction of knowledge base unit, is suitable to carry out horizontal merger to the first tree construction and the second tree construction；Or, to
One tree construction and the second tree construction carry out vertical merger.
Alternatively,
The construction of knowledge base unit, is suitable to calculate the similarity of the first tree construction and the second tree construction；When described similar
When degree is higher than five predetermined threshold values, determine that the first tree construction is similar tree construction to the second tree construction；By the first similar tree
Structure and the second tree construction are merged in the horizontal direction of tree construction.
Alternatively,
The construction of knowledge base unit, is suitable to correspond to identical mark when the root node of the first tree construction and the second tree construction
During label, the Jaccard similarities that the leaf node set of the first tree construction is combined with the leaf node of the second tree construction are calculated, made
It is the first tree construction and the similarity of the second tree construction；It is described by similar the first tree construction and the second tree construction in tree construction
Horizontal direction merge including：The leaf node of the first tree construction of same layer is carried out with the leaf node of the second tree construction
Merge.
Alternatively,
The construction of knowledge base unit, is suitable to a leaf node of the root node and the second tree construction when the first tree construction
It is identical and the leaf node does not have branch, by the leaf node of the tree construction of replacement second of the first tree construction, as merger
One branch of tree construction afterwards.
Alternatively,
The construction of knowledge base unit, is further adapted for being modified the tree construction after merger, including following one or more：
Optimize the position of tree construction interior joint, the mount point of branch in adjustment tree construction adds label addition corresponding for each node
One or more synonyms so that each node one TongYiCi CiLin of correspondence.
According to another aspect of the invention, there is provided a kind of application searches server, the server includes：
Database maintenance unit, is suitable to maintenance application database, and the label of each application is preserved in the application database
List；
Interactive unit, is suitable to receive the search word of client upload, obtains the list of labels of the search word；
Search processing, the list of labels for being suitable to be calculated based on label repository the search word is each with database
Correlation degree between the list of labels of application；
The interactive unit, is further adapted for when the pass between the list of labels of the search word and a list of labels for application
When connection degree is higher than predetermined threshold value, the relevant information of the application is back to client and is shown；
The server is also including the construction device of the label repository any one of the 3rd aspect in the present invention.
According to the construction method of label repository, device, application searches method and server in the present invention, base can be made up
In the defect of the application searches engine of keyword match, thus solve caused by being mismatched by semanteme, user is in magnanimity application
Search cannot usually obtain the problem of the application for meeting search intention when applying in storehouse, achieve more convenient, accurate, efficiently
The beneficial effect of searching functions.
Described above is only the general introduction of technical solution of the present invention, in order to better understand technological means of the invention,
And can be practiced according to the content of specification, and in order to allow the above and other objects of the present invention, feature and advantage can
Become apparent, below especially exemplified by specific embodiment of the invention.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit is common for this area
Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings：
Fig. 1 shows a kind of flow chart of the construction method of label repository according to an embodiment of the invention；
Fig. 2 shows a schematic diagram for tree in label repository in an embodiment of the invention；
Fig. 3 shows correlation rule stipulations schematic diagram in an embodiment of the invention；
Fig. 4 shows two horizontal merger schematic diagrames of tree in an embodiment of the invention；
Fig. 5 shows two horizontal vertical merger schematic diagrames of tree in an embodiment of the invention；
Fig. 6 shows the one tree topological structure schematic diagram in label repository in an embodiment of the invention；
Fig. 7 shows the partial schematic diagram of label repository in an embodiment of the invention；
Fig. 8 shows a kind of flow chart of application searches method according to an embodiment of the invention；
Fig. 9 shows a kind of schematic diagram of the construction device of label repository according to an embodiment of the invention；And
Figure 10 shows a kind of schematic diagram of application searches server according to an embodiment of the invention
Specific embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although showing the disclosure in accompanying drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
Limited.Conversely, there is provided these embodiments are able to be best understood from the disclosure, and can be by the scope of the present disclosure
Complete conveys to those skilled in the art.
Hereinafter, application is represented with app, query represents search word, and tag represents label, and TagNet represents label data
Storehouse.
The present invention is to build label repository TagNet, is the key for realizing searching functions technology.TagNet similar to
WordNet, builds the knowledge hierarchy of a set of tissue tag incidence relations, and user view and app functions are mapped into same tag
In system, the semantic matches of application searches engine are solved the problems, such as.
Fig. 1 shows a kind of flow chart of the construction method of label repository according to an embodiment of the invention.Such as Fig. 1
Shown, the method includes
Step S110, obtains the list of labels of multiple search words on applying.
Step S120, obtains the list of labels of multiple applications.
Step S130, the list of labels and preset strategy of list of labels, the application according to the search word build
Label repository, for scanning for the matching of word and application in application searches.
Wherein, marked including one or more in the list of labels of the search word of each application or in the list of labels of application
Sign.
The search word of application is the keyword that user is input into when application is searched for, and existing each app should
With there is corresponding keyword to describe.
Label repository TagNet, similar to word knowledge base WordNet, is a set of comprising bookmark name and each label
Between incidence relation knowledge base.The structure of label repository is the key for realizing searching functions technology.
It can be seen that, the method shown in Fig. 1 is to build label repository, and based on label repository, we solve query and app
The technical barrier of correlation calculations, can search the app applications being associated with query intention, the layout exhibition of Optimizing Search result
It is existing, realize search plus the effect recommended.User view and app are mapped in same semantic space, semantic is solved
With problem, searching functions technology is effectively realized.
Label repository, is a set of knowledge base for describing label relation, and tag is organized into tree, the stratification of tree
Show the hypernym-hyponym subordinate relation between Tag, by similar or related tag clusters to one tree, from root node to leaf
The concept that node is represented from typically to specific, from being abstracted into tool as realizing the concept hierarchy cluster of tag.Fig. 2 is shown at this
A schematic diagram for tree in label repository in invention one embodiment, as shown in Figure 2, label relation is organized into tree
Shape structure.Each node represents a label, such as " game " " risk " " forest ".200th, 210 and 220 three of tree are represented
Level, expression belongs to the subordinate relation of the hypernym-hyponym between the label of different levels.Similar or related label is gathered
Class to one tree, the tag concept represented from root node to leaf node from typically to specific, from being abstracted into tool as realizing mark
The concept hierarchy cluster of label.
In one embodiment of the invention, according to the label column of the search word in the step of Fig. 1 methods describeds S130
Table, the list of labels and preset strategy of the application, building label repository includes：Collect the list of labels of multiple search words
With the list of labels of multiple applications, the set of list of labels is obtained, using the set of the list of labels as training data；To institute
State training data and be associated rule digging, label repository is built according to the correlation rule excavated.
In the list of labels of the list of labels of application and the search word of application, there is a kind of label co-occurrence rule：Tool as
Label frequent item set, is usually associated with an abstract label, the label of this co-occurrence namely the tool as the hypernym of label.
Table 1 shows tool in an embodiment of the invention as label and the co-occurrence phenomenon of abstract label.As the institute of table 1
Show have label such as " life " " instrument " " leisure " " unit " label of elephant, represent specific type of play.And " game " is marked
Label are a concept, both high frequency co-occurrences.
Table 1
Assuming that the list of labels of the search word of the application of selection 2,000,000, the list of labels of 2,000,000 applications, only retain therein
List of labels, removes the search word of corresponding application and corresponding application.Every one list of labels of row, totally 400 ten thousand, output is arrived
In one file, the part labels content of part row is shown by table 2.I.e. table 2 shows instruction in an embodiment of the invention
Practice collection part row.
Table 2
In one embodiment of the invention, it is described rule digging is associated to the training data to include：
The association rule mining that N takes turns iteration is carried out to the training data using Apriori algorithm, every wheel iteration is obtained and is dug
The correlation rule for excavating；In every wheel iteration, obtain a plurality of including preceding paragraph and consequent rule, if the preceding paragraph of a rule with
Consequent support is not less than the pass of the wheel not less than the minimum support and preceding paragraph of the frequent episode of the wheel with consequent confidence level
The regular min confidence of connection, it is determined that the rule is correlation rule and is excavated.
In specific setting, in every wheel iteration, the preceding paragraph in the every rule for obtaining includes one or more labels, after
Item includes a label.This is that, based on above-mentioned co-occurrence phenomenon, with association rule mining, the preceding paragraph of correlation rule is the label for having elephant
Frequent Set, can be comprising multiple labels, and consequent is abstract label frequent episode, and consequent takes a label.
In one embodiment of the invention, the minimum support of the frequent episode of the 1st wheel is the first predetermined threshold value, the 2nd wheel
The minimum support of frequent episode often taken turns into N-1 wheels successively decreases the second predetermined threshold value, the minimum support of the frequent episode of N wheels
It is the 3rd predetermined threshold value；The correlation rule min confidence of each wheel is the 4th predetermined threshold value.
Table 3 shows the output result part of the wheel of execution one Apriori in an embodiment of the invention.
Table 3
| Rule is consequent<- regular preceding paragraph (enhancing rate, confidence level %) |
| Game<- leisure (19.7175,99.5196) |
| Game<- unit (16.3539,99.6702) |
| Game<- unit lies fallow (7.80958,99.9702) |
| Game<- intelligence development (12.7918,99.4027) |
| Game<(10.635,99.8265) is lain fallow in-intelligence development |
| Game<- intelligence development unit (5.0142,99.9752) |
| Leisure<- intelligence development single-play game (5.01296,92.2045) |
| Leisure<- intelligence development unit (5.0142,92.204) |
| Game<- shooting (11.9815,99.3075) |
| Game<- risk (9.61166,99.4388) |
| Game<- simulation (9.46392,90.016) |
| Game<- leisure intelligence development (7.73153,99.8159) |
| Game<- leisure intelligence development leisure (5.69434,99.8965) |
| Game<- cool run (6.90657,98.924) |
| Instrument<- system (6.43224,89.0376) |
| Game<- fight (6.3949,99.1897) |
| Game<- strategy (6.3933,94.2213) |
| Game<- 360 fine work play (5.76818,99.8123) |
| Game<- action (5.57768,98.9898) |
| Game<- war (5.35935,98.6695) |
| Game<- manage (5.34743,98.1273) |
| Game<- tactful (5.29036,99.3829) |
It is assumed that when certain Selection utilization Apriori algorithm does association rule mining, 11 wheel iteration need to be performed altogether, i.e. N is
11.First predetermined threshold value is 5.0, and the second predetermined threshold value is that each round successively decreases 0.5 on the basis of 5.0, and the 3rd predetermined threshold value is
0.1, the 4th preset value is 85%, constant in iteration, only generate it is consequent comprising a correlation rule for frequent episode, as shown in Figure 5.
The result that each round iteration is excavated is required for being merged together, and the merger operation of tree is performed after merging, starts next round iteration.
In one embodiment of the invention, the correlation rule that the basis is excavated builds label repository to be included：It is right
The correlation rule that each wheel iteration is excavated merges treatment, obtains the corresponding tree construction of each wheel iteration；To each wheel iteration correspondence
Tree construction carry out merger, obtain one or more tree constructions after merger；Using one or more tree constructions after merger as
Constructed label repository；Wherein, each node one label of correspondence of each tree construction, the topology knot of tree construction interior joint
Structure is used to represent the incidence relation between label.
Wherein, it is described that treatment is merged to the correlation rule that each wheel iteration is excavated, obtain the corresponding tree of each wheel iteration
Structure includes：
In the correlation rule that every wheel iteration is excavated, when multiple correlation rules have identical consequent, to described many
The preceding paragraph of individual correlation rule is merged and obtains preceding paragraph set；Using described consequent as root node, using the preceding paragraph set as
The set of leaf node, obtains the corresponding tree construction of wheel iteration.
Fig. 3 shows correlation rule stipulations schematic diagram in an embodiment of the invention.
I.e. to newly-generated some correlation rules, there is identical consequent preceding paragraph to merge, it is consequent as root node, it is preceding
Item collection cooperation is leaf node.For example, if a certain wheel only generates 3 correlation rules, " game ← war ", " play ← run
Extremely ", " game ← trivial games "." game " is the consequent of rule, and preceding paragraph is " war ", " cool run ", " trivial games ", is merged one
Rise, constitute preceding paragraph set " war, cool run, trivial games ".The newly-generated correlation rule of this wheel is converted into a tree construction：
" game ← war, cool run, trivial games ", " game " is the root node of tree, and remaining is leaf node, as shown in Figure 3.
It is described merger is carried out to the corresponding tree construction of each wheel iteration to include additionally, in one embodiment of the invention：
Iteration is taken turns from the 2nd wheel iteration to N, the corresponding tree construction of the i-th wheel iteration and the i-1 wheels before the wheel iteration are changed
Merger is carried out for corresponding tree construction, the corresponding tree construction of preceding i wheels iteration is obtained；Wherein, i be more than 1 and less than or equal to N just
Integer；Preceding N takes turns the corresponding tree construction of iteration as one or more tree constructions after merger.
When the data of a certain wheel Apriori outputs merge with the tree of iteration before, exist as root node before
Label, is specifically abandoned, it is impossible to be added in tree as leaf node.
In one embodiment of the invention, the corresponding tree construction of the i-th wheel iteration is changed the wheel as the first tree construction
The instead of preceding corresponding tree construction of i-1 wheel iteration is used as the second tree construction；Described i-th takes turns the corresponding tree construction of iteration changes with the wheel
The instead of preceding corresponding tree construction of i-1 wheel iteration carries out merger to be included：Level is carried out to the first tree construction and the second tree construction to return
And；Or, vertical merger is carried out to the first tree construction and the second tree construction.
Specifically, it is described horizontal merger is carried out to the first tree construction and the second tree construction to include：Calculate the first tree construction and
The similarity of the second tree construction；When the similarity is higher than five predetermined threshold value, the first tree construction and the second tree construction are determined
It is similar tree construction；Similar the first tree construction and the second tree construction are merged in the horizontal direction of tree construction.
Fig. 4 shows two horizontal merger schematic diagrames of tree in an embodiment of the invention, as shown in Figure 4, tree
(A, B, C, D) and tree (A, B, C, E) merge into one tree (A, B, C, D, E).
Wherein, first tree construction and the similarity of the second tree construction of calculating includes：
When the root node of the first tree construction and the second tree construction corresponds to identical label, the leaf of the first tree construction is calculated
The Jaccard similarities that child node set is combined with the leaf node of the second tree construction, tie as the first tree construction and the second tree
The similarity of structure；It is described by similar the first tree construction and the second tree construction the horizontal direction of tree construction merge including：
The leaf node of the first tree construction of same layer and the leaf node of the second tree construction are merged.
Specifically, in one embodiment of the invention, it is described that first tree construction and the second tree construction are vertically returned
And including：
When the root node of the first tree construction, and a leaf node of the second tree construction is identical and the leaf node does not divide
Branch, by the leaf node of the tree construction of replacement second of the first tree construction, as a branch of the tree construction after merger.
Fig. 5 shows in an embodiment of the invention the two vertical merger schematic diagrames of tree, newly-generated tree, such as
First step output, before being merged into as a branch in the another one tree of grey iterative generation.As shown in Figure 5, (B, T, G) is set
A leaf node of tree (A, B, C, E) is substituted, as a branch, one tree (A, B, (B, T, G) C, E) is merged into.
After completing above-mentioned steps, you can tentatively generate a label repository, containing treated label word in the storehouse, such as
Same to set the forest for constituting by some, each tree represents a concept or entity sets, and structure is similar to word knowledge base
WordNet。
Fig. 6 shows the one tree topological structure schematic diagram in label repository in an embodiment of the invention.Have two
Individual primary root node, respectively " health " and " investment ".As a rule, the label repository that this step is obtained is original tag knowledge
Storehouse, accuracy rate about 60%~65%.
In one embodiment of the invention, the correlation rule that the basis is excavated builds label repository also to be included：
Tree construction after merger is modified, including following one or more：Optimize the position of tree construction interior joint, adjust tree construction
The mount point of middle branch, adds label corresponding for each node and adds one or more synonyms so that each node correspondence
One TongYiCi CiLin.
Amendment by this step to original tag knowledge base, finally gives the label repository of a high-accuracy, accurate
True rate is about 96%.This is that this programme calculates the search word of application and the basic data structure of app application semantics correlations.
Fig. 7 shows the partial schematic diagram of label repository in an embodiment of the invention, there is two primary root nodes,
Respectively " play " and " phone ".
Fig. 8 shows a kind of flow chart of application searches method according to an embodiment of the invention.As shown in figure 8, should
Method includes：
Step 810, maintenance application database preserves the list of labels of each application in the application database.
Step 820, receives the search word of client upload, obtains the list of labels of the search word.
Step 830, based on label repository, calculates the list of labels of the search word and the label of each application in database
Correlation degree between list.
Step 840, when the correlation degree between the list of labels of the search word and a list of labels applied is higher than
During predetermined threshold value, the relevant information of the application is back to client and is shown.
The label repository is built by the method in the present invention described in any embodiment.
Fig. 9 shows a kind of construction device of label repository according to an embodiment of the invention, and the device 900 is wrapped
Include：
Search word label acquiring unit 910, is suitable to obtain the list of labels of multiple search words on applying；
Using label acquiring unit 920, it is suitable to obtain the list of labels of multiple applications；
Construction of knowledge base unit 930, be suitable to list of labels according to the search word, the list of labels of the application and
Preset strategy, builds label repository, for scanning for the matching of word and application in application searches；Wherein, each should
Include one or more labels in the list of labels of search word or in the list of labels of application.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to collect the label of multiple search words
List and the list of labels of multiple applications, obtain the set of list of labels, using the set of the list of labels as training data；
Rule digging is associated to the training data, label repository is built according to the correlation rule excavated.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to using Apriori algorithm to institute
Stating training data carries out the association rule mining that N takes turns iteration, obtains the correlation rule that every wheel iteration is excavated；In every wheel iteration
In, obtain a plurality of including preceding paragraph and consequent rule, if the preceding paragraph of a rule with consequent support not less than the wheel
Correlation rule min confidence of the minimum support and preceding paragraph of frequent episode with consequent confidence level not less than the wheel, it is determined that
The rule is correlation rule and is excavated.
In one embodiment of the invention, the construction device 900 of the label repository, in every wheel iteration, obtains
Every rule in preceding paragraph include one or more labels, it is consequent including a label.
In one embodiment of the invention, the frequent episode that the construction device the 900, the 1st of the label repository is taken turns is most
Small support is the first predetermined threshold value, and the minimum support of frequent episode often taken turns in the 2nd wheel to N-1 wheels successively decreases the second default threshold
Value, the minimum support of the frequent episode of N wheels is the 3rd predetermined threshold value；The correlation rule min confidence of each wheel is preset for the 4th
Threshold value.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to the pass excavated to each wheel iteration
Connection rule merges treatment, obtains the corresponding tree construction of each wheel iteration；Merger is carried out to the corresponding tree construction of each wheel iteration, is obtained
One or more tree constructions after to merger；Using one or more tree constructions after merger as constructed label repository；
Wherein, each node one label of correspondence of each tree construction, the topological structure of tree construction interior joint is used to representing between label
Incidence relation.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to the pass excavated in every wheel iteration
In connection rule, when multiple correlation rules have identical consequent, the preceding paragraph of the multiple correlation rule is merged and is obtained
Preceding paragraph set；Using described consequent as root node, using the preceding paragraph set as the set of leaf node, the wheel iteration pair is obtained
The tree construction answered.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to be changed from the 2nd wheel iteration to N wheels
In generation, the corresponding tree construction of the i-th wheel iteration tree construction corresponding with the i-1 wheel iteration before the wheel iteration is carried out into merger, obtained
Preceding i takes turns the corresponding tree construction of iteration；Wherein, i is the positive integer more than 1 and less than or equal to N；Preceding N takes turns the corresponding tree construction of iteration
As one or more tree constructions after merger.
In one embodiment of the invention, the construction device 900 of the label repository, the i-th wheel iteration is corresponding
The i-1 before the wheel iteration is taken turns the corresponding tree construction of iteration as the second tree construction as the first tree construction for tree construction；It is described
Construction of knowledge base unit, is suitable to carry out horizontal merger to the first tree construction and the second tree construction；Or, to the first tree construction and
Two tree constructions carry out vertical merger.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to calculate the first tree construction and second
The similarity of tree construction；When the similarity is higher than five predetermined threshold value, determine that the first tree construction is phase with the second tree construction
As tree construction；Similar the first tree construction and the second tree construction are merged in the horizontal direction of tree construction.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to when the first tree construction and the second tree
When the root node of structure corresponds to identical label, the leaf node set of the first tree construction and the leaf of the second tree construction are calculated
The Jaccard similarities that node is combined, as the first tree construction and the similarity of the second tree construction；It is described by the first similar tree
Structure and the second tree construction the horizontal direction of tree construction merge including：By the leaf node of the first tree construction of same layer
Leaf node with the second tree construction is merged.
In one embodiment of the invention, the construction of knowledge base unit 930, is suitable to the root node when the first tree construction
It is identical with a leaf node of the second tree construction and the leaf node does not have branch, the tree of replacement second of the first tree construction is tied
The leaf node of structure, as a branch of the tree construction after merger.
In one embodiment of the invention, the construction of knowledge base unit 930, is further adapted for entering the tree construction after merger
Row amendment, including following one or more：Optimize the position of tree construction interior joint, the mount point of branch, adds in adjustment tree construction
Plus label corresponding for each node adds one or more synonyms so that each node one TongYiCi CiLin of correspondence.
Figure 10 shows a kind of application searches server according to an embodiment of the invention, as shown in Figure 10, should
Included with search server 1000：
Database maintenance unit 1010, is suitable to maintenance application database, and each application is preserved in the application database
List of labels.
Interactive unit 1020, is suitable to receive the search word of client upload, obtains the list of labels of the search word.
Search processing 1030, is suitable to based on label repository, calculates the list of labels and database of the search word
In each application list of labels between correlation degree.
The interactive unit 1020, is further adapted for when between the list of labels of the search word and a list of labels for application
Correlation degree be higher than predetermined threshold value when, the relevant information of the application is back to client and is shown.
The server 1000 also includes the construction device 900 of the label repository as described in any embodiment of the present invention.
In sum, in the inventive solutions, it is proposed that the construction method and device of a kind of label repository, solve
Determine the technical barrier that search word and application relativity calculate, the application being associated with query intention, Optimizing Search can have been searched
The layout of result represents, and realizes search plus the effect recommended.The invention, same semanteme is mapped to by user view and application
In space, semantic matches are solved the problems, such as, effectively realize searching functions technology.In application intelligent recommendation, search is benefited wide
Accuse, app application searches result personalized layouts app application aspects similar with discovery bring the effect being highly profitable.
It should be noted that：
Algorithm and display be not inherently related to any certain computer, virtual bench or miscellaneous equipment provided herein.
Various fexible units can also be used together with based on teaching in this.As described above, construct required by this kind of device
Structure be obvious.Additionally, the present invention is not also directed to any certain programmed language.It is understood that, it is possible to use it is various
Programming language realizes the content of invention described herein, and the description done to language-specific above is to disclose this hair
Bright preferred forms.
In specification mentioned herein, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be put into practice in the case of without these details.In some instances, known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it will be appreciated that in order to simplify one or more that the disclosure and helping understands in each inventive aspect, exist
Above to the description of exemplary embodiment of the invention in, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：I.e. required guarantor
The application claims of shield features more more than the feature being expressly recited in each claim.More precisely, such as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, and wherein each claim is in itself
All as separate embodiments of the invention.
Those skilled in the art are appreciated that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.Can be the module or list in embodiment
Unit or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, can use any
Combine to all features disclosed in this specification (including adjoint claim, summary and accompanying drawing) and so disclosed appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power
Profit is required, summary and accompanying drawing) disclosed in each feature can the alternative features of or similar purpose identical, equivalent by offer carry out generation
Replace.
Although additionally, it will be appreciated by those of skill in the art that some embodiments described herein include other embodiments
In included some features rather than further feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment required for protection is appointed
One of meaning mode can be used in any combination.
All parts embodiment of the invention can be realized with hardware, or be run with one or more processor
Software module realize, or with combinations thereof realize.It will be understood by those of skill in the art that can use in practice
Microprocessor or digital signal processor (DSP) come realize label repository according to embodiments of the present invention construction device,
The some or all functions of some or all parts in application searches server.The present invention is also implemented as holding
Some or all equipment or program of device of row method as described herein are (for example, computer program and computer
Program product).It is such to realize that program of the invention be stored on a computer-readable medium, or can have one or
The form of person's multiple signal.Such signal can be downloaded from internet website and obtained, or be provided on carrier signal, or
Person provides in any other form.
It should be noted that above-described embodiment the present invention will be described rather than limiting the invention, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol being located between bracket should not be configured to limitations on claims.Word "comprising" is not excluded the presence of not
Element listed in the claims or step.Word "a" or "an" before element is not excluded the presence of as multiple
Element.The present invention can come real by means of the hardware for including some different elements and by means of properly programmed computer
It is existing.If in the unit claim for listing equipment for drying, several in these devices can be by same hardware branch
To embody.The use of word first, second, and third does not indicate that any order.These words can be explained and run after fame
Claim.
Claims (10)
1. a kind of construction method of label repository, wherein, including：
Obtain the list of labels of multiple search words on applying；
Obtain the list of labels of multiple applications；
The list of labels and preset strategy of list of labels, the application according to the search word, build label repository, with
Matching for scanning for word and application in application searches；
Wherein, each list of labels includes one or more labels.
2. the method for claim 1, wherein list of labels, the label of the application according to the search word
List and preset strategy, building label repository includes：
Collect the list of labels of multiple search words and the list of labels of multiple applications, the set of list of labels is obtained, by the mark
The set of list is signed as training data；
Rule digging is associated to the training data, label repository is built according to the correlation rule excavated.
3. method as claimed in claim 1 or 2, wherein, it is described rule digging is associated to the training data to include：
The association rule mining that N takes turns iteration is carried out to the training data using Apriori algorithm, every wheel iteration is obtained and is excavated
Correlation rule；
In every wheel iteration, obtain a plurality of including preceding paragraph and consequent rule, if the preceding paragraph of a rule and consequent support
Degree is minimum not less than the correlation rule of the wheel with consequent confidence level not less than the minimum support and preceding paragraph of the frequent episode of the wheel
Confidence level, it is determined that the rule is correlation rule and is excavated.
4. the method as any one of claim 1-3, wherein,
In every wheel iteration, the preceding paragraph in the every rule for obtaining includes one or more labels, consequent including a label.
5. a kind of application searches method, wherein, including：
Maintenance application database, preserves the list of labels of each application in the application database；
The search word of client upload is received, the list of labels of the search word is obtained；
Based on label repository, the pass between the list of labels of each application in the list of labels and database of the search word is calculated
Connection degree；
When the correlation degree between the list of labels of the search word and a list of labels applied is higher than predetermined threshold value, will
The relevant information of the application is back to client and is shown；
The label repository is built by the method as any one of claim 1-4.
6. a kind of construction device of label repository, wherein, including：
Search word label acquiring unit, is suitable to obtain the list of labels of multiple search words on applying；
Using label acquiring unit, it is suitable to obtain the list of labels of multiple applications；
Construction of knowledge base unit, is suitable to list of labels according to the search word, the list of labels of the application and default plan
Slightly, label repository is built, for scanning for the matching of word and application in application searches；Wherein, each list of labels
Include one or more labels.
7. device as claimed in claim 6, wherein,
The construction of knowledge base unit, is suitable to collect the list of labels of the list of labels of multiple search words and multiple applications, obtains
The set of list of labels, using the set of the list of labels as training data；Regular digging is associated to the training data
Pick, label repository is built according to the correlation rule excavated.
8. device as claimed in claims 6 or 7, wherein,
The construction of knowledge base unit, is suitable to carry out the training data correlation rule that N takes turns iteration using Apriori algorithm
Excavate, obtain the correlation rule that every wheel iteration is excavated；In every wheel iteration, obtain a plurality of including preceding paragraph and consequent rule,
If the preceding paragraph of a rule and consequent support not less than the frequent episode of the wheel minimum support and preceding paragraph with it is consequent
Correlation rule min confidence of the confidence level not less than the wheel, it is determined that the rule is correlation rule and is excavated.
9. the device as any one of claim 6-8, wherein,
In every wheel iteration, the preceding paragraph in the every rule for obtaining includes one or more labels, consequent including a label.
10. a kind of application searches server, wherein, including：
Database maintenance unit, is suitable to maintenance application database, and the list of labels of each application is preserved in the application database；
Interactive unit, is suitable to receive the search word of client upload, obtains the list of labels of the search word；
Search processing, is suitable to based on label repository, calculates and respectively apply in the list of labels of the search word and database
List of labels between correlation degree；
The interactive unit, is further adapted for working as associating journey between the list of labels of the search word and a list of labels for application
When degree is higher than predetermined threshold value, the relevant information of the application is back to client and is shown；
The server also includes the construction device of the label repository as any one of claim 6-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611248542.6A CN106682190B (en) | 2016-12-29 | 2016-12-29 | Construction method and device of tag knowledge base, application search method and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611248542.6A CN106682190B (en) | 2016-12-29 | 2016-12-29 | Construction method and device of tag knowledge base, application search method and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106682190A true CN106682190A (en) | 2017-05-17 |
| CN106682190B CN106682190B (en) | 2020-12-15 |
Family
ID=58872531
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611248542.6A Active CN106682190B (en) | 2016-12-29 | 2016-12-29 | Construction method and device of tag knowledge base, application search method and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106682190B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107688614A (en) * | 2017-08-04 | 2018-02-13 | 平安科技（深圳）有限公司 | It is intended to acquisition methods, electronic installation and computer-readable recording medium |
| CN107798082A (en) * | 2017-10-16 | 2018-03-13 | 广东欧珀移动通信有限公司 | A kind of processing method and processing device of file label |
| CN109471888A (en) * | 2018-11-15 | 2019-03-15 | 广东电网有限责任公司信息中心 | A kind of method of invalid information in quick filtering xml document |
| CN111158699A (en) * | 2019-12-31 | 2020-05-15 | 青岛海尔科技有限公司 | Application optimization method and device based on Apriori algorithm and intelligent equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110302191A1 (en) * | 2010-06-04 | 2011-12-08 | Evgenios Moroz | System and Method for Locating Business Verifications from Trusted Persons |
| US20140074816A1 (en) * | 2012-06-25 | 2014-03-13 | Rediff.Com India Limited | Method and apparatus for generating a query candidate set |
| CN104281656A (en) * | 2014-09-18 | 2015-01-14 | 广州三星通信技术研究有限公司 | Method and device for adding label information into application program |
| CN105719189A (en) * | 2016-01-15 | 2016-06-29 | 天津大学 | Tag recommendation method for effectively increasing tag diversity in social network |
| CN105893441A (en) * | 2015-12-15 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Application recommendation method and application recommendation system for terminal |
| CN105893440A (en) * | 2015-12-15 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Associated application recommendation method and apparatus |
-
2016
- 2016-12-29 CN CN201611248542.6A patent/CN106682190B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110302191A1 (en) * | 2010-06-04 | 2011-12-08 | Evgenios Moroz | System and Method for Locating Business Verifications from Trusted Persons |
| US20140074816A1 (en) * | 2012-06-25 | 2014-03-13 | Rediff.Com India Limited | Method and apparatus for generating a query candidate set |
| CN104281656A (en) * | 2014-09-18 | 2015-01-14 | 广州三星通信技术研究有限公司 | Method and device for adding label information into application program |
| CN105893441A (en) * | 2015-12-15 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Application recommendation method and application recommendation system for terminal |
| CN105893440A (en) * | 2015-12-15 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Associated application recommendation method and apparatus |
| CN105719189A (en) * | 2016-01-15 | 2016-06-29 | 天津大学 | Tag recommendation method for effectively increasing tag diversity in social network |
Non-Patent Citations (1)
| Title |
|---|
| 罗可: ""数据库中数据挖掘理论方法及应用研究"", 《中国优秀博硕士学位论文全文数据库(博士)信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107688614A (en) * | 2017-08-04 | 2018-02-13 | 平安科技（深圳）有限公司 | It is intended to acquisition methods, electronic installation and computer-readable recording medium |
| CN107688614B (en) * | 2017-08-04 | 2018-08-10 | 平安科技（深圳）有限公司 | It is intended to acquisition methods, electronic device and computer readable storage medium |
| CN107798082A (en) * | 2017-10-16 | 2018-03-13 | 广东欧珀移动通信有限公司 | A kind of processing method and processing device of file label |
| CN107798082B (en) * | 2017-10-16 | 2020-07-17 | Oppo广东移动通信有限公司 | File label processing method and device |
| CN109471888A (en) * | 2018-11-15 | 2019-03-15 | 广东电网有限责任公司信息中心 | A kind of method of invalid information in quick filtering xml document |
| CN109471888B (en) * | 2018-11-15 | 2021-11-09 | 广东电网有限责任公司信息中心 | Method for rapidly filtering invalid information in xml file |
| CN111158699A (en) * | 2019-12-31 | 2020-05-15 | 青岛海尔科技有限公司 | Application optimization method and device based on Apriori algorithm and intelligent equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106682190B (en) | 2020-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pan et al. | Examining the usage, citation, and diffusion patterns of bibliometric mapping software: A comparative study of three tools | |
| Georgiev et al. | Enhancing user creativity: Semantic measures for idea generation | |
| Chen et al. | Merging domain ontologies based on the WordNet system and fuzzy formal concept analysis techniques | |
| CN104412265B (en) | Update for promoting the search of application searches to index | |
| Jun et al. | Examining technological innovation of Apple using patent analysis | |
| CN112037920A (en) | Medical knowledge map construction method, device, equipment and storage medium | |
| CN106682190A (en) | Construction method and device of label knowledge base, application search method and server | |
| CN102937976B (en) | A kind of drop-down reminding method based on input prefix and device | |
| CN110427478B (en) | Knowledge graph-based question and answer searching method and system | |
| US20180232351A1 (en) | Joining web data with spreadsheet data using examples | |
| EP3940582A1 (en) | Method for disambiguating between authors with same name on basis of network representation and semantic representation | |
| Kargar et al. | Efficient duplication free and minimal keyword search in graphs | |
| KR20060122276A (en) | Relation extraction from documents for the automatic construction of ontologies | |
| Du et al. | An approach for selecting seed URLs of focused crawler based on user-interest ontology | |
| CN103810159B (en) | Machine translation data processing method, system and terminal | |
| CN107562966B (en) | Intelligent learning-based optimization system and method for webpage link retrieval sequencing | |
| CN110516164B (en) | Information recommendation method, device, equipment and storage medium | |
| An et al. | Automatic generation of ontology from the deep web | |
| Simperl et al. | Combining human and computation intelligence: the case of data interlinking tools | |
| Han et al. | DTaxa: An actor–critic for automatic taxonomy induction | |
| CN105824976A (en) | Method and device for optimizing word segmentation banks | |
| JP2019020939A (en) | Information processing system, information processing method, and program | |
| Naredi et al. | Improved extraction of quantitative rules using best M positive negative association rules algorithm | |
| Kejriwal et al. | Unsupervised real-time induction and interactive visualization of taxonomies over domain-specific concepts | |
| Zhang et al. | A twig-based algorithm for top-k subgraph matching in large-scale graph data |
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 |