CN109657060A - safety production accident case pushing method and system - Google Patents
safety production accident case pushing method and system Download PDFInfo
- Publication number
- CN109657060A CN109657060A CN201811571338.7A CN201811571338A CN109657060A CN 109657060 A CN109657060 A CN 109657060A CN 201811571338 A CN201811571338 A CN 201811571338A CN 109657060 A CN109657060 A CN 109657060A
- Authority
- CN
- China
- Prior art keywords
- case
- similarity
- data
- industrial accident
- cases
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention relates to the technical field of intelligent pushing, in particular to a safe production accident case pushing method and system. The pushing method comprises the following steps: analyzing the data information of the case to respectively acquire attribute text data and numerical data of the case; respectively calculating the overall similarity of each two cases; clustering the multiple cases according to the overall similarity between every two cases to generate multiple case clusters; and pushing at least one case in the case cluster where the target case is located. According to the safety production accident case pushing method and system, data information of cases is divided into attribute text data and numerical data, the similarity of the two types of data between the cases is calculated respectively, then the overall similarity of the two cases is calculated according to the similarity of the two types of data between the cases, and finally clustering is carried out based on the overall similarity between the cases to form case clusters, so that the calculation efficiency and the pushing accuracy are improved.
Description
Technical field
The present invention relates to intelligent push fields, and in particular to a kind of industrial accident case method for pushing and is
System.
Background technique
It is analysis and decision to effectively assist solution of emergent event when safety in production emergency event next time occurs
Personnel provide the reference information of similar case, it is necessary to which the intelligent push of research safety production accident case passes through input
The target case of one concern, system automatically retrieval goes out similar cases and carries out being pushed to policymaker's offer reference, and then improves
The decision-making capability of decision support system for emergency response.
Intelligent push in the prior art generally uses single similarity calculation, and the scope of application is more limited to, so
And aspect data structure of keeping the safety in production is divided into text and two kinds of numerical value, intelligently pushing method in the prior art, which is not able to satisfy, to be needed
It asks.
In consideration of it, overcoming the above defect in the prior art, a kind of new industrial accident case method for pushing is provided
And system becomes this field technical problem urgently to be resolved.
Summary of the invention
It is an object of the invention in view of the above drawbacks of the prior art, provide a kind of industrial accident case push side
Method and system.
First aspect present invention provides a kind of industrial accident case method for pushing, which includes:
The data information of industrial accident case is analyzed, to obtain the attribute text of industrial accident case
Data and numeric type data;
Calculate separately the first similarity between the attribute text data of every two case and between numeric type data
Two similarities, it is similar according to the totality between the first similarity of every two case and the second similarity calculation every two case
Degree;
Multiple cases are clustered according to the overall similarity between the every two case, to generate multiple cases
Cluster;
Target case is received, obtains the case cluster where target case, and select the case cluster where the target case
In at least one case pushed.
Preferably, the calculating step of the first similarity between the attribute text data of every two case includes:
Case G is calculated separately based on WordNet semantic concept tree distanceiWith case GjThe similarity of each single item attribute text
simAk(Gi,Gj),
Wherein, depth (Gi,k) it is case GiDepth of the K attribute texts in semantic concept tree, depth (GJ, k)
For case GjDepth of the K attribute texts in semantic concept tree, depth (lso (Gi,k,Gj,k) it is case GiK
Attribute text and case GjK attribute texts depth of the nearest common initial data in semantic concept tree, k 1,
2 ..., n, n are the item number of attribute text in attribute text data;
According to case GiWith case GjThe similarity simA of each single item attribute textk(Gi, Gj) and each single item attribute text
The weight coefficient w in bulk properties text data similaritykCalculate case GiWith case GjThe first similarity, whereinK is 1,2 ..., n, and n is attribute text in attribute text data
This item number.
Preferably, the calculating step of the second similarity between the numeric type data of every two case includes:
Case G is calculated based on European space distanceiWith case GjNumeric type data the second similarity simC (Gi,Gj),
Wherein,For case GiY numeric type datas numerical value,For case GjY numeric type numbers
According to numerical value, w 'yFor y numeric type datas, weight coefficient, y 1,2 ..., m, m are in overall numerical value type data similarity
The item number of numeric type data.
Preferably, described " according between the first similarity of every two case and the second similarity calculation every two case
Overall similarity " the step of include:
Overall similarity sim (Gi,Gj)=α × simA (Gi,Gj)+β×simC(Gi,Gj),
Wherein, α, β are respectively the first similarity and the second similarity shared weight in overall similarity.
Preferably, described " multiple cases to be clustered according to the overall similarity between the every two case, with life
At multiple case clusters " the step of include:
S1 randomly chooses a untreated case from case total collection, searches in case total collection untreated with this
The overall similarity of case is less than all cases of first threshold, when the quantity of the case searched is greater than or equal to second threshold
When, case cluster is established by core of the untreated case, and the case searched is added in the candidate collection of the case cluster;
It is noise by the untreated case marker when the quantity of the case searched is less than second threshold;
To the untreated case of each of candidate collection the case cluster is added, and always collect in case in the case by S2
All cases for being less than first threshold with the overall similarity of the case are searched in conjunction, be greater than when the quantity of the case searched or
When equal to second threshold, the case searched is continuously added in candidate collection;
S3 repeats step S2, until untreated case is not present in candidate collection;
S4 repeats step S1 to S3, until untreated case is not present in case total collection.
Preferably, the method for pushing further include:
History industrial accident is obtained, the data information of the history industrial accident case is analyzed, with
The attribute text data and numeric type data of history industrial accident case are obtained, respectively to establish case library.
Preferably, the attribute text data includes accident pattern, accident description, cause of accident, accident spot and calls to account
Situation;The numeric type data includes story time, causality loss and punishment situation.
Second aspect of the present invention provides a kind of industrial accident case supplying system, which includes:
Analysis of cases module is analyzed for the data information to industrial accident case, to obtain safety in production
The attribute text data and numeric type data of accident case;
Similarity calculation module, the first similarity between attribute text data for calculating separately every two case,
The second similarity between numeric type data, according to the first similarity of every two case and the second similarity calculation every two
Overall similarity between case;
Cluster module, for being clustered according to the overall similarity between the every two case to multiple cases, with
Generate multiple case clusters;
Pushing module obtains the case cluster where target case, and select the target case for receiving target case
At least one case in the case cluster at place is pushed.
Preferably, the supplying system further include:
Case library, for storing the attribute text data and numeric type data of history industrial accident case.
Preferably, the supplying system further include:
Interactive module, for receiving the target case query information of user's input.
The data information of case is divided into attribute textual data by industrial accident case method for pushing of the invention and system
According to and numeric type data, the similarity of two types data between case is calculated separately first, then according to two kinds between case
The overall similarity of two cases of similarity calculation of data, finally carries out cluster formation based on the overall similarity between case
Case cluster improves the efficiency of calculating and the precision of push.
Detailed description of the invention
Fig. 1 is the flow chart of the industrial accident case method for pushing of first embodiment of the invention.
Fig. 2 be in the method for pushing of the preferred embodiment of the present invention semantic concept tree apart from schematic illustration.
Fig. 3 is that DBSCAN clusters candidate collection schematic illustration in the method for pushing of the preferred embodiment of the present invention.
Fig. 4 is the flow chart of the industrial accident case method for pushing of second embodiment of the invention.
Fig. 5 is the software architecture diagram of the industrial accident case supplying system of first embodiment of the invention.
Fig. 6 is the software architecture diagram of the industrial accident case supplying system of second embodiment of the invention.
Fig. 7 is the hardware block diagram of the industrial accident case supplying system of third embodiment of the invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawing and specific implementation
Invention is further described in detail for example.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention,
It is not intended to limit the present invention.
In order to keep the narration of this disclosure more detailed with it is complete, below for embodiments of the present invention and specific real
It applies example and proposes illustrative description;But this not implements or uses the unique forms of the specific embodiment of the invention.Embodiment
In cover multiple specific embodiments feature and to construction with operate these specific embodiments method and step it is suitable with it
Sequence.However, can also reach identical or impartial function and sequence of steps using other specific embodiments.
The industrial accident case method for pushing that the embodiment of the present invention proposes, the data structure based on industrial accident
The data information of industrial accident case is divided into attribute text data (semantic category) and numeric type data (numerical value by feature
Class), specifically, attribute text data can include but is not limited to following: accident pattern, accident description, cause of accident, accident
Point and situation of calling to account;Numeric type data can include but is not limited to following: story time, causality loss and punishment situation.This hair
The method for pushing of bright embodiment calculates separately two types data between case when calculating similarity between two cases first
Similarity, then according to the overall similarity of two cases of similarity calculation of two kinds of data between case.Finally, being based on again
Overall similarity between case carries out cluster and forms case cluster, makees at least one case where target case in case cluster
For target case, my similar cases are pushed.
Fig. 1 is industrial accident case method for pushing according to an embodiment of the invention, refering to Figure 1, should
Method for pushing includes:
S101 analyzes the data information of industrial accident case, to obtain the category of industrial accident case
Property text data and numeric type data.
S102, calculate separately the first similarity between the attribute text data of every two case and numeric type data it
Between the second similarity, according to the totality between the first similarity of every two case and the second similarity calculation every two case
Similarity.
S103 clusters multiple cases according to the overall similarity between the every two case, to generate multiple cases
Example cluster.
S104 receives target case, obtains the case cluster where target case, and select the case where the target case
At least one case in cluster is pushed.
In step s101, industrial accident case include target case and except goal-trail exception case to be pushed,
The data information of all cases is analyzed, the data information of case is divided into attribute text data and numeric type data two
Class.Wherein, attribute text data includes multinomial attribute text, it may for example comprise the 1st attribute text, the 2nd attribute text ...,
N-th attribute text;Numeric type data includes multinomial data, it may for example comprise the 1st numeric type data, the 2nd numeric type number
According to ..., m numeric type datas;N and m is natural number.
In step s 102, based on the Semantic feature of attribute text data, consider two case GiAnd GjKth item category
Property text GI, kAnd GJ, kSimilitude when, by the attribute text of two cases in the distance in WordNet semantic concept tree in terms of
Calculate similarity simAk(Gi, Gj), then the similarity of each single item attribute text is weighted and averaged to calculate the first similarity
simA(Gi, Gj)。
Specifically, firstly, calculating separately case G according to following formula based on WordNet semantic concept tree distanceiAnd case
GjThe similarity simA of each single item attribute textk(Gi,Gj),
Wherein, depth (Gi,k) it is case GiDepth of the K attribute texts in semantic concept tree, depth (Gj,k)
For case GjDepth of the K attribute texts in semantic concept tree, depth (lso (Gi,k, GJ, k) it is case GiK
Attribute text and case GjK attribute texts depth of the nearest common initial data in semantic concept tree, k 1,
2 ..., n, n are the item number of attribute text in attribute text data.
Specifically, it please refers to shown in Fig. 2, in Fig. 2, Gi,kAnd Gj,kDepth be 4, that is to say, that depth (Gi,k)
For 4, depth (Gj,k) it is the nearest common initial data that A point shown in 4, Fig. 2 is, the depth of A point is 3, that is to say, that lso
(GI, k,Gj,k) it is 3.
Then, according to case GiWith case GjThe similarity sjmA of each single item attribute textk(Gi,Gj) and each single item attribute
Text weight coefficient w in bulk properties text data similaritykCalculate case GiWith case GjThe first similarity, whereinK is 1,2 ..., n, and n is attribute text in attribute text data
This item number.
In step s 102, the characteristics of being based on numeric type data calculates case G based on European space distanceiWith case Gj's
Second similarity simC (G of numeric type datai, Gj),
Wherein,For case GiY numeric type datas numerical value,For case GjY numeric type numbers
According to numerical value, w 'yFor y numeric type datas, weight coefficient, y 1,2 ..., m, m are in overall numerical value type data similarity
The item number of numeric type data.
Finally, according to the first similarity and the second similarity calculation overall similarity, overall similarity sim (Gi, Gj)=α
×simA(Gi, Gj)+β×simC(Gi, Gj), wherein α, β are respectively the first similarity and the second similarity in overall similarity
In shared weight.
In step s 103, with DBSCAN (Density-Based Spatial Clustering of
Applications with Noise) clustering algorithm clusters all cases, and DBSCAN is to have noisy to be based on density
Clustering method, be a kind of density-based spatial clustering algorithm, which is cluster by the region division with sufficient density, and
Cluster, is defined as the maximum set of the connected point of density by the cluster that arbitrary shape is found in having noisy spatial database.Tool
Body, all cases are all sorted out or labeled as noise, in the present specification, the untreated case representation case did not both have
It is included into any one case cluster and is also not flagged as noise.
Firstly, all cases are included into case total collection Z, when initial, all cases are in case total collection Z
Untreated case determines search radius e and minimal amount minPts, for example, e is first threshold, minPts is second threshold.
Then, initial case is randomly choosed from case total collection Z, e neighborhood density measurement, search is carried out to initial case
It is less than all cases of first threshold with the overall similarity of the initial case, when the quantity of the case searched is greater than or equal to
When second threshold, case cluster C1 is established by core of the initial case, and candidate collection N1 is added in the case searched;When searching
It is noise by initial case marker when the quantity for the case that rope arrives is less than second threshold.That is, if the phase of initial case
It is sufficiently large like case density, so that it may case cluster to be established using it as core case, if the similar cases density of initial case is not
It is enough big, just it is regarded as noise.
Then, to the untreated case of each of candidate collection N1, above-mentioned e neighborhood density measurement will be carried out, first will
Selected case is added in case cluster C1, is further continued for searching for the overall similarity with the case in case total collection Z less than first
The case searched is continued to add by all cases of threshold value when the quantity of the case searched is greater than or equal to second threshold
Enter candidate collection N1, at this point, candidate collection N1 continues to expand, please refers to shown in Fig. 3.This step is repeated, until candidate collection N1
In be not present untreated case, at this point, case cluster C1 foundation finish.
Then, second untreated case is randomly choosed from case total collection Z, continues above-mentioned steps, it is total in case
All cases for being less than first threshold with the overall similarity of the case are searched in set Z, when the quantity of the case searched is big
When second threshold, case cluster C2 is established by core of the case, and case cluster C2 is added in the case searched
Candidate collection N2 in;It is noise by the case marker when the quantity of the case searched is less than second threshold.
To the untreated case of each of candidate collection N2, which is added case cluster C2, and always collect in case
All cases that search in Z is less than first threshold with the overall similarity of the case are closed, when the quantity of the case searched is greater than
Or when being equal to second threshold, the case searched is continuously added in candidate collection N2;This step is repeated, until candidate collection N2
In be not present untreated case.
It repeats the above steps, until untreated case is not present in case total collection Z.
In step S104, according to step S103 cluster as a result, other cases of case cluster are where target case
The similar cases of target case can be pushed.
Fig. 4 is industrial accident case method for pushing according to an embodiment of the invention, is please referred to shown in Fig. 2, should
Method for pushing includes:
S201 obtains history industrial accident, analyzes the data information of the history industrial accident case,
To obtain the attribute text data and numeric type data of history industrial accident case, to establish case library.
S202 receives the target case or target case query information of user's input.
S203, calculate separately the first similarity between the attribute text data of every two case and numeric type data it
Between the second similarity, according to the totality between the first similarity of every two case and the second similarity calculation every two case
Similarity.
S204 clusters multiple cases according to the overall similarity between the every two case, to generate multiple cases
Example cluster.
S205 pushes at least one case in the case cluster where target case.
In step s 201, the data information of history industrial accident case is analyzed, the history peace that will acquire
The attribute text data and numeric type data of full production accident case are stored in case library, to accelerate calculating speed.
In step S202, the target case query information of user's input is received, when the target case is stored in case library
When middle, it is directly entered subsequent similarity calculation and sorting procedure.When the target case is not stored in case library, by user
Input the information of the target case, or search for the information of the target case automatically from network, then to the information of target case into
Row analysis, to obtain the attribute text data and numeric type data of target case respectively, and is stored in case library.
In step S203, the overall similarity of every two case in case library is calculated, is clustered in step S204,
Similarity calculation and clustering processing please refer to the related description of first embodiment, herein without repeating one by one.
Based on the same inventive concept, a kind of industrial accident case supplying system is additionally provided in the embodiment of the present invention,
Such as the following examples.The principle and above-mentioned industrial accident solved the problems, such as due to industrial accident case supplying system
Case method for pushing is similar, therefore the implementation of industrial accident case supplying system may refer to above-mentioned industrial accident
The implementation of case method for pushing, overlaps will not be repeated.It is used below, term " unit " either " submodule " or
The combination of the software and/or hardware of predetermined function may be implemented in " module ".Although mobile terminal described in following embodiment
Functional module preferably realized with software, but the combined realization of hardware or software and hardware be also may and by structure
Think.
Fig. 5 is the functional module signal of the industrial accident case supplying system of one embodiment provided by the invention
Figure.The industrial accident case supplying system 100 of the present embodiment include: analysis of cases module 10, similarity calculation module 20,
Cluster module 30 and pushing module 40, wherein analysis of cases module 10 be used for the data information of industrial accident case into
Row analysis, to obtain the attribute text data and numeric type data of industrial accident case;Similarity calculation module 20 is used for
Calculate separately the first similarity between the attribute text data of every two case and second between numeric type data similar
Degree, according to the overall similarity between the first similarity of every two case and the second similarity calculation every two case;Cluster
Module 30 is for clustering multiple cases according to the overall similarity between the every two case, to generate multiple cases
Cluster;Pushing module 40 obtains the case cluster where target case for receiving target case, and where selecting the target case
At least one case in case cluster is pushed.
Fig. 6 is the functional module signal of the industrial accident case supplying system of second embodiment provided by the invention
Figure.Further comprise on the basis of the present embodiment first embodiment shown in Fig. 5: case library 50 and interactive module 60, wherein case
Example library 50 is used to store the attribute text data and numeric type data of history industrial accident case;Interactive module 60 is for connecing
Receive the target case query information of user's input.
Fig. 7 is the hardware module signal of the industrial accident case supplying system of third embodiment provided by the invention
Figure.The system 100 may include: processor 1001, such as CPU, communication bus 1002, memory 1003.
Wherein, communication bus 1002 is for realizing the connection communication between these components.Memory 1003 can be high speed
RAM memory is also possible to stable memory (non-volatile memory), such as magnetic disk storage.Memory 1003
It optionally can also be the storage device independently of aforementioned processor 1001.
It will be understood by those skilled in the art that the restriction of the not structure paired terminal of terminal structure shown in Fig. 7, can wrap
It includes than illustrating more or fewer components, perhaps combines certain components or different component layouts.
As shown in fig. 7, as may include operating system and safety in a kind of memory 1003 of computer storage medium
Production accident case pushes processing routine.Processor 1001 can be used for calling the industrial accident stored in memory 1003
Case pushes processing routine, and executes the operating procedure in industrial accident case method for pushing.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of industrial accident case method for pushing, which is characterized in that the method for pushing includes:
The data information of industrial accident case is analyzed, to obtain the attribute text data of industrial accident case
And numeric type data;
Calculate separately the first similarity between the attribute text data of every two case and the second phase between numeric type data
Like degree, and according to the overall similarity between the first similarity of every two case and the second similarity calculation every two case;
Multiple cases are clustered according to the overall similarity between the every two case, to generate multiple case clusters;
Target case is received, the case cluster where target case is obtained, and is selected in the case cluster where the target case
At least one case is pushed.
2. industrial accident case method for pushing according to claim 1, which is characterized in that the attribute of every two case
The calculating step of the first similarity between text data includes:
Case G is calculated separately based on WordNet semantic concept tree distanceiWith case GjThe similarity simA of each single item attribute textk
(Gi, Gj),
Wherein, depth (GI, k) it is case GiDepth of the K attribute texts in semantic concept tree, depth (GJ, k) it is case
Example GjDepth of the K attribute texts in semantic concept tree, depth (lso (GI, k, GJ, k) it is case GiK attributes
Text and case GjK attribute texts depth of the nearest common initial data in semantic concept tree, k 1,2 ...,
N, n are the item number of attribute text in attribute text data;
According to case GiWith case GjThe similarity simA of each single item attribute textk(Gi, Gj) and each single item attribute text total
Weight coefficient w in body attribute text data similaritykCalculate case GiWith case GjThe first similarity, whereinK is 1,2 ..., n, and n is attribute text in attribute text data
This item number.
3. industrial accident case method for pushing according to claim 1 or 2, which is characterized in that every two case
The calculating step of the second similarity between numeric type data includes:
Case G is calculated based on European space distanceiWith case GjNumeric type data the second similarity simC (Gi, Gj),
Wherein,For case GiY numeric type datas numerical value,For case GjY numeric type datas
Numerical value, w 'yFor y numeric type datas, weight coefficient, y 1,2 ..., m, m are numerical value in overall numerical value type data similarity
The item number of type data.
4. industrial accident case method for pushing according to claim 3, which is characterized in that described " according to every two
The step of overall similarity between first similarity of case and the second similarity calculation every two case " includes:
Overall similarity sim (Gi, Gj)=α × simA (Gi, Gj)+β×simC(Gi, Gj),
Wherein, α, β are respectively the first similarity and the second similarity shared weight in overall similarity.
5. industrial accident case method for pushing according to claim 1, which is characterized in that described " according to described every
Overall similarity between two cases clusters multiple cases, to generate multiple case clusters " the step of include:
S1 randomly chooses a untreated case from case total collection, search and the untreated case in case total collection
Overall similarity be less than first threshold all cases, when the quantity of the case searched be greater than or equal to second threshold when,
Case cluster is established by core of the untreated case, and the case searched is added in the candidate collection of the case cluster;When
It is noise by the untreated case marker when quantity of the case searched is less than second threshold;
To the untreated case of each of candidate collection the case cluster is added, and in case total collection in the case by S2
Search is less than all cases of first threshold with the overall similarity of the case, when the quantity of the case searched is greater than or equal to
When second threshold, the case searched is continuously added in candidate collection;
S3 repeats step S2, until untreated case is not present in candidate collection;
S4 repeats step S1 to S3, until untreated case is not present in case total collection.
6. industrial accident case method for pushing according to claim 1, which is characterized in that the method for pushing also wraps
It includes:
History industrial accident is obtained, the data information of the history industrial accident case is analyzed, with respectively
The attribute text data and numeric type data of history industrial accident case are obtained, to establish case library.
7. industrial accident case method for pushing according to claim 1, which is characterized in that the attribute text data
Including accident pattern, accident description, cause of accident, accident spot and situation of calling to account;The numeric type data include story time,
Causality loss and punishment situation.
8. a kind of industrial accident case supplying system, which is characterized in that the supplying system includes:
Analysis of cases module is analyzed for the data information to industrial accident case, to obtain industrial accident
The attribute text data and numeric type data of case;
Similarity calculation module, the first similarity between attribute text data, sum number for calculating separately every two case
The second similarity between value type data, according to the first similarity of every two case and the second similarity calculation every two case
Between overall similarity;
Cluster module, for being clustered according to the overall similarity between the every two case to multiple cases, to generate
Multiple case clusters;
Pushing module obtains the case cluster where target case, and select the target case place for receiving target case
Case cluster at least one case pushed.
9. industrial accident case supplying system according to claim 8, which is characterized in that the supplying system also wraps
It includes:
Case library, for storing the attribute text data and numeric type data of history industrial accident case.
10. industrial accident case supplying system according to claim 9, which is characterized in that the supplying system also wraps
It includes:
Interactive module, for receiving the target case query information of user's input.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811571338.7A CN109657060B (en) | 2018-12-21 | 2018-12-21 | Safety production accident case pushing method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811571338.7A CN109657060B (en) | 2018-12-21 | 2018-12-21 | Safety production accident case pushing method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109657060A true CN109657060A (en) | 2019-04-19 |
CN109657060B CN109657060B (en) | 2021-01-12 |
Family
ID=66116083
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811571338.7A Active CN109657060B (en) | 2018-12-21 | 2018-12-21 | Safety production accident case pushing method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109657060B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111309716A (en) * | 2020-01-22 | 2020-06-19 | 广东电网有限责任公司电力科学研究院 | Maintenance method and device applied to PAS case base and computer equipment |
CN113807940A (en) * | 2020-06-17 | 2021-12-17 | 马上消费金融股份有限公司 | Information processing and fraud identification method, device, equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101644572A (en) * | 2009-06-24 | 2010-02-10 | 中国科学院地理科学与资源研究所 | Detection method of ocean eddy variation based on historical similarity cases |
CN101930462A (en) * | 2010-08-20 | 2010-12-29 | 华中科技大学 | Comprehensive body similarity detection method |
CN104050196A (en) * | 2013-03-15 | 2014-09-17 | 阿里巴巴集团控股有限公司 | Point of interest (POI) data redundancy detection method and device |
CN104636430A (en) * | 2014-12-30 | 2015-05-20 | 东软集团股份有限公司 | Case knowledge base representation and case similarity obtaining method and system |
JP2017163421A (en) * | 2016-03-10 | 2017-09-14 | 富士通株式会社 | Moving image encoding method, computer program, and moving image encoding device |
-
2018
- 2018-12-21 CN CN201811571338.7A patent/CN109657060B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101644572A (en) * | 2009-06-24 | 2010-02-10 | 中国科学院地理科学与资源研究所 | Detection method of ocean eddy variation based on historical similarity cases |
CN101930462A (en) * | 2010-08-20 | 2010-12-29 | 华中科技大学 | Comprehensive body similarity detection method |
CN104050196A (en) * | 2013-03-15 | 2014-09-17 | 阿里巴巴集团控股有限公司 | Point of interest (POI) data redundancy detection method and device |
CN104636430A (en) * | 2014-12-30 | 2015-05-20 | 东软集团股份有限公司 | Case knowledge base representation and case similarity obtaining method and system |
JP2017163421A (en) * | 2016-03-10 | 2017-09-14 | 富士通株式会社 | Moving image encoding method, computer program, and moving image encoding device |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111309716A (en) * | 2020-01-22 | 2020-06-19 | 广东电网有限责任公司电力科学研究院 | Maintenance method and device applied to PAS case base and computer equipment |
CN113807940A (en) * | 2020-06-17 | 2021-12-17 | 马上消费金融股份有限公司 | Information processing and fraud identification method, device, equipment and storage medium |
CN113807940B (en) * | 2020-06-17 | 2024-04-12 | 马上消费金融股份有限公司 | Information processing and fraud recognition method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109657060B (en) | 2021-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10109275B2 (en) | Word hash language model | |
WO2020192401A1 (en) | System and method for generating answer based on clustering and sentence similarity | |
US10289717B2 (en) | Semantic search apparatus and method using mobile terminal | |
CN101055585B (en) | System and method for clustering documents | |
WO2018004829A1 (en) | Methods and apparatus for subgraph matching in big data analysis | |
CN100442284C (en) | Search system for providing information of keyword input frequency by category and method thereof | |
Aznag et al. | Leveraging formal concept analysis with topic correlation for service clustering and discovery | |
CN101563685A (en) | System and method for processing a query by using user feedback | |
US9009029B1 (en) | Semantic hashing in entity resolution | |
Usman et al. | Filter-based multi-objective feature selection using NSGA III and cuckoo optimization algorithm | |
CN105787097A (en) | Distributed index establishment method and system based on text clustering | |
US20070016600A1 (en) | System and method for index reorganization using partial index transfer in spatial data warehouse | |
Korobkin et al. | A multi-stage algorithm for text documents filtering based on physical knowledge | |
CN110580255A (en) | method and system for storing and retrieving data | |
CN109657060A (en) | safety production accident case pushing method and system | |
Le Nguyen et al. | A combined approach for filter feature selection in document classification | |
US6286012B1 (en) | Information filtering apparatus and information filtering method | |
CN103678513A (en) | Interactive search generation method and system | |
Basha et al. | An improved similarity matching based clustering framework for short and sentence level text | |
Vidhya et al. | Hybrid text mining model for document classification | |
Melethadathil et al. | Classification and clustering for neuroinformatics: Assessing the efficacy on reverse-mapped NeuroNLP data using standard ML techniques | |
Horban et al. | Principles of Searching for a Variety of Types of Associative Rules in OLAP-cubes. | |
KR102268570B1 (en) | Apparatus and method for generating document cluster | |
EP4002151A1 (en) | Data tagging and synchronisation system | |
Bedo et al. | A k–skyband approach for feature selection |
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 |