CN109857872A - The information recommendation method and device of knowledge based map - Google Patents
The information recommendation method and device of knowledge based map Download PDFInfo
- Publication number
- CN109857872A CN109857872A CN201910120048.9A CN201910120048A CN109857872A CN 109857872 A CN109857872 A CN 109857872A CN 201910120048 A CN201910120048 A CN 201910120048A CN 109857872 A CN109857872 A CN 109857872A
- Authority
- CN
- China
- Prior art keywords
- similar entities
- target entity
- type
- similarity
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000004364 calculation method Methods 0.000 claims abstract description 56
- 238000013507 mapping Methods 0.000 claims abstract description 36
- 238000012512 characterization method Methods 0.000 claims description 61
- 230000011218 segmentation Effects 0.000 claims description 21
- 238000010276 construction Methods 0.000 claims description 17
- 238000005516 engineering process Methods 0.000 claims description 15
- 238000001228 spectrum Methods 0.000 claims 1
- 230000008569 process Effects 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000003889 chemical engineering Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides the information recommendation methods and device of a kind of knowledge based map, the information recommendation method of the knowledge based map includes: when receiving the searching request for carrying target entity, knowledge mapping is generated for target entity, knowledge mapping includes that being associated between incidence relation and target entity and similar entities between each similar entities, each similar entities contacts;Similarity in calculation knowledge map between each similar entities and target entity;From high to low according to similarity, recommend similar entities for target entity.Scheme provided by the invention effectively improves the accuracy of recommendation.
Description
Technical field
The present invention relates to field of computer technology, in particular to a kind of the information recommendation method and dress of knowledge based map
It sets.
Background technique
Knowledge mapping (Knowledge Graph) is also known as mapping knowledge domains, and being known as knowledge domain in books and information group can
Map is mapped depending on change or ken, is explicit knowledge's development process and structural relation, to describe knowledge resource and its carrier,
Excavation, analysis, building, drafting and explicit knowledge and connecting each other between them.But the information content ratio that knowledge mapping provides
Larger, user is difficult quickly to search oneself desired content from knowledge mapping sometimes.
Summary of the invention
The embodiment of the invention provides the information recommendation methods and device of a kind of knowledge based map, effectively improve and push away
The accuracy recommended.
A kind of information recommendation method of knowledge based map, comprising:
When receiving the searching request for carrying target entity, knowledge mapping is generated for the target entity, it is described to know
Knowing map includes incidence relation between each similar entities, each similar entities and the target entity and the phase
Like the association connection between entity;
Calculate the similarity in the knowledge mapping between each similar entities and the target entity;
From high to low according to the similarity, recommend the similar entities for the target entity.
Preferably, the similarity calculated between each described similar entities and the target entity, comprising:
Using following first calculation formula, calculate similar between each described similar entities and the target entity
Degree;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information
The weighted value of type;Sj(A, Bji) characterize the corresponding message segment of j-th of information type and target entity A in i-th of similar entities B
Between similarity;N characterizes the corresponding information category total number of similar entities.
Preferably, the information recommendation method of above-mentioned knowledge based map further comprises: building class weight table, described
It include: the information type and be described that scientific and technological resources classification, each described scientific and technological resources classification include in class weight table
The weighted value of information type distribution;
Each described scientific and technological resources classification, meets following second calculation formula;
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N characterizes a kind of scientific and technological resources
The total number of information type in classification;
Following first calculation formula are utilized described, are calculated between each described similar entities and the target entity
Before similarity, further comprise:
Based on the class weight table, a target science and technology resource class is selected for the similar entities;
Determine the weighted value of the corresponding information type of the target science and technology resource class.
Preferably, when j-th of information type is numeric type, using following third calculation formula, it is real to calculate target
Similarity between body A and i-th of similar entities B;
Third calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of numerical value class and target entity A
Similarity;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) characterization i-th it is similar
In entity B in the corresponding message segment of numerical value class each element minimum value;Numeric type entity is equal in characterization target entity A
Value;Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B.
Preferably, when j-th of information type is list type, using following 4th calculation formula, it is real to calculate target
Similarity between body A and i-th of similar entities B;
4th calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;The element that characterization target entity A list type information includes is corresponding with list type in i-th of similar entities B
The element that message segment includes intersects number;GAThe element number that list type information includes in characterization target entity A;Characterization the
The element number that the corresponding message segment of list type includes in i similar entities B.
Preferably, when j-th of information type is text-type,
Using participle tool respectively to the text type information section and the target entity for including in the similar entities
It is segmented;
Each of word segmentation result participle is converted into corresponding participle vector;
Using following 5th calculation formula, the corresponding message segment of i-th of similar entities B, j-th of information type and mesh are calculated
Mark the similarity between entity A;
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;G characterizes mesh in word segmentation result
The total number for the participle that mark entity A includes;Characterize text-type message segment includes in i-th of similar entities B k-th point
The corresponding participle vector of word;Total of the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result
Number.
Preferably, the information recommendation method of above-mentioned knowledge based map further comprises:
For identical two similar entities of similarity, execute:
According between the incidence relation and the target entity and the similar entities between each similar entities
Association connection, identical two similar entities of statistics similarity arrive the shortest path of the target entity respectively;
It is described to recommend the similar entities for the target entity, comprising: identical two similar entities of similarity are directed to,
For the small similar entities of the target entity preferential recommendation shortest path.
A kind of information recommending apparatus of knowledge based map, comprising: map construction unit, similarity calculated and push away
Recommend unit, wherein
The map construction unit, for when receiving the searching request for carrying target entity, being that the target is real
Body generates knowledge mapping, the knowledge mapping include the incidence relation between each similar entities, each similar entities with
And being associated between the target entity and the similar entities contacts;
The similarity calculated, for calculating each in the knowledge mapping that the map construction unit generates
Similarity between similar entities and the target entity;
The recommendation unit is used for according to the calculated similarity of the similarity calculated from high to low, is described
Target entity recommends the similar entities.
Preferably,
The similarity calculated, for utilize following first calculation formula, calculate each described similar entities with
Similarity between the target entity;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information
The weighted value of type;Sj(A, Bji) characterize the corresponding message segment of j-th of information type and target entity A in i-th of similar entities B
Between similarity;N characterizes the corresponding information category total number of similar entities.
Preferably, above-mentioned apparatus further comprises: weight table construction unit, wherein
The weight table construction unit includes: scientific and technological resources class in the class weight table for constructing class weight table
Not, the information type that each described scientific and technological resources classification includes and the weighted value for information type distribution,
Each described scientific and technological resources classification, meets following second calculation formula,
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N characterizes a kind of scientific and technological resources
The total number of information type in classification;
The similarity calculated is further used for the class weight constructed based on the weight table construction unit
Table selectes a target science and technology resource class for the similar entities;Determine the corresponding information of the target science and technology resource class
The weighted value of type.
Preferably, the similarity calculated, is further used for
When j-th of information type is numeric type, using following third calculation formula, target entity A and i-th is calculated
Similarity between a similar entities B;
Third calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of numerical value class and target entity A
Similarity;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) characterization i-th it is similar
In entity B in the corresponding message segment of numerical value class each element minimum value;Numeric type entity is equal in characterization target entity A
Value;Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B;
When j-th of information type is list type, using following 4th calculation formula, target entity A and i-th is calculated
Similarity between a similar entities B;
4th calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;The element that characterization target entity A list type information includes is corresponding with list type in i-th of similar entities B
The element that message segment includes intersects number;GAThe element number that list type information includes in characterization target entity A;Characterization i-th
The element number that the corresponding message segment of list type includes in a similar entities B;
When j-th of information type is text-type,
Using participle tool respectively to the text type information section and the target entity for including in the similar entities
It is segmented;
Each of word segmentation result participle is converted into corresponding participle vector;
Using following 5th calculation formula, the corresponding message segment of i-th of similar entities B, j-th of information type and mesh are calculated
Mark the similarity between entity A;
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;G characterizes mesh in word segmentation result
The total number for the participle that mark entity A includes;Characterize text-type message segment includes in i-th of similar entities B k-th point
The corresponding participle vector of word;Total of the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result
Number.
The embodiment of the invention provides the information recommendation method and device of a kind of knowledge based map, the knowledge based maps
Information recommendation method generate knowledge mapping for target entity by when receiving the searching request for carrying target entity,
Knowledge mapping includes between incidence relation and target entity and similar entities between each similar entities, each similar entities
Association connection;I.e. knowledge mapping completely searches out similar entities relevant to target entity to come as far as possible, passes through calculating
Similarity in knowledge mapping between each similar entities and target entity;That is the high explanation of similarity is closer in user's
Search Requirement, then from high to low according to similarity, recommending similar entities for target entity, effectively improving the accurate of recommendation
Property.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is a kind of flow chart of the information recommendation method of knowledge based map provided by one embodiment of the present invention;
Fig. 2 is a kind of flow chart of the information recommendation method for knowledge based map that another embodiment of the present invention provides;
Fig. 3 is a kind of part-structure figure for knowledge mapping that another embodiment of the present invention provides;
Fig. 4 is a kind of structural representation of the information recommending apparatus of knowledge based map provided by one embodiment of the present invention
Figure;
Fig. 5 is a kind of structural representation of the information recommending apparatus for knowledge based map that another embodiment of the present invention provides
Figure.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments, based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
As shown in Figure 1, the embodiment of the invention provides a kind of information recommendation method of knowledge based map, this method can be with
The following steps are included:
Step 101: when receiving the searching request for carrying target entity, generating knowledge mapping for target entity, know
Knowledge map includes between incidence relation and target entity and similar entities between each similar entities, each similar entities
Association connection;
Step 102: the similarity in calculation knowledge map between each similar entities and target entity;
Step 103: from high to low according to similarity, recommending similar entities for target entity.
Wherein target entity can be word, figure, table, number etc..
Existing knowledge mapping generation technique can be directly used to complete in the generating process of above-mentioned knowledge mapping.
For sciemtifec and technical sphere, similar entities can be title such as document topic, can be the text in document text
Information, number or table etc..
In the embodiment shown in fig. 1, by when receiving the searching request for carrying target entity, being target entity
Knowledge mapping is generated, knowledge mapping includes the incidence relation and target entity between each similar entities, each similar entities
Being associated between similar entities contacts;I.e. knowledge mapping completely searches for similar entities relevant to target entity as far as possible
Out, pass through the similarity in calculation knowledge map between each similar entities and target entity;That is the high explanation of similarity
Closer in the Search Requirement of user, then from high to low according to similarity, recommending similar entities for target entity, effectively mentioning
The high accuracy recommended.
In an alternative embodiment of the invention in order to keep similarity result more accurate, by different type in similar entities
Message segment calculate separately similarity, and distribute corresponding weighted value for different information types, be based on different types of message segment
Similarity and weighted value, calculate the similarity of similar entities.Its specific embodiment can are as follows: calculates using following first public
Formula calculates the similarity between each similar entities and target entity;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information
The weighted value of type;Sj(A, Bji) characterize the corresponding message segment of j-th of information type and target entity A in i-th of similar entities B
Between similarity;N characterizes the corresponding information category total number of similar entities.
In an alternative embodiment of the invention, the information recommendation method of above-mentioned knowledge based map can also further comprise: structure
Build class weight table, include: scientific and technological resources classification, each scientific and technological resources classification information type for including in class weight table with
And the weighted value for information type distribution;Each scientific and technological resources classification meets following second calculation formula;
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N characterizes a kind of scientific and technological resources
The total number of information type in classification;
Following first calculation formula are being utilized, before calculating the similarity between each similar entities and target entity,
Further comprise: based on class weight table, selecting a target science and technology resource class for similar entities;Determine target scientific and technological resources
The weighted value of the corresponding information type of classification.
Such as: scientific and technological resources classification be title, it includes information type be text and numerical value, wherein the letter of text class
Ceasing the corresponding weighted value of type can be 0.8, and the corresponding weighted value of the information type of numerical value class can be 0.2;Scientific and technological resources classification is
Text, it includes information type be text, numerical value and list type, wherein the corresponding weighted value of the information type of text class
It can be 0.5, the corresponding weighted value of the information type of numerical value class can be 0.3;The corresponding weighted value of the information type of list type can be
0.2;Above-mentioned weighted value, scientific and technological resources classification and information type can further be adjusted according to search result.
For the similar entities that information type is numeric type, calculates its similarity between target entity and be embodied
Mode can are as follows: when j-th of information type is numeric type, using following third calculation formula, calculates target entity A and i-th
Similarity between similar entities B;
Third calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of numerical value class and target entity A
Similarity;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) characterization i-th it is similar
In entity B in the corresponding message segment of numerical value class each element minimum value;Numeric type entity is equal in characterization target entity A
Value;Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B.Wherein, for numeric type entity or number
Value type similar entities, element are numerical value one by one or numerical value a group by a group.
For the similar entities that information type is list type, calculates its similarity between target entity and be embodied
Mode can are as follows: when j-th of information type is list type, using following 4th calculation formula, calculates target entity A and i-th
Similarity between similar entities B;
4th calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;The element that characterization target entity A list type information includes is corresponding with list type in i-th of similar entities B
The element that message segment includes intersects number;GAThe element number that list type information includes in characterization target entity A;Characterization i-th
The element number that the corresponding message segment of list type includes in a similar entities B.Wherein, similar for list type entity or list type
Entity, element are the information in list in each table.
For the similar entities that information type is text-type, calculates its similarity between target entity and be embodied
Mode can are as follows: when j-th of information type is text-type, using participle tool respectively to including in the similar entities
Text type information section and the target entity are segmented;Each of word segmentation result participle is converted to corresponding point
Term vector;Using following 5th calculation formula, the corresponding message segment of i-th of similar entities B, j-th of information type and target are calculated
Similarity between entity A;
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;G characterizes mesh in word segmentation result
The total number for the participle that mark entity A includes;Characterize text-type message segment includes in i-th of similar entities B k-th point
The corresponding participle vector of word;Total of the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result
Number.Each participle, which is converted to corresponding participle vector, can be used existing LDA model calculating.
In an alternative embodiment of the invention, in order to further increase the accuracy of search, the above method further comprises: needle
Two similar entities identical to similarity, execute: according between each similar entities incidence relation and target entity with
Association connection between similar entities, identical two similar entities of statistics similarity arrive the shortest path of target entity respectively;
Recommend similar entities for target entity, comprising: be directed to identical two similar entities of similarity, most for target entity preferential recommendation
The small similar entities of short path.Wherein, shortest path refers to a similar entities to interval similar entities between target entity
Minimum number.
As shown in Fig. 2, the embodiment of the invention provides a kind of information recommendation method of knowledge based map, this method can be with
The following steps are included:
Step 200: building class weight table;
It include: that scientific and technological resources classification, each scientific and technological resources classification include in the class weight table that the step constructs
Information type and the weighted value distributed for information type;Meanwhile each scientific and technological resources classification, meet following second and calculates public affairs
Formula;
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N characterizes a kind of scientific and technological resources
The total number of information type in classification.
The information type and be letter that scientific and technological resources classification, each scientific and technological resources classification in category weight table include
The weighted value of breath type distribution can be adjusted according to the actual situation.
Step 201: when receiving the searching request for carrying target entity, generating knowledge mapping for target entity;
Knowledge mapping include incidence relation between each similar entities, each similar entities and target entity to it is similar
Association connection between entity, as shown in figure 3, the incidence relation between entity is indicated by connection mode.
Step 202: being based on class weight table, select a target science and technology resource class for similar entities;
The scientific and technological resources classification can be divided according to different demands, such as according to subject Type division economics class, society
Class, chemical engineering etc. can be learned, is for another example classified according to title and text etc..
Step 203: determining the weighted value of the corresponding information type of target science and technology resource class;
The weighted value can carry out setting according to the actual situation and correspondingly adjust.
Step 204: using the weighted value for the information type determined, each similar entities and mesh in calculation knowledge map
Mark the similarity between entity;
The step can utilize following first calculation formula, calculate between each described similar entities and the target entity
Similarity;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information
The weighted value of type;Sj(A, Bji) characterize the corresponding message segment of j-th of information type and target entity A in i-th of similar entities B
Between similarity;N characterizes the corresponding information category total number of similar entities.
Wherein, S (A, Bi) the specific calculating process of similarity between characterization target entity A and i-th of similar entities B can
Include:
When j-th of information type is numeric type, using following third calculation formula, target entity A and i-th of phase are calculated
Like the similarity between entity B;
Third calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of numerical value class and target entity A
Similarity;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) characterization i-th it is similar
In entity B in the corresponding message segment of numerical value class each element minimum value;Numeric type entity is equal in characterization target entity A
Value;Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B;
When j-th of information type is list type, using following 4th calculation formula, target entity A and i-th of phase are calculated
Like the similarity between entity B;
4th calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;The element that characterization target entity A list type information includes is corresponding with list type in i-th of similar entities B
The element that message segment includes intersects number;GAThe element number that list type information includes in characterization target entity A;Characterization the
The element number that the corresponding message segment of list type includes in i similar entities B;
When j-th of information type is text-type, using participle tool respectively to the text for including in the similar entities
Type information section and target entity are segmented;Each of word segmentation result participle is converted into corresponding participle vector;
Using following 5th calculation formula, calculate the corresponding message segment of i-th of similar entities B, j-th of information type and target entity A it
Between similarity;
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;G characterizes mesh in word segmentation result
The total number for the participle that mark entity A includes;Characterize text-type message segment includes in i-th of similar entities B k-th point
The corresponding participle vector of word;Total of the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result
Number.
Step 205: according to similarity from high to low and similar entities to target entity shortest path, be target entity
Recommend similar entities.
In this step, the shortest path of similar entities to target entity is for identical two similar entities of similarity
Execute, specific embodiment can include: according between each similar entities incidence relation and target entity to it is similar
Association connection between entity, identical two similar entities of statistics similarity arrive the shortest path of target entity respectively;For mesh
Mark the small similar entities of entity preferential recommendation shortest path.Knowledge mapping as shown in Figure 3, such as: similar entities 16 and target
The similarity of entity is greater than the similarity of similar entities 2 and target entity, then when recommending, similar entities 16 are in similar entities 2
Preferential recommendation before, for another example: similar entities 20 are similar to target entity equal to similar entities 6 to the similarity of target entity
The shortest path of degree, similar entities 20 to target entity is similar entities 20- similar entities 5- target entity, and similar entities 6 arrive
The shortest path of target entity is similar entities 6- target entity, then similar entities 6 are compared with similar entities 20, preferential recommendation phase
Like entity 6.
As shown in figure 4, the embodiment of the present invention provides a kind of information recommending apparatus of knowledge based map, comprising: map structure
Build unit 401, similarity calculated 402 and recommendation unit 403, wherein
Map construction unit 401, for being generated for target entity when receiving the searching request for carrying target entity
Knowledge mapping, knowledge mapping include incidence relation between each similar entities, each similar entities and target entity and phase
Like the association connection between entity;
Similarity calculated 402, for calculating each similar reality in the knowledge mapping that map construction unit 401 generates
Similarity between body and target entity;
Recommendation unit 403, for from high to low, being that target is real according to the calculated similarity of similarity calculated 402
Body recommends similar entities.
In an alternative embodiment of the invention, similarity calculated 402 are calculated for utilizing following first calculation formula
Similarity between each similar entities and target entity;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information
The weighted value of type;Sj(A, Bji) characterize the corresponding message segment of j-th of information type and target entity A in i-th of similar entities B
Between similarity;N characterizes the corresponding information category total number of similar entities.
As shown in figure 5, in still another embodiment of the process, the information recommending apparatus of above-mentioned knowledge based map, further
It include: weight table construction unit 501, wherein
Weight table construction unit 501 includes: scientific and technological resources classification in class weight table, every for constructing class weight table
A kind of information type that scientific and technological resources classification includes and the weighted value for information type distribution,
Each scientific and technological resources classification meets following second calculation formula,
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N characterizes a kind of scientific and technological resources
The total number of information type in classification;
Similarity calculated 402 is further used for the class weight table constructed based on weight table construction unit 501, is
Similar entities select a target science and technology resource class;Determine the weighted value of the corresponding information type of target science and technology resource class.
In an alternative embodiment of the invention, similarity calculated 402 are further used for:
When j-th of information type is numeric type, using following third calculation formula, target entity A and i-th is calculated
Similarity between a similar entities B;
Third calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of numerical value class and target entity A
Similarity;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) characterization i-th it is similar
In entity B in the corresponding message segment of numerical value class each element minimum value;Numeric type entity is equal in characterization target entity A
Value;Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B;
When j-th of information type is list type, using following 4th calculation formula, target entity A and i-th is calculated
Similarity between a similar entities B;
4th calculation formula:
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;The element that characterization target entity A list type information includes is corresponding with list type in i-th of similar entities B
The element that message segment includes intersects number;GAThe element number that list type information includes in characterization target entity A;Characterization the
The element number that the corresponding message segment of list type includes in i similar entities B;
When j-th of information type is text-type,
Using participle tool respectively to the text type information section and the target entity for including in the similar entities
It is segmented;
Each of word segmentation result participle is converted into corresponding participle vector;
Using following 5th calculation formula, the corresponding message segment of i-th of similar entities B, j-th of information type and mesh are calculated
Mark the similarity between entity A;
Wherein, Si(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of list type and target entity A
Similarity;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;G characterizes mesh in word segmentation result
The total number for the participle that mark entity A includes;Characterize text-type message segment includes in i-th of similar entities B k-th point
The corresponding participle vector of word;Total of the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result
Number.
The contents such as the information exchange between each unit, implementation procedure in above-mentioned apparatus, due to implementing with the method for the present invention
Example is based on same design, and for details, please refer to the description in the embodiment of the method for the present invention, and details are not described herein again.
The embodiment of the invention provides a kind of readable mediums, including execute instruction, when the processor of storage control executes
Described when executing instruction, the storage control executes the method that any of the above-described embodiment of the present invention provides.
The embodiment of the invention provides a kind of storage controls, comprising: processor, memory and bus;The memory
It is executed instruction for storing, the processor is connect with the memory by the bus, when the storage control is run
When, the processor executes the described of memory storage and executes instruction, so that the storage control executes in the present invention
The method that any embodiment offer is provided.
In conclusion more than the present invention each embodiment at least has the following beneficial effects:
1, in embodiments of the present invention, by when receiving the searching request for carrying target entity, being target entity
Knowledge mapping is generated, knowledge mapping includes the incidence relation and target entity between each similar entities, each similar entities
Being associated between similar entities contacts;I.e. knowledge mapping completely searches for similar entities relevant to target entity as far as possible
Out, pass through the similarity in calculation knowledge map between each similar entities and target entity;That is the high explanation of similarity
Closer in the Search Requirement of user, then from high to low according to similarity, recommending similar entities for target entity, effectively mentioning
The high accuracy recommended.
It should be noted that, in this document, such as first and second etc relational terms are used merely to an entity
Or operation is distinguished with another entity or operation, is existed without necessarily requiring or implying between these entities or operation
Any actual relationship or order.Moreover, the terms "include", "comprise" or its any other variant be intended to it is non-
It is exclusive to include, so that the process, method, article or equipment for including a series of elements not only includes those elements,
It but also including other elements that are not explicitly listed, or further include solid by this process, method, article or equipment
Some elements.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including
There is also other identical factors in the process, method, article or equipment of the element.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can store in computer-readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed;And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
In the various media that can store program code such as disk.
Finally, it should be noted that the foregoing is merely presently preferred embodiments of the present invention, it is merely to illustrate skill of the invention
Art scheme, is not intended to limit the scope of the present invention.Any modification for being made all within the spirits and principles of the present invention,
Equivalent replacement, improvement etc., are included within the scope of protection of the present invention.
Claims (10)
1. a kind of information recommendation method of knowledge based map characterized by comprising
When receiving the searching request for carrying target entity, knowledge mapping, the knowledge graph are generated for the target entity
Spectrum includes incidence relation between each similar entities, each similar entities and the target entity and the similar reality
Association connection between body;
Calculate the similarity in the knowledge mapping between each similar entities and the target entity;
From high to low according to the similarity, recommend the similar entities for the target entity.
2. the information recommendation method of knowledge based map according to claim 1, which is characterized in that described to calculate each
Similarity between the similar entities and the target entity, comprising:
Using following first calculation formula, the similarity between each described similar entities and the target entity is calculated;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information type
Weighted value;Sj(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of j-th of information type and target entity A
Similarity;N characterizes the corresponding information category total number of similar entities.
3. the information recommendation method of knowledge based map according to claim 2, which is characterized in that
Further comprise: constructing class weight table, include: scientific and technological resources classification, each described section in the class weight table
The information type that skill resource class includes and the weighted value for information type distribution;
Each described scientific and technological resources classification, meets following second calculation formula;
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N is characterized in a kind of scientific and technological resources classification
The total number of information type;
Following first calculation formula are utilized described, are calculated similar between each described similar entities and the target entity
Before degree, further comprise:
Based on the class weight table, a target science and technology resource class is selected for the similar entities;
Determine the weighted value of the corresponding information type of the target science and technology resource class.
4. the information recommendation method of knowledge based map according to claim 2 or 3, which is characterized in that
When j-th of information type is numeric type, using following third calculation formula, target entity A and i-th of phase are calculated
Like the similarity between entity B;
Third calculation formula:
Wherein, Si(A, Bji) characterization i-th of similar entities B in it is similar between the corresponding message segment of numerical value class and target entity A
Degree;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) i-th of similar entities of characterization
In B in the corresponding message segment of numerical value class each element minimum value;Characterize the mean value of numeric type entity in target entity A;
Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B.
5. the information recommendation method of knowledge based map according to claim 2 or 3, which is characterized in that
When j-th of information type is list type, using following 4th calculation formula, target entity A and i-th of phase are calculated
Like the similarity between entity B;
4th calculation formula:
Wherein, Si(A, Bji) characterization i-th of similar entities B in it is similar between the corresponding message segment of list type and target entity A
Degree;The element information corresponding with list type in i-th of similar entities B that characterization target entity A list type information includes
The element intersection number that section includes;GAThe element number that list type information includes in characterization target entity A;I-th of characterization
The element number that the corresponding message segment of list type includes in similar entities B.
6. the information recommendation method of knowledge based map according to claim 2 or 3, which is characterized in that
When j-th of information type is text-type, using participle tool respectively to the text for including in the similar entities
Type information section and the target entity are segmented;By each of word segmentation result participle be converted to it is corresponding segment to
Amount;Using following 5th calculation formula, the corresponding message segment of i-th of similar entities B, j-th of information type and target entity are calculated
Similarity between A;
Wherein, Si(A, Bji) characterization i-th of similar entities B in it is similar between the corresponding message segment of list type and target entity A
Degree;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;It is real that g characterizes target in word segmentation result
The total number for the participle that body A includes;Characterize k-th of participle pair that text-type message segment includes in i-th of similar entities B
The participle vector answered;The total number for the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result.
7. the information recommendation method of knowledge based map according to any one of claims 1 to 3, which is characterized in that further
Include:
For identical two similar entities of similarity, execute:
According to the pass between the incidence relation and the target entity and the similar entities between each similar entities
Connection connection, identical two similar entities of statistics similarity arrive the shortest path of the target entity respectively;
It is described to recommend the similar entities for the target entity, comprising: identical two similar entities of similarity to be directed to, for institute
State the small similar entities of target entity preferential recommendation shortest path.
8. a kind of information recommending apparatus of knowledge based map characterized by comprising map construction unit, similarity calculation
Unit and recommendation unit, wherein
The map construction unit, for when receiving the searching request for carrying target entity, being that the target entity is raw
At knowledge mapping, the knowledge mapping includes the incidence relation and institute between each similar entities, each similar entities
Being associated between target entity and the similar entities is stated to contact;
The similarity calculated, it is similar for calculating each in the knowledge mapping that the map construction unit generates
Similarity between entity and the target entity;
The recommendation unit, for from high to low, being the target according to the calculated similarity of the similarity calculated
Entity recommends the similar entities.
9. device according to claim 8, which is characterized in that
The similarity calculated, for utilize following first calculation formula, calculate each described similar entities with it is described
Similarity between target entity;
Wherein, S (A, Bi) characterization target entity A and i-th of similar entities B between similarity;wjCharacterize j-th of information type
Weighted value;Sj(A, Bji) characterize in i-th of similar entities B between the corresponding message segment of j-th of information type and target entity A
Similarity;N characterizes the corresponding information category total number of similar entities.
10. device according to claim 9, which is characterized in that
Further comprise: weight table construction unit, wherein
The weight table construction unit, for constructing class weight table, include: in the class weight table scientific and technological resources classification,
The information type that each described scientific and technological resources classification includes and the weighted value for information type distribution,
Each described scientific and technological resources classification, meets following second calculation formula,
Wherein, wjCharacterize a kind of weighted value of j-th of information type in scientific and technological resources classification;N is characterized in a kind of scientific and technological resources classification
The total number of information type;
The similarity calculated is further used for the class weight table constructed based on the weight table construction unit,
A target science and technology resource class is selected for the similar entities;Determine the corresponding information type of the target science and technology resource class
Weighted value;
And/or
The similarity calculated, is further used for:
When j-th of information type is numeric type, using following third calculation formula, target entity A and i-th of phase are calculated
Like the similarity between entity B;
Third calculation formula:
Wherein, Si(A, Bji) characterization i-th of similar entities B in it is similar between the corresponding message segment of numerical value class and target entity A
Degree;Max (A) characterizes the maximum value of each element in numeric type entity in target entity A;min(Bji) i-th of similar entities of characterization
In B in the corresponding message segment of numerical value class each element minimum value;Characterize the mean value of numeric type entity in target entity A;
Characterize the average value of the corresponding message segment of numerical value class in i-th of similar entities B;
When j-th of information type is list type, using following 4th calculation formula, target entity A and i-th of phase are calculated
Like the similarity between entity B;
4th calculation formula:
Wherein, Si(A, Bji) characterization i-th of similar entities B in it is similar between the corresponding message segment of list type and target entity A
Degree;The element information corresponding with list type in i-th of similar entities B that characterization target entity A list type information includes
The element intersection number that section includes;GAThe element number that list type information includes in characterization target entity A;I-th of characterization
The element number that the corresponding message segment of list type includes in similar entities B;
When j-th of information type is text-type,
The text type information section for including in the similar entities and the target entity are carried out respectively using participle tool
Participle;
Each of word segmentation result participle is converted into corresponding participle vector;
Using following 5th calculation formula, calculates the corresponding message segment of i-th of similar entities B, j-th of information type and target is real
Similarity between body A;
Wherein, Si(A, Bji) characterization i-th of similar entities B in it is similar between the corresponding message segment of list type and target entity A
Degree;FAtThe corresponding participle vector of t-th of participle that characterization target entity A text information includes;It is real that g characterizes target in word segmentation result
The total number for the participle that body A includes;Characterize k-th of participle pair that text-type message segment includes in i-th of similar entities B
The participle vector answered;The total number for the participle that text-type message segment includes in i-th of similar entities B in m characterization word segmentation result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910120048.9A CN109857872A (en) | 2019-02-18 | 2019-02-18 | The information recommendation method and device of knowledge based map |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910120048.9A CN109857872A (en) | 2019-02-18 | 2019-02-18 | The information recommendation method and device of knowledge based map |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109857872A true CN109857872A (en) | 2019-06-07 |
Family
ID=66898279
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910120048.9A Pending CN109857872A (en) | 2019-02-18 | 2019-02-18 | The information recommendation method and device of knowledge based map |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109857872A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111091454A (en) * | 2019-11-05 | 2020-05-01 | 新华智云科技有限公司 | Financial public opinion recommendation method based on knowledge graph |
CN111325033A (en) * | 2020-03-20 | 2020-06-23 | 中国建设银行股份有限公司 | Entity identification method, entity identification device, electronic equipment and computer readable storage medium |
CN112015911A (en) * | 2020-08-24 | 2020-12-01 | 上海松鼠课堂人工智能科技有限公司 | Method for searching massive knowledge maps |
CN112765367A (en) * | 2021-01-28 | 2021-05-07 | 杭州泰一指尚科技有限公司 | Method and device for constructing theme knowledge graph |
CN112988135A (en) * | 2021-05-20 | 2021-06-18 | 中国人民解放军国防科技大学 | Task unit recommendation method and device for open source software and computer equipment |
CN113806558A (en) * | 2021-09-22 | 2021-12-17 | 湖北天天数链技术有限公司 | Question selection method, knowledge graph construction method and device and electronic equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013061371A1 (en) * | 2011-10-26 | 2013-05-02 | 株式会社日立製作所 | Information processing device and method |
CN103631859A (en) * | 2013-10-24 | 2014-03-12 | 杭州电子科技大学 | Intelligent review expert recommending method for science and technology projects |
CN106250412A (en) * | 2016-07-22 | 2016-12-21 | 浙江大学 | The knowledge mapping construction method merged based on many source entities |
CN107169010A (en) * | 2017-03-31 | 2017-09-15 | 北京奇艺世纪科技有限公司 | A kind of determination method and device of recommendation search keyword |
WO2018049960A1 (en) * | 2016-09-14 | 2018-03-22 | 厦门幻世网络科技有限公司 | Method and apparatus for matching resource for text information |
-
2019
- 2019-02-18 CN CN201910120048.9A patent/CN109857872A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013061371A1 (en) * | 2011-10-26 | 2013-05-02 | 株式会社日立製作所 | Information processing device and method |
CN103631859A (en) * | 2013-10-24 | 2014-03-12 | 杭州电子科技大学 | Intelligent review expert recommending method for science and technology projects |
CN106250412A (en) * | 2016-07-22 | 2016-12-21 | 浙江大学 | The knowledge mapping construction method merged based on many source entities |
WO2018049960A1 (en) * | 2016-09-14 | 2018-03-22 | 厦门幻世网络科技有限公司 | Method and apparatus for matching resource for text information |
CN107169010A (en) * | 2017-03-31 | 2017-09-15 | 北京奇艺世纪科技有限公司 | A kind of determination method and device of recommendation search keyword |
Non-Patent Citations (1)
Title |
---|
李阳等: "知识图谱中实体相似度计算研究", 《中文信息学报》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111091454A (en) * | 2019-11-05 | 2020-05-01 | 新华智云科技有限公司 | Financial public opinion recommendation method based on knowledge graph |
CN111325033A (en) * | 2020-03-20 | 2020-06-23 | 中国建设银行股份有限公司 | Entity identification method, entity identification device, electronic equipment and computer readable storage medium |
CN112015911A (en) * | 2020-08-24 | 2020-12-01 | 上海松鼠课堂人工智能科技有限公司 | Method for searching massive knowledge maps |
CN112015911B (en) * | 2020-08-24 | 2021-07-20 | 上海松鼠课堂人工智能科技有限公司 | Method for searching massive knowledge maps |
CN112765367A (en) * | 2021-01-28 | 2021-05-07 | 杭州泰一指尚科技有限公司 | Method and device for constructing theme knowledge graph |
CN112988135A (en) * | 2021-05-20 | 2021-06-18 | 中国人民解放军国防科技大学 | Task unit recommendation method and device for open source software and computer equipment |
CN112988135B (en) * | 2021-05-20 | 2021-07-27 | 中国人民解放军国防科技大学 | Task unit recommendation method and device for open source software and computer equipment |
CN113806558A (en) * | 2021-09-22 | 2021-12-17 | 湖北天天数链技术有限公司 | Question selection method, knowledge graph construction method and device and electronic equipment |
CN113806558B (en) * | 2021-09-22 | 2024-03-26 | 湖北天天数链技术有限公司 | Question selection method, knowledge graph construction device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109857872A (en) | The information recommendation method and device of knowledge based map | |
CN106528656B (en) | A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter | |
CN103544663B (en) | The recommendation method of network open class, system and mobile terminal | |
Luo et al. | A context-aware personalized resource recommendation for pervasive learning | |
CN103365997B (en) | A kind of opining mining method based on integrated study | |
CN107451619A (en) | A kind of small target detecting method that confrontation network is generated based on perception | |
CN107256230B (en) | Fusion method based on diversified geographic information points | |
CN104536881A (en) | Public testing error report priority sorting method based on natural language analysis | |
CN103714413A (en) | Position information-based competence model construction system and method | |
JP2016538615A (en) | Main knowledge point recommendation method and system | |
CN103744889B (en) | A kind of method and apparatus for problem progress clustering processing | |
CN103559303A (en) | Evaluation and selection method for data mining algorithm | |
CN105373546B (en) | A kind of information processing method and system for knowledge services | |
CN109657159A (en) | The determination method of the transfer learning boundary of isomeric relationship data in public sentiment data role identification | |
KR102203048B1 (en) | Method for recommending education program in education service based on the simirality and system thereof | |
Eikelboom et al. | A spatial optimization algorithm for geodesign | |
CN108628967A (en) | A kind of e-learning group partition method generating network similarity based on study | |
CN107391577A (en) | A kind of works label recommendation method and system based on expression vector | |
Ran et al. | Combining grey relational analysis and TOPSIS concepts for evaluating the technical innovation capability of high technology enterprises with fuzzy information | |
CN107305559A (en) | Method and apparatus are recommended in one kind application | |
CN103218419B (en) | Web tab clustering method and system | |
CN111339258B (en) | University computer basic exercise recommendation method based on knowledge graph | |
Salehi et al. | Attribute-based recommender system for learning resource by learner preference tree | |
CN109299884A (en) | A kind of influence power appraisal procedure and assessment device | |
CN110262906B (en) | Interface label recommendation method and device, storage medium and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |