CN108153843A - A kind of medical information personalized recommendation method and system based on cloud platform - Google Patents
A kind of medical information personalized recommendation method and system based on cloud platform Download PDFInfo
- Publication number
- CN108153843A CN108153843A CN201711367661.8A CN201711367661A CN108153843A CN 108153843 A CN108153843 A CN 108153843A CN 201711367661 A CN201711367661 A CN 201711367661A CN 108153843 A CN108153843 A CN 108153843A
- Authority
- CN
- China
- Prior art keywords
- keyword
- cloud platform
- cluster
- medical
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5017—Task decomposition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of medical information personalized recommendation method and system based on cloud platform, wherein, the medical information personalized recommendation method includes:The medical information of user is collected based on user terminal application layer, obtains the first keyword of medical information;First keyword is sent to cloud platform by user terminal application layer based on security protocol, and retrieval type is generated based on the first keyword;It is retrieved based on retrieval type generation retrieval tasks, keyword extraction is carried out to the medical literature retrieved, obtains the second keyword;Clustering processing is carried out to the second keyword, degree of membership calculating is carried out according to cluster result, build target keywords neighbor set, according to the medical profession information that target keywords neighbor set is accustomed to obtaining and user matches with target keywords based on user, and medical profession information is pushed into user terminal layer.In embodiments of the present invention, it is inaccurate to solve existing medical profession information push, improves the experience property of entire platform.
Description
Technical field
The present invention relates to digital medical technical field more particularly to a kind of medical information personalized recommendations based on cloud platform
Method and system.
Background technology
China's heart brain blood disease and chronic are numerous at present, and there is cardiovascular patient about 300,000,000, at least 5.8 hundred million in the whole nation
People has at least one or more risk factor related with slow disease, and to the year two thousand thirty, Chinese slow disease burden will increase by 50%.
2016, the joint publication of seven ministries and commissions of State Council《About the notice for printing and distributing propulsion family doctor's subscribed services instruction》(traditional Chinese medical science
Correct hair (2016) 1), it is desirable that by 2017, family doctor's subscribed services coverage rate reached more than 30%, key population signing
Service coverage rate reaches more than 60%, and key population mainly includes the patients with chronic diseases such as hypertension, diabetes, tuberculosis.It arrives
The year two thousand twenty strives subscribed services being expanded to full crowd, forms contractual service relationship steady in a long-term, realize family doctor substantially
The all standing of subscribed services system.Authoritative survey data shows, have in the crowd seen a doctor in large hospital 70% patient and be not required to
Want on-scene care, it is only necessary to online or mobile terminal carries out interrogation service, can solve the demand of this part population significantly,
Mitigate the work load of doctor, improve medical service level and efficiency.
However, in online interrogation, after uploading some medical informations about illness, the medical treatment of platform feedback or push
Information suggests that not high quick and accurate or push sequence and method do not meet the browsing custom of user.
Invention content
It is an object of the invention to overcome the deficiencies in the prior art, and the present invention provides a kind of medical treatment letters based on cloud platform
Personalized recommendation method and system are ceased, existing medical profession information push inaccuracy is solved, improves the experience of entire platform
Property.
In order to solve the above-mentioned technical problem, an embodiment of the present invention provides a kind of medical information based on cloud platform is personalized
Recommendation method, the medical information personalized recommendation method include:
The medical information of user is collected based on user terminal application layer, keyword extraction is carried out based on the medical information,
Obtain the first keyword of the medical information;
First keyword is sent to cloud platform by the user terminal application layer based on security protocol, based on described
One keyword generates retrieval type;
Retrieval tasks are generated based on the retrieval type, according to the type of database, the retrieval tasks are converted into described
Multiple retrieval streams of cloud platform will be deployed on each retrieval unit in cloud platform server cluster a retrieval stream respectively;
The medical literature retrieved is pooled to cloud platform server by each retrieval unit, using TF-IDF algorithms to collecting
Medical literature on to cloud platform server carries out keyword extraction and calculates to form keyword calculating task, and calculating task is switched to
It is multiple to calculate stream, multiple calculating streams are evenly distributed on each computing unit in server cluster, it is crucial to obtain second
Word;
Cluster calculation is carried out in cloud platform according to second keyword and forms cluster task, cluster task is divided into more
A clustering flow, and multiple clustering flows are evenly distributed on each cluster cell in server cluster, obtain described second
The degree of membership of the cluster centre of keyword and each second keyword in cluster;
Cloud platform server according to the degree of membership calculate between target keywords and each second keyword
Similarity obtains the second keyword composition target keywords neighbor set higher with target keywords similarity;
Cloud platform is based on user according to the target keywords neighbor set and is accustomed to obtaining and user and target keywords phase
The medical profession information matched, and medical profession information is pushed into user terminal layer.
Preferably, the target keywords are the second keyword of cluster centre.
Preferably, the cluster centre for obtaining second keyword and each second keyword being subordinate in cluster
Degree, including:
Determine the second keyword number to be clustered;
Fuzzy clustering is carried out according to the second keyword number, determines the cluster centre of second keyword, is obtained
Second keyword of the cluster centre periphery aggregation;
Degree of membership processing is carried out to the second keyword that the cluster centre periphery is assembled using Subject Matrix, is obtained each
Degree of membership of second keyword in cluster.
Preferably, it is described according to similar between degree of membership calculating target keywords and each second keyword
Degree, including:
The similarity between target keywords and each second keyword is calculated, obtains result of calculation;
Sequencing of similarity is carried out to the second keyword according to the result of calculation, obtains ranking results;
It is adjacent to choose the second keyword composition target keywords higher with the target keywords similarity in ranking results
Nearly collection.
Preferably, the cloud platform is based on the acquisition of user's custom and user and target according to the target keywords neighbor set
The medical profession information that keyword matches, and medical profession information is pushed into user terminal layer, including:
Cloud platform obtains the user information browsing custom of user terminal;
Integration processing is carried out to the medical profession information to match according to user information browsing custom;
It will integrate that treated that medical profession information pushes to user terminal layer.
In addition, the embodiment of the present invention additionally provides a kind of medical information personalized recommendation system based on cloud platform, it is described
Medical information personalized recommendation system includes user terminal application layer and cloud platform;
The user terminal application layer:For collecting the medical information of user based on user terminal application layer, based on described
Medical information carries out keyword extraction, obtains the first keyword of the medical information;The user terminal application layer is based on peace
First keyword is sent to cloud platform by full agreement, and retrieval type is generated based on first keyword;
The cloud platform includes:
Retrieve module:For being based on the retrieval type generation retrieval tasks, according to the type of database, the retrieval is appointed
Business is converted into multiple retrieval streams of the cloud platform, each in cloud platform server cluster by being deployed in respectively to a retrieval stream
On retrieval unit;
Second keyword-extraction module:The medical literature retrieved is pooled to cloud platform service for each retrieval unit
Device carries out keyword extraction to the medical literature being pooled on cloud platform server using TF-IDF algorithms and calculates to form keyword
Calculating task is switched to multiple calculate and flowed, multiple calculating streams are evenly distributed to each in server cluster by calculating task
On computing unit, the second keyword is obtained;
Cluster module:Cluster task is formed for carrying out cluster calculation in cloud platform according to second keyword, it will
Cluster task is divided into multiple clustering flows, and multiple clustering flows are evenly distributed to each cluster cell in server cluster
On, obtain the degree of membership of the cluster centre and each second keyword of second keyword in cluster;
Similarity calculation module:For cloud platform server according to institute's degree of membership calculate target keywords with it is described each
Similarity between a second keyword obtains the second keyword composition target keywords higher with target keywords similarity
Neighbor set;
Pushing module:User is based on for cloud platform according to the target keywords neighbor set to be accustomed to obtaining and user and mesh
The medical profession information that mark keyword matches, and medical profession information is pushed into user terminal layer.
Preferably, the target keywords are the second keyword of cluster centre.
Preferably, the cluster module includes:
Determination unit:For determining the second keyword number to be clustered;
Fuzzy clustering unit:For carrying out fuzzy clustering according to the second keyword number, determine that described second is crucial
The cluster centre of word obtains the second keyword of the cluster centre periphery aggregation;
Degree of membership acquiring unit:For being carried out using Subject Matrix to the second keyword that the cluster centre periphery is assembled
Degree of membership processing obtains degree of membership of each second keyword in cluster.
Preferably, the similarity calculation module includes:
Computing unit:For calculating the similarity between target keywords and each second keyword, obtain and calculate
As a result;
Sequencing unit:For carrying out sequencing of similarity to the second keyword according to the result of calculation, ranking results are obtained;
Selection unit:It is formed for choosing in ranking results with the second higher keyword of the target keywords similarity
Target keywords neighbor set.
Preferably, the pushing module includes:
Browsing custom acquiring unit:The user information that user terminal is obtained for cloud platform browses custom;
Integral unit:For being carried out at integration to the medical profession information to match according to user information browsing custom
Reason;
Push unit:For that will integrate that treated, medical profession information will push to user terminal layer.
In embodiments of the present invention, keyword extraction is carried out to the medical information that user uploads first, according to these keys
Word structure retrieval type is retrieved, and obtains searching document, carries out keyword extraction to these searching documents, it is crucial to obtain second
Word, carries out the second keyword clustering processing and degree of membership calculates, and obtains target keywords neighbor set;It is real in these processing
Now shunting is handled, and is improved the ability of processing, is ensured the quick operation of platform;Using the above method, user can be needed
Medical profession information quickly pushes to user terminal, improves the Experience Degree of user.
Description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it is clear that, the accompanying drawings in the following description is only this
Some embodiments of invention, for those of ordinary skill in the art, without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the method flow signal of the medical information personalized recommendation method based on cloud platform in the embodiment of the present invention
Figure;
Fig. 2 is the system structure composition of the medical information personalized recommendation system based on cloud platform in the embodiment of the present invention
Schematic diagram.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained all other without creative efforts
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 is the method flow signal of the medical information personalized recommendation method based on cloud platform in the embodiment of the present invention
Figure, as shown in Figure 1, the medical information personalized recommendation method includes:
S11:The medical information of user is collected based on user terminal application layer, carrying out keyword based on the medical information carries
It takes, obtains the first keyword of the medical information;
Specifically, user terminal application layer can be connect with the user terminal for the medical information signal that can receive user,
For receiving user's medical information of user terminal uploads, in user terminal application layer, these medical informations are carried out crucial
Word extracts, and obtains the first keyword.
S12:First keyword is sent to cloud platform by the user terminal application layer based on security protocol, based on institute
State the first keyword generation retrieval type;
Specifically, the security protocol is tls protocol, ssl protocol etc.;It, can after the first keyword is received in cloud platform
According to the database retrieval format needs in cloud platform, the retrieval type of the first keyword of corresponding generation.
S13:Retrieval tasks are generated based on the retrieval type, according to the type of database, the retrieval tasks are converted into
Multiple retrieval streams of the cloud platform, each retrieval unit that a retrieval stream will be deployed in respectively in cloud platform server cluster
On;
Specifically, after retrieval type is generated, cloud platform can generate corresponding retrieval tasks according to retrieval type, according to database
Type retrieval type is converted into multiple retrieval streams in cloud platform, the speed of retrieval can be accelerated in this way;By multiple retrieval streams
It is deployed in respectively on each retrieval unit in cloud platform server cluster, realizes the quick-searching to database data.
S14:The medical literature retrieved is pooled to cloud platform server by each retrieval unit, using TF-IDF algorithms pair
The medical literature being pooled on cloud platform server carries out keyword extraction and calculates to form keyword calculating task, by calculating task
Switch to multiple calculate to flow, multiple calculating streams are evenly distributed on each computing unit in server cluster, obtain second
Keyword;
Specifically, in above-mentioned steps, the medical literature that each retrieval unit retrieves is pooled to cloud platform clothes
It is engaged on device, then carrying out keyword extraction to the medical literature being pooled on cloud platform server using TF-IDF algorithms calculates shape
Into keyword calculating task, and calculating task is converted into multiple calculate and is flowed, multiple calculating streams are evenly distributed to server set
On each computing unit in group, the second keyword is obtained.
Wherein, the main thought of TF-IDF is:If the frequency TF high that some word or phrase occur in an article, and
And seldom occur in other articles, then it is assumed that this word or phrase have good class discrimination ability, are adapted to classify.
TF-IDF is actually:TF*IDF, TF word frequency (Term Frequency), the reverse document-frequencies of IDF (Inverse Document
Frequency).TF represents the frequency that entry occurs in document d.The main thought of IDF is:If include the document of entry t
It is fewer, that is, n is smaller, IDF is bigger, then illustrates that entry t has good class discrimination ability.If in certain a kind of document C
Number of files comprising entry t is m, and the total number of documents that other classes include t is k, it is clear that all number of files n=m+k comprising t,
When m is big, n is also big, can be small according to the value of IDF that IDF formula obtain, and just illustrates that entry t class discriminations are indifferent.
If but in fact, an entry frequently occurs in the document of a class, illustrate that the entry can represent this very well
The feature of the text of class, such entry should assign higher weight to them, and select the Feature Words for being used as the class text
With difference and other class documents.
S15:Cluster calculation is carried out in cloud platform according to second keyword and forms cluster task, by cluster task point
For multiple clustering flows, and multiple clustering flows are evenly distributed on each cluster cell in server cluster, described in acquisition
The degree of membership of the cluster centre of second keyword and each second keyword in cluster;
Specifically, carrying out clustering processing to the second keyword using fuzzy clustering method in cloud platform, generation cluster is appointed
It is engaged in and is converted into multiple clustering flows, and multiple clustering flows are evenly distributed on each cluster cell in server cluster,
Obtain the degree of membership of the cluster centre and each second keyword of second keyword in cluster.
In the above process is performed, the second keyword number to be clustered is determined;According to the second keyword number into
Row fuzzy clustering determines the cluster centre of second keyword, obtains the second keyword of the cluster centre periphery aggregation;
Degree of membership processing is carried out to the second keyword that the cluster centre periphery is assembled using Subject Matrix, it is crucial to obtain each second
Degree of membership of the word in cluster.
Further, when initial clustering number is determined reference resources quantity and user demand come select one it is appropriate
Clusters number, then by analyze cluster result adjust clusters number, optimize Clustering Effect;Carry out fuzzy clustering when
It waits, may be used and share the mode of cluster and clustered, it is high, mutual similar according to the frequency of occurrences after cluster is completed
The second relatively low keyword of property is as cluster centre, and centered on the cluster centre, the situation of the second keyword of surrounding;Just
Beginningization the second keyword Subject Matrix carries out the second keyword Subject Matrix U=of initialization using the random number between 0 to 1
{ul,u2,…,un, wherein uj=(u1j,u2j..., ucj) T, uijRepresent certain degrees of membership of the second keyword j in the i-th class, uij
Between 0 to 1 and it is made to meet the following formula:
Cluster centre Ci(i=1,2 ..., c) be by the use of the second keyword be subordinate to the degree of membership of each classification as weighting because
Son and Weighted Index calculate the weighted average of all second keyword feature vectors, and calculation formula is specially:
Wherein, uijRepresent degrees of membership of the second keyword j in the i-th class;xjVector space mould for the second keyword j
Type;N is the second overall number of keywords mesh;m(m>1) it is Weighted Index, usually takes 1 can be adjusted according to actual conditions;For
User j is under the jurisdiction of the weighted sum of each classification.
S16:Cloud platform server according to institute's degree of membership calculate target keywords and each second keyword it
Between similarity, obtain with target keywords similarity it is higher the second keyword composition target keywords neighbor set;
Specifically, calculating the similarity between target keywords and each second keyword, result of calculation is obtained;Root
Sequencing of similarity is carried out to the second keyword according to the result of calculation, obtains ranking results;Choose ranking results in the mesh
Mark higher the second keyword composition target keywords neighbor set of keyword similarity.
Further, it is obtained by calculating all second keyword feature vector data points to cluster apart from summation,
In a second keyword feature vector data point to cluster distance be to be subordinate to the person in servitude of each classification using second keyword
The weighted value that category degree calculates the second number of keyword strong point to each cluster centre distance as weighted factor and Weighted Index is total
With obtain, cost function J (U, c1,…,cc) calculation formula be specially:
JiObject function for the i-th class cluster;c1,…,ccIt is input parameter for cluster centre;C is the number of cluster;n
Number for the second keyword;M is Weighted Index;U be the second keyword Subject Matrix, matrix element uijRepresent user j i-th
Degree of membership in a classification;dijFor the euclidean between ith cluster center and j-th of second keyword feature vector data points
Distance namely dijCalculation formula be specially:dij=| | ci-xj||。
It is integrated according to the weighted value of above-mentioned centre distance of wrestling and carries out similarity calculation, so as to get target keywords
With the similarity between each second keyword, these similarities are carried out with a sequence processing, is chosen according to ranking results
The neighbor set of the second keyword composition target keywords of the more forward ranking results of sequence.
S17:Cloud platform is based on user according to the target keywords neighbor set and is accustomed to obtaining and user and target keywords
The medical profession information to match, and medical profession information is pushed into user terminal layer.
Specifically, cloud platform obtains the user information browsing custom of user terminal;It is browsed and is accustomed to according to the user information
Integration processing is carried out to the medical profession information to match;It will integrate that treated that medical profession information pushes to user terminal
Layer.
Further, the browsing custom of medical information believes the medical profession to match on the subscriber terminal according to user
Breath carries out integration processing, obtains to browse with user and is accustomed to relevant information, and push and cure to user terminal layer according to these information
Business information is treated, user is facilitated to browse and is used.
Fig. 2 is the system structure composition of the medical information personalized recommendation system based on cloud platform in the embodiment of the present invention
Schematic diagram, as shown in Fig. 2, the medical information personalized recommendation system includes user terminal application layer 11 and cloud platform 12;
The user terminal application layer 11:For collecting the medical information of user based on user terminal application layer, based on institute
It states medical information and carries out keyword extraction, obtain the first keyword of the medical information;The user terminal application layer is based on
First keyword is sent to cloud platform by security protocol, and retrieval type is generated based on first keyword;
The cloud platform 12 includes:
Retrieve module 121:For being based on the retrieval type generation retrieval tasks, according to the type of database, by the inspection
Rope task is converted into multiple retrieval streams of the cloud platform, and a retrieval stream will be deployed in respectively in cloud platform server cluster
On each retrieval unit;
Second keyword-extraction module 122:The medical literature retrieved is pooled to cloud platform for each retrieval unit
Server carries out keyword extraction to the medical literature being pooled on cloud platform server using TF-IDF algorithms and calculates to form pass
Calculating task is switched to multiple calculate and flowed by key word calculating task, multiple calculating streams is evenly distributed to every in server cluster
On one computing unit, the second keyword is obtained;
Cluster module 123:Cluster task is formed for carrying out cluster calculation in cloud platform according to second keyword,
Cluster task is divided into multiple clustering flows, and multiple clustering flows are evenly distributed to each cluster cell in server cluster
On, obtain the degree of membership of the cluster centre and each second keyword of second keyword in cluster;
Similarity calculation module 124:Target keywords and institute are calculated according to institute's degree of membership for cloud platform server
The similarity between each second keyword is stated, the second keyword composition target higher with target keywords similarity is obtained and closes
Key word neighbor set;
Pushing module 125:User is based on according to the target keywords neighbor set for cloud platform and is accustomed to acquisition and user
The medical profession information to match with target keywords, and medical profession information is pushed into user terminal layer.
Preferably, the target keywords are the second keyword of cluster centre.
Preferably, the cluster module 123 includes:
Determination unit:For determining the second keyword number to be clustered;
Fuzzy clustering unit:For carrying out fuzzy clustering according to the second keyword number, determine that described second is crucial
The cluster centre of word obtains the second keyword of the cluster centre periphery aggregation;
Degree of membership acquiring unit:For being carried out using Subject Matrix to the second keyword that the cluster centre periphery is assembled
Degree of membership processing obtains degree of membership of each second keyword in cluster.
Preferably, the similarity calculation module 124 includes:
Computing unit:For calculating the similarity between target keywords and each second keyword, obtain and calculate
As a result;
Sequencing unit:For carrying out sequencing of similarity to the second keyword according to the result of calculation, ranking results are obtained;
Selection unit:It is formed for choosing in ranking results with the second higher keyword of the target keywords similarity
Target keywords neighbor set.
Preferably, the pushing module 125 includes:
Browsing custom acquiring unit:The user information that user terminal is obtained for cloud platform browses custom;
Integral unit:For being carried out at integration to the medical profession information to match according to user information browsing custom
Reason;
Push unit:For that will integrate that treated, medical profession information will push to user terminal layer.
Specifically, the operation principle of the system related functions module of the embodiment of the present invention can be found in the correlation of embodiment of the method
Description, which is not described herein again.
In embodiments of the present invention, keyword extraction is carried out to the medical information that user uploads first, according to these keys
Word structure retrieval type is retrieved, and obtains searching document, carries out keyword extraction to these searching documents, it is crucial to obtain second
Word, carries out the second keyword clustering processing and degree of membership calculates, and obtains target keywords neighbor set;It is real in these processing
Now shunting is handled, and is improved the ability of processing, is ensured the quick operation of platform;Using the above method, user can be needed
Medical profession information quickly pushes to user terminal, improves the Experience Degree of user.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium can include:Read-only memory (ROM, Read Only Memory), random access memory (RAM, Random
Access Memory), disk or CD etc..
In addition, a kind of medical information personalized recommendation method based on cloud platform provided above the embodiment of the present invention
And system is described in detail, and should employ specific case herein and the principle of the present invention and embodiment are explained
It states, the explanation of above example is only intended to facilitate the understanding of the method and its core concept of the invention;Meanwhile for this field
Those skilled in the art, thought according to the present invention, in specific embodiments and applications there will be changes, to sum up institute
It states, the content of the present specification should not be construed as limiting the invention.
Claims (10)
1. a kind of medical information personalized recommendation method based on cloud platform, which is characterized in that the medical information personalization pushes away
The method of recommending includes:
The medical information of user is collected based on user terminal application layer, keyword extraction is carried out based on the medical information, is obtained
First keyword of the medical information;
First keyword is sent to cloud platform by the user terminal application layer based on security protocol, is closed based on described first
Key word generates retrieval type;
Retrieval tasks are generated based on the retrieval type, according to the type of database, the retrieval tasks is converted into the cloud and are put down
Multiple retrieval streams of platform will be deployed on each retrieval unit in cloud platform server cluster a retrieval stream respectively;
The medical literature retrieved is pooled to cloud platform server by each retrieval unit, using TF-IDF algorithms to being pooled to cloud
Medical literature on Platform Server carries out keyword extraction and calculates to form keyword calculating task, calculating task is switched to multiple
Stream is calculated, multiple calculating streams are evenly distributed on each computing unit in server cluster, obtain the second keyword;
Cluster calculation is carried out in cloud platform according to second keyword and forms cluster task, cluster task is divided into multiple poly-
Class stream, and multiple clustering flows are evenly distributed on each cluster cell in server cluster, it is crucial to obtain described second
The degree of membership of the cluster centre of word and each second keyword in cluster;
Cloud platform server according to the degree of membership calculate it is similar between target keywords and each second keyword
Degree obtains the second keyword composition target keywords neighbor set higher with target keywords similarity;
Cloud platform is accustomed to obtaining what is with user with target keywords matched based on user according to the target keywords neighbor set
Medical profession information, and medical profession information is pushed into user terminal layer.
2. the medical information personalized recommendation method according to claim 1 based on cloud platform, which is characterized in that the mesh
Mark the second keyword that keyword is cluster centre.
3. the medical information personalized recommendation method according to claim 1 based on cloud platform, which is characterized in that described to obtain
The degree of membership of the cluster centre and each second keyword of second keyword in cluster is taken, including:
Determine the second keyword number to be clustered;
Fuzzy clustering is carried out according to the second keyword number, determines the cluster centre of second keyword, described in acquisition
Second keyword of cluster centre periphery aggregation;
Degree of membership processing is carried out to the second keyword that the cluster centre periphery is assembled using Subject Matrix, obtains each second
Degree of membership of the keyword in cluster.
4. the medical information personalized recommendation method according to claim 1 based on cloud platform, which is characterized in that described
The similarity between target keywords and each second keyword is calculated according to the degree of membership, including:
The similarity between target keywords and each second keyword is calculated, obtains result of calculation;
Sequencing of similarity is carried out to the second keyword according to the result of calculation, obtains ranking results;
It chooses in ranking results and forms target keywords neighbor set with the second higher keyword of the target keywords similarity.
5. the medical information personalized recommendation method according to claim 1 based on cloud platform, which is characterized in that the cloud
Platform is accustomed to obtaining the hospitality industry with target keywords to match with user based on user according to the target keywords neighbor set
Business information, and medical profession information is pushed into user terminal layer, including:
Cloud platform obtains the user information browsing custom of user terminal;
Integration processing is carried out to the medical profession information to match according to user information browsing custom;
It will integrate that treated that medical profession information pushes to user terminal layer.
6. a kind of medical information personalized recommendation system based on cloud platform, which is characterized in that the medical information personalization pushes away
It recommends system and includes user terminal application layer and cloud platform;
The user terminal application layer:For collecting the medical information of user based on user terminal application layer, based on the medical treatment
Information carries out keyword extraction, obtains the first keyword of the medical information;The user terminal application layer is based on safety and assists
First keyword is sent to cloud platform by view, and retrieval type is generated based on first keyword;
The cloud platform includes:
Retrieve module:For being based on the retrieval type generation retrieval tasks, according to the type of database, the retrieval tasks are turned
Multiple retrieval streams of the cloud platform are turned to, each retrieval being respectively deployed in a retrieval stream in cloud platform server cluster
On unit;
Second keyword-extraction module:The medical literature retrieved is pooled to cloud platform server for each retrieval unit,
Keyword extraction is carried out using TF-IDF algorithms to the medical literature being pooled on cloud platform server to calculate to form keyword meter
Calculating task is switched to multiple calculate and flowed by calculation task, each meter multiple calculating streams being evenly distributed in server cluster
It calculates on unit, obtains the second keyword;
Cluster module:Cluster task is formed for carrying out cluster calculation in cloud platform according to second keyword, will be clustered
Task is divided into multiple clustering flows, and multiple clustering flows are evenly distributed on each cluster cell in server cluster, obtains
Take the degree of membership of the cluster centre and each second keyword of second keyword in cluster;
Similarity calculation module:Target keywords and described each the are calculated according to institute's degree of membership for cloud platform server
It is neighbouring to obtain the second keyword composition target keywords higher with target keywords similarity for similarity between two keywords
Collection;
Pushing module:The acquisition of user's custom is based on for cloud platform according to the target keywords neighbor set to close with user and target
The medical profession information that key word matches, and medical profession information is pushed into user terminal layer.
7. the medical information personalized recommendation system according to claim 6 based on cloud platform, which is characterized in that the mesh
Mark the second keyword that keyword is cluster centre.
8. the medical information personalized recommendation system according to claim 6 based on cloud platform, which is characterized in that described poly-
Generic module includes:
Determination unit:For determining the second keyword number to be clustered;
Fuzzy clustering unit:For carrying out fuzzy clustering according to the second keyword number, second keyword is determined
Cluster centre obtains the second keyword of the cluster centre periphery aggregation;
Degree of membership acquiring unit:For being subordinate to using the second keyword that Subject Matrix assembles the cluster centre periphery
Degree processing obtains degree of membership of each second keyword in cluster.
9. the medical information personalized recommendation system according to claim 6 based on cloud platform, which is characterized in that the phase
Include like degree computing module:
Computing unit:For calculating the similarity between target keywords and each second keyword, result of calculation is obtained;
Sequencing unit:For carrying out sequencing of similarity to the second keyword according to the result of calculation, ranking results are obtained;
Selection unit:For choosing in ranking results target is formed with the second higher keyword of the target keywords similarity
Keyword neighbor set.
10. the medical information personalized recommendation system according to claim 6 based on cloud platform, which is characterized in that described
Pushing module includes:
Browsing custom acquiring unit:The user information that user terminal is obtained for cloud platform browses custom;
Integral unit:For carrying out integration processing to the medical profession information to match according to user information browsing custom;
Push unit:For that will integrate that treated, medical profession information will push to user terminal layer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711367661.8A CN108153843A (en) | 2017-12-18 | 2017-12-18 | A kind of medical information personalized recommendation method and system based on cloud platform |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711367661.8A CN108153843A (en) | 2017-12-18 | 2017-12-18 | A kind of medical information personalized recommendation method and system based on cloud platform |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108153843A true CN108153843A (en) | 2018-06-12 |
Family
ID=62467575
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711367661.8A Pending CN108153843A (en) | 2017-12-18 | 2017-12-18 | A kind of medical information personalized recommendation method and system based on cloud platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108153843A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109065127A (en) * | 2018-08-22 | 2018-12-21 | 重庆华医康道科技有限公司 | Medical treatment & health cloud intelligent information pushes working platform method |
CN109559832A (en) * | 2018-10-23 | 2019-04-02 | 平安医疗健康管理股份有限公司 | A kind of information processing method, device, server and medium |
CN111341457A (en) * | 2020-02-25 | 2020-06-26 | 广州七乐康药业连锁有限公司 | Medical diagnosis information visualization method and device based on big data retrieval |
CN111540421A (en) * | 2020-04-19 | 2020-08-14 | 周玉娟 | Big data based intelligent medical condition monitoring method and big data medical cloud platform |
CN113536110A (en) * | 2021-06-09 | 2021-10-22 | 四川数字链享科技有限公司 | Medical information pushing system based on big data |
CN111341457B (en) * | 2020-02-25 | 2024-05-10 | 广州七乐康药业连锁有限公司 | Medical diagnosis information visualization method and device based on big data retrieval |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2016085717A (en) * | 2014-10-22 | 2016-05-19 | 財団法人 資訊工業策進会Institute For Information Industry | Service request analysis system, method and computer readable recording medium |
CN105760547A (en) * | 2016-03-16 | 2016-07-13 | 中山大学 | Book recommendation method and system based on user clustering |
CN106960215A (en) * | 2017-02-21 | 2017-07-18 | 广州七乐康药业连锁有限公司 | A kind of method and system that the matching of medical image depth is realized based on cloud platform |
CN107016231A (en) * | 2017-02-21 | 2017-08-04 | 广州七乐康药业连锁有限公司 | It is a kind of that the method and system that medical data is calculated are realized based on cloud platform |
-
2017
- 2017-12-18 CN CN201711367661.8A patent/CN108153843A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2016085717A (en) * | 2014-10-22 | 2016-05-19 | 財団法人 資訊工業策進会Institute For Information Industry | Service request analysis system, method and computer readable recording medium |
CN105760547A (en) * | 2016-03-16 | 2016-07-13 | 中山大学 | Book recommendation method and system based on user clustering |
CN106960215A (en) * | 2017-02-21 | 2017-07-18 | 广州七乐康药业连锁有限公司 | A kind of method and system that the matching of medical image depth is realized based on cloud platform |
CN107016231A (en) * | 2017-02-21 | 2017-08-04 | 广州七乐康药业连锁有限公司 | It is a kind of that the method and system that medical data is calculated are realized based on cloud platform |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109065127A (en) * | 2018-08-22 | 2018-12-21 | 重庆华医康道科技有限公司 | Medical treatment & health cloud intelligent information pushes working platform method |
CN109065127B (en) * | 2018-08-22 | 2021-11-23 | 重庆华医康道科技有限公司 | Working method of medical health cloud intelligent information pushing platform |
CN109559832A (en) * | 2018-10-23 | 2019-04-02 | 平安医疗健康管理股份有限公司 | A kind of information processing method, device, server and medium |
CN111341457A (en) * | 2020-02-25 | 2020-06-26 | 广州七乐康药业连锁有限公司 | Medical diagnosis information visualization method and device based on big data retrieval |
CN111341457B (en) * | 2020-02-25 | 2024-05-10 | 广州七乐康药业连锁有限公司 | Medical diagnosis information visualization method and device based on big data retrieval |
CN111540421A (en) * | 2020-04-19 | 2020-08-14 | 周玉娟 | Big data based intelligent medical condition monitoring method and big data medical cloud platform |
CN111540421B (en) * | 2020-04-19 | 2021-01-08 | 深圳坐标软件集团有限公司 | Big data based intelligent medical condition monitoring method and big data medical cloud platform |
CN113536110A (en) * | 2021-06-09 | 2021-10-22 | 四川数字链享科技有限公司 | Medical information pushing system based on big data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Akbari et al. | From tweets to wellness: Wellness event detection from twitter streams | |
CN108153843A (en) | A kind of medical information personalized recommendation method and system based on cloud platform | |
CN111883228B (en) | Knowledge graph-based health information recommendation method, device, equipment and medium | |
Zhang et al. | Personalized recommendation of social images by constructing a user interest tree with deep features and tag trees | |
Farseev et al. | Tweetfit: Fusing multiple social media and sensor data for wellness profile learning | |
WO2014186387A1 (en) | Context-aware prediction in medical systems | |
Wu et al. | Flickr distance: a relationship measure for visual concepts | |
CN106528676B (en) | Entity Semantics search processing method and device based on artificial intelligence | |
Komkhao et al. | Incremental collaborative filtering based on Mahalanobis distance and fuzzy membership for recommender systems | |
Nawar et al. | Cross-content recommendation between movie and book using machine learning | |
Farseev et al. | Tweet can be fit: Integrating data from wearable sensors and multiple social networks for wellness profile learning | |
Yigit et al. | Extended topology based recommendation system for unidirectional social networks | |
Zhu et al. | Inferring personality traits from attentive regions of user liked images via weakly supervised dual convolutional network | |
Wang et al. | Analysis of regional group health persona based on image recognition | |
Kacchi et al. | Friend recommendation system based on lifestyles of users | |
Huang et al. | Social friend recommendation based on network correlation and feature co-clustering | |
Pendyala et al. | Towards a truthful world wide web from a humanitarian perspective | |
Li et al. | Boosting Healthiness Exposure in Category-constrained Meal Recommendation Using Nutritional Standards | |
Bu et al. | Unsupervised face-name association via commute distance | |
Farhana et al. | Predict Diabetes Mellitus Using Machine Learning Algorithms | |
Hussein et al. | Smart collaboration framework for managing chronic disease using recommender system | |
Vijayarajeswari et al. | An enhanced symptom clustering with profile based prescription suggestion in biomedical application | |
CN108172304A (en) | A kind of medical information visible processing method and system based on user's medical treatment feedback | |
Liu et al. | Recommendations for different tasks based on the uniform multimodal joint representation | |
Abu et al. | Biodiversity image retrieval framework for monogeneans |
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 |
Application publication date: 20180612 |
|
RJ01 | Rejection of invention patent application after publication |