CN109376901A - A kind of service quality prediction technique based on decentralization matrix decomposition - Google Patents
A kind of service quality prediction technique based on decentralization matrix decomposition Download PDFInfo
- Publication number
- CN109376901A CN109376901A CN201811043938.6A CN201811043938A CN109376901A CN 109376901 A CN109376901 A CN 109376901A CN 201811043938 A CN201811043938 A CN 201811043938A CN 109376901 A CN109376901 A CN 109376901A
- Authority
- CN
- China
- Prior art keywords
- user
- service
- indicate
- potential
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 239000011159 matrix material Substances 0.000 title claims abstract description 38
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 27
- 230000003993 interaction Effects 0.000 claims abstract description 21
- 230000006870 function Effects 0.000 claims description 7
- 230000004044 response Effects 0.000 claims description 7
- 230000002452 interceptive effect Effects 0.000 claims description 6
- 238000005295 random walk Methods 0.000 claims description 6
- 238000003786 synthesis reaction Methods 0.000 claims description 5
- 230000006399 behavior Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 abstract description 18
- 239000002699 waste material Substances 0.000 abstract description 9
- 238000002474 experimental method Methods 0.000 description 3
- 101100379079 Emericella variicolor andA gene Proteins 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012856 packing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 210000003739 neck Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of service quality prediction techniques based on decentralization matrix decomposition, comprising the following steps: (1) establishes user's adjacent map, (2) determine the range of interaction, and (3) determine the information of interaction, and (4) synthesize and predict qos value.Through the above way; the present invention is based on the service quality prediction techniques of decentralization matrix decomposition to predict qos value by using decentralization matrix disassembling method; solve the problems, such as computing resource waste caused by centralized training; furthermore; each user is stored in user oneself hand the qos value of Web service; to protect the individual privacy of user well, there are extensive market prospects in popularizing for the service quality prediction technique based on decentralization matrix decomposition.
Description
Technical field
The present invention relates to network service fields, pre- more particularly to a kind of service quality based on decentralization matrix decomposition
Survey method.
Background technique
With the fast development of Internet technology, more and more Web services are emerged, Internet user wants in magnanimity
Web service in search out Web service required for oneself more difficult.Therefore, more and more Web service recommendation systems are pregnant
It educates and gives birth to, and have received widespread attention and study.
In order to realize that personalized Web service recommendation, Web service recommendation system need to collect user to service quality QoS
The assessment situation of (Quality of Service).QoS is widely used in the non-functional feature of description Web service, such as: when response
Between, handling capacity, price, reliability etc..Therefore, the service recommendation system based on QoS has obtained in society extensively
Application.It is existing always that a kind of common hypothesis, which is exactly the qos value of Web service, in these researchs.
But it is unpractical that system, which obtains user to assess the qos value of each Web service, on the one hand, by Web service
The qos value that provider or third party community announce be for a user it is inaccurate because actual qos value be easy by
The influence of nondeterministic network environment and the region where user;On the other hand, due to the limit of other resources such as time, cost
System.So how to obtain accurate qos value is a main problem.For this problem, it is pre- to propose various Web service QoS
Survey method, wherein matrix disassembling method is one of common technology, it in many recommendations application accuracy with higher and
Good performance.
And existing matrix decomposition Web service QoS prediction technique is all the method for centralized training, is specifically exactly,
The platform for first constructing a recommender system, collects all users to the qos value of all Web services, then using these data come
Construct a matrix decomposition model.Do so that there are several drawbacks:
(1) storage resource is wasted, recommender system needs to collect the qos value of all users, so all users are to Web service
Qos value needs to be centrally stored on some server;
(2) waste computing resource needs to train on the server when training matrix decomposition model, the training speed of model
Degree is limited to the machine quantity of server-side;
(3) private data of user cannot be protected, user is obtained the qos value of Web service by server, the preference of user
Information can may also be leaked to therewith malicious attacker, and there are privacy of user security hidden troubles.
Summary of the invention
The service quality prediction based on decentralization matrix decomposition that the invention mainly solves the technical problem of providing a kind of
Method predicts qos value by using decentralization matrix disassembling method, on the one hand, the data of user are stored in of oneself
Without being uploaded to server-side in people's equipment, the problem of waste this addresses the problem storage resource caused by centralized training model,
On the other hand, the training of model is also completed in user terminal, completes model by the non-primary data information of interaction between user
Coorinated training, the problem of can solve computing resource waste caused by centralized training, in addition, each user is to Web service
Qos value is stored in user oneself hand, to protect the individual privacy of user well, is being based on decentralization matrix decomposition
Service quality prediction technique it is universal on have extensive market prospects.
In order to solve the above technical problems, the present invention provides a kind of service quality prediction side based on decentralization matrix decomposition
Method, comprising the following steps:
DefinitionIndicate user's set, definitionIndicate Web service set, wherein
It is sharedA user andA Web service,Indicate i-th of user to the interactive information of j-th of Web service, wherein, and indicate userTo serviceThe qos value of assessment,
DefinitionThe potential eigenmatrix of user is indicated, wherein every a lineIndicate user'sPotential feature vector is tieed up, it is fixed
JusticeIt indicates to service potential characteristic tensor, definitionIndicate userThe potential eigenmatrix of service, whereinTable
Show userTo serviceK tie up potential feature vector,
(1) user's adjacent map is established:
User's adjacent map is established according to the geographical location of userTo indicate the level of intimate between user, definitionIt indicates
UserAnd userThe distance between, then, userAnd userBetween similarity can indicate are as follows:
,
Wherein:If userAnd userIn areal, then in formulaIf userWith
UserNot in areal, then in formula,It is the mapping function of the distance between user He similarity, uses
The distance between family is smaller, and similarity is bigger,
(2) range of interaction is determined:
User interacts the information of other neighbours using the method for the neighboring user interaction based on random walk:
Establishing user's adjacent mapLater, using user's adjacency matrixIndicate the adjacent map of user, definitionTable
In showing?A neighbours,Indicate the number of neighbours,, it is clear thatIt means that
UserImmediate neighbor,
Work as userThink and his direct neighboursInteractive information,Indicate userIt is chosen from his neighbours
The behavior of one user, then,
,
According to Markov probability, userChoose he the second neighbours () probability are as follows:
,
Wherein, D indicates the maximum distance of random walk,
User carries out migration to surrounding neighbours in preset maximum number of iterations with interactive information,
(3) information of interaction is determined:
In order to protect the privacy of user, which information of interaction determined between user, by qos value be decomposed into user preference it is potential to
AmountWith service preferences latent variable, that is,
,
Wherein, for each user, theThe potential feature vector of the service of a service, can decompose are as follows:
,
WhereinIt indicates global potential feature vector, indicates the common preference of all users,Indicate privately owned potential feature to
Amount indicates the personal preference of user, then loss function can indicate are as follows:
,
VectorWithIt only depends on and is stored in userIn information, andDepend not only upon userIn information, also according to
Rely the information in other adjacent users, sets secret protection agreement to exchange the information between user, thus Learning Service
Global potential feature vector, the secret protection agreement be by send the loss function of each user about gradient to him
Neighbours, to learn the potential feature vector of global service, for each userLoss functionAbout,WithGradient are as follows:
,
(4) it synthesizes and predicts qos value:
User after the completion of interaction is by global potential feature vectorWith privately owned potential feature vectorIt is potential that synthesis obtains service
Feature vector, i.e.,
,
Again by user preference latent variableWith service preferences latent variableSynthesis obtains userTo the QoS of service valuation
Value, i.e.,
。
In a preferred embodiment of the present invention, each user is stored in the qos value of service and potential feature each
At user, i.e., each userThe K for only needing to save him ties up potential feature vectorWith the potential eigenmatrix of service。
In a preferred embodiment of the present invention, the attribute of the qos value includes response time and handling capacity.
The beneficial effects of the present invention are: the present invention is based on the service quality prediction techniques of decentralization matrix decomposition by adopting
Qos value is predicted with decentralization matrix disassembling method, on the one hand, the data of user are stored on the personal device of oneself and are not necessarily to
The problem of uploading to server-side, being wasted this addresses the problem storage resource caused by centralized training model, on the other hand, model
Training also completed in user terminal, the coorinated training of model is completed between user by the non-primary data information of interaction, can be with
Computing resource waste caused by centralized training is solved the problems, such as, in addition, each user is stored in use to the qos value of Web service
In the hand of family oneself, so that the individual privacy of user is protected well, it is pre- in the service quality based on decentralization matrix decomposition
Popularizing for survey method has extensive market prospects.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing, in which:
Fig. 1 is user-clothes of service quality prediction technique one preferred embodiment of the invention based on decentralization matrix decomposition
Business calling figure;
Fig. 2 is that the observation of service quality prediction technique one preferred embodiment of the invention based on decentralization matrix decomposition obtains
QoS matrix;
Fig. 3 is that the prediction of service quality prediction technique one preferred embodiment of the invention based on decentralization matrix decomposition obtains
QoS matrix;
Fig. 4 is the decomposition diagram of prior art centralization matrix disassembling method;
Fig. 5 is the exploded pictorial of service quality prediction technique one preferred embodiment of the invention based on decentralization matrix decomposition
Figure.
Specific embodiment
The technical scheme in the embodiments of the invention will be clearly and completely described below, it is clear that described implementation
Example is only a part of the embodiments of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, this field is common
Technical staff's all other embodiment obtained without making creative work belongs to the model that the present invention protects
It encloses.
Fig. 1-Fig. 5 is please referred to, the embodiment of the present invention includes:
A kind of service quality prediction technique based on decentralization matrix decomposition, comprising the following steps:
DefinitionIndicate user's set, definitionIndicate Web service set, wherein
It is sharedA user andA Web service,Indicate i-th of user to the interactive information of j-th of Web service, wherein, andIndicate userTo serviceThe qos value of assessment,
DefinitionThe potential eigenmatrix of user is indicated, wherein every a lineIndicate user'sPotential feature vector is tieed up, is definedIt indicates to service potential characteristic tensor, definitionIndicate userThe potential eigenmatrix of service, whereinIt indicates
UserTo serviceK tie up potential feature vector,
(1) user's adjacent map is established:
User's adjacent map is established according to the geographical location of userTo indicate the level of intimate between user, definitionIt indicates
UserAnd userThe distance between, then, userAnd userBetween similarity can indicate are as follows:
,
Wherein:If userAnd userIn areal, then in formulaIf userWith
UserNot in areal, then in formula,It is the mapping function of the distance between user He similarity, uses
The distance between family is smaller, and similarity is bigger,
(2) range of interaction is determined:
User interacts the information of other neighbours using the method for the neighboring user interaction based on random walk:
Establishing user's adjacent mapLater, using user's adjacency matrixIndicate the adjacent map of user, definitionTable
ShowIn?A neighbours,Indicate the number of neighbours,, it is clear thatWith regard to table
Show userImmediate neighbor,
Work as userThink and his direct neighboursInteractive information,Indicate userIt is chosen from his neighbours
The behavior of one user, then,
,
According to Markov probability, userChoose he the second neighbours () probability are as follows:
,
Wherein, D indicates the maximum distance of random walk,
User carries out migration to surrounding neighbours in preset maximum number of iterations with interactive information,
(3) information of interaction is determined:
In order to protect the privacy of user, which information of interaction determined between user, by qos value be decomposed into user preference it is potential to
AmountWith service preferences latent variable, that is,
,
Wherein, for each user, theThe potential feature vector of the service of a service, can decompose are as follows:
,
It wherein indicates global potential feature vector, indicates the common preference of all users,Indicate privately owned potential feature to
Amount indicates userPersonal preference, then loss function can indicate are as follows:
,
VectorUser is stored in only depending onIn information, andDepend not only upon userIn information, be also relied on
Information in other adjacent users sets secret protection agreement to exchange the information between user, so that Learning Service is complete
The potential feature vector of office, the secret protection agreement are by sending each userLoss functionAboutGradient give
His neighbours, to learn the potential feature vector of global service, for each userLoss functionAbout,
WithGradient are as follows:
,
(4) it synthesizes and predicts qos value:
User after the completion of interaction is by global potential feature vectorWith privately owned potential feature vectorIt is potential that synthesis obtains service
Feature vector, i.e.,
,
Again by user preference latent variableWith service preferences latent variableSynthesis obtains userTo serviceThe QoS of assessment
Value, i.e.,
。
Preferably, each user is stored at each user the qos value of service and potential feature, i.e., each user
The K for only needing to save him ties up potential feature vectorWith the potential eigenmatrix of service。
Preferably, the attribute of the qos value includes response time and handling capacity.
This example is implemented under the premise of the technical scheme of the present invention, gives detailed embodiment and process,
But protection scope of the present invention is not limited to following examples.
The present invention is tested on disclosed obtainable QoS data collection, which contains 339 users and comment
The qos value of 5825 Web services of valence, present invention primarily contemplates two representative QoS attributes: the response time (RT) and
Handling capacity (TP), response time indicate that user issues request and receives the duration between response, and handling capacity indicates user's tune
With the message transmission rate of service.The details of data set is as follows;
QoS | User | Service | Range | Average value | Variance | Packing density |
RT(sec) | 339 | 5825 | 0~20 | 0.909 | 1.973 | 94.8% |
TP(kbps) | 339 | 5825 | 0~1000 | 47.562 | 110.797 | 92.7% |
In experiment, user calls the qos value of service with 3395825 QoS matrix indicates, wherein -1 indicates unknown qos value,
The qos value for needing to predict, this experiment random erasure partial value only save the qos value of small data density, such as: packing density=
10%, indicate that qos value is evaluated in Web service of the user only to 10%, the QoS using decentralization matrix decomposition of the invention calculates in advance
Method predicts remaining qos value, with RMSE(root-mean-square error) come evaluate QoS prediction accuracy:
Wherein,Indicate the user that observation obtainsTo serviceThe qos value of evaluation,Indicate the user that prediction obtainsTo clothes
BusinessQos value,Indicate in training set the qos value of prediction in need obtain number, RMSE is smaller, and forecasting accuracy is got over
It is high.
In order to verify feasibility and validity of the invention, this experiment is by the present invention is based on decentralization matrix decompositions
QoS prediction technique (DMF) is compared with following three QoS prediction technique:
MF: being the most basic QoS based on matrix decomposition and side method;
RMF: being the QoS prediction technique based on data Random-fuzzy protection privacy, and user's maintenance data randomized technique comes random
Their qos value is obscured, specifically, each user adds a certain range of random number on original qos value, is then sent
It is concentrated to recommender system and carries out QoS prediction.
LMF: being the QoS prediction technique of the protection privacy based on difference privacy technology.
It was found that the present invention has in the qos value to the Web service that do not evaluated in prediction user compares outstanding performance.
In the process of implementation, the data of user are stored in user terminal to the present invention, without being uploaded to service recommendation system, which reduces
Storage resource waste;The training of model is also completed in user terminal, and the collaboration for completing model between user by interaction gradient is instructed
Practice, reduces the waste of computing resource;Each user is stored in user terminal to the qos value of Web service, protects user well
Individual privacy;And in terms of forecasting accuracy, good effect has also been obtained.The contrast effect of method is as follows in above-mentioned 4;
Method | RMSE(K=5) | RMSE(K=10) | RMSE(K=15) | RMSE(K=20) |
MF | 219.59 | 193.62 | 162.39 | 143.83 |
RMF | 415.90 | 365.71 | 305.54 | 272.11 |
LMF | 277.65 | 240.81 | 206.16 | 180.85 |
DMF | 108.80 | 106.48 | 105.97 | 103.20 |
Beneficial effect the present invention is based on the service quality prediction technique of decentralization matrix decomposition is:
Qos value is predicted by using decentralization matrix disassembling method, on the one hand, the data of user are stored in the individual of oneself
Without being uploaded to server-side in equipment, the problem of waste this addresses the problem storage resource caused by centralized training model, separately
On the one hand, the training of model is also completed in user terminal, completes the association of model between user by the non-primary data information of interaction
The problem of with training, can solve computing resource waste caused by centralized training, in addition, QoS of each user to Web service
Value is stored in user oneself hand, to protect the individual privacy of user well.
The above description is only an embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright description is applied directly or indirectly in other relevant technology necks
Domain is included within the scope of the present invention.
Claims (3)
1. a kind of service quality prediction technique based on decentralization matrix decomposition, which comprises the following steps:
Definition indicates user's set, definitionWeb service set is indicated, wherein sharingA user andIt is a
Web service,Indicate i-th of user to the interactive information of j-th of Web service, wherein, andTable
Show userTo serviceThe qos value of assessment,
DefinitionThe potential eigenmatrix of user is indicated, wherein every a lineIndicate user'sPotential feature vector is tieed up, it is fixed
JusticeIt indicates to service potential characteristic tensor, definitionIndicate userThe potential eigenmatrix of service, wherein indicate use
FamilyTo serviceK tie up potential feature vector,
(1) user's adjacent map is established:
User's adjacent map is established according to the geographical location of userTo indicate the level of intimate between user, definitionIt indicates to use
FamilyAnd userThe distance between, then, userAnd userBetween similarity can indicate are as follows:
,
Wherein:If userAnd userIn areal, then in formulaIf userWith with
FamilyNot in areal, then in formula,It is the mapping function of the distance between user He similarity, user
The distance between it is smaller, similarity is bigger,
(2) range of interaction is determined:
User interacts the information of other neighbours using the method for the neighboring user interaction based on random walk:
Establishing user's adjacent mapLater, using user's adjacency matrixIndicate the adjacent map of user, definitionIt indicatesIn?A neighbours,Indicate the number of neighbours, it is clear thatMean that userImmediate neighbor,
Work as userThink and his direct neighboursInteractive information,Indicate userOne is chosen from his neighbours
The behavior of a user, then,
,
According to Markov probability, userChoose he the second neighbours () probability are as follows:
,
Wherein, D indicates the maximum distance of random walk,
User carries out migration to surrounding neighbours in preset maximum number of iterations with interactive information,
(3) information of interaction is determined:
In order to protect the privacy of user, which information of interaction determined between user, by qos value be decomposed into user preference it is potential to
AmountWith service preferences latent variable, that is,
,
Wherein, for each user, theThe potential feature vector of the service of a service, can decompose are as follows:
,
WhereinIt indicates global potential feature vector, indicates the common preference of all users,Indicate privately owned potential feature to
Amount indicates userPersonal preference, then loss function can indicate are as follows:
,
Vector sum, which only depends on, is stored in userIn information, andDepend not only upon userIn information, also rely on
Information in other adjacent users sets secret protection agreement to exchange the information between user, so that Learning Service is global
Potential feature vector, the secret protection agreement is by sending each userLoss functionAbout gradient give he
Neighbours, to learn the potential feature vector of global service, for each userLoss functionAbout,With
Gradient are as follows:
,
(4) qos value is predicted:
User after the completion of interaction is by global potential feature vector and privately owned potential feature vectorSynthesis, which obtains, services potential spy
Levy vector, i.e.,
,
Again by user preference latent variableIt is synthesized with service preferences latent variable and obtains userTo serviceThe qos value of assessment,
I.e.
。
2. the service quality prediction technique according to claim 1 based on decentralization matrix decomposition, which is characterized in that every
A user is stored at each user the qos value of service and potential feature, i.e., each userOnly need to save his K dimension
Potential feature vectorWith the potential eigenmatrix of service。
3. the service quality prediction technique according to claim 1 based on decentralization matrix decomposition, which is characterized in that institute
The attribute for stating qos value includes response time and handling capacity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811043938.6A CN109376901B (en) | 2018-09-07 | 2018-09-07 | Service quality prediction method based on decentralized matrix decomposition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811043938.6A CN109376901B (en) | 2018-09-07 | 2018-09-07 | Service quality prediction method based on decentralized matrix decomposition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109376901A true CN109376901A (en) | 2019-02-22 |
CN109376901B CN109376901B (en) | 2021-09-03 |
Family
ID=65404858
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811043938.6A Active CN109376901B (en) | 2018-09-07 | 2018-09-07 | Service quality prediction method based on decentralized matrix decomposition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109376901B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110443430A (en) * | 2019-08-13 | 2019-11-12 | 汕头大学 | A kind of service quality prediction technique based on block chain |
CN111541570A (en) * | 2020-04-22 | 2020-08-14 | 北京交通大学 | Cloud service QoS prediction method based on multi-source feature learning |
CN112600697A (en) * | 2020-12-07 | 2021-04-02 | 中山大学 | QoS prediction method and system based on federal learning, client and server |
CN112700067A (en) * | 2021-01-14 | 2021-04-23 | 安徽师范大学 | Method and system for predicting service quality under unreliable mobile edge environment |
WO2021077814A1 (en) * | 2019-10-23 | 2021-04-29 | 支付宝(杭州)信息技术有限公司 | Push model optimization method and device executed by user terminal |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104778636A (en) * | 2015-04-08 | 2015-07-15 | 中国人民解放军海军工程大学 | Ship equipment maintenance support information service recommendation method |
CN104915444A (en) * | 2015-06-29 | 2015-09-16 | 北京邮电大学 | Information recommending method and device |
CN105024886A (en) * | 2015-07-31 | 2015-11-04 | 浙江大学 | Rapid Web server QoS (Quality of Service) prediction method based on user metadata |
CN105791010A (en) * | 2016-03-07 | 2016-07-20 | 天津职业技术师范大学 | Quality-of-service (QoS) predicting method based on user implicit characteristics and collaborative filtering |
-
2018
- 2018-09-07 CN CN201811043938.6A patent/CN109376901B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104778636A (en) * | 2015-04-08 | 2015-07-15 | 中国人民解放军海军工程大学 | Ship equipment maintenance support information service recommendation method |
CN104915444A (en) * | 2015-06-29 | 2015-09-16 | 北京邮电大学 | Information recommending method and device |
CN105024886A (en) * | 2015-07-31 | 2015-11-04 | 浙江大学 | Rapid Web server QoS (Quality of Service) prediction method based on user metadata |
CN105791010A (en) * | 2016-03-07 | 2016-07-20 | 天津职业技术师范大学 | Quality-of-service (QoS) predicting method based on user implicit characteristics and collaborative filtering |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110443430A (en) * | 2019-08-13 | 2019-11-12 | 汕头大学 | A kind of service quality prediction technique based on block chain |
CN110443430B (en) * | 2019-08-13 | 2023-08-22 | 汕头大学 | Block chain-based service quality prediction method |
WO2021077814A1 (en) * | 2019-10-23 | 2021-04-29 | 支付宝(杭州)信息技术有限公司 | Push model optimization method and device executed by user terminal |
CN111541570A (en) * | 2020-04-22 | 2020-08-14 | 北京交通大学 | Cloud service QoS prediction method based on multi-source feature learning |
CN112600697A (en) * | 2020-12-07 | 2021-04-02 | 中山大学 | QoS prediction method and system based on federal learning, client and server |
CN112700067A (en) * | 2021-01-14 | 2021-04-23 | 安徽师范大学 | Method and system for predicting service quality under unreliable mobile edge environment |
CN112700067B (en) * | 2021-01-14 | 2024-05-24 | 安徽师范大学 | Method and system for predicting service quality in unreliable mobile edge environment |
Also Published As
Publication number | Publication date |
---|---|
CN109376901B (en) | 2021-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yu et al. | Personalized travel package with multi-point-of-interest recommendation based on crowdsourced user footprints | |
Roick et al. | L ocation B ased S ocial N etworks–Definition, Current State of the Art and Research Agenda | |
US10332150B2 (en) | Location event advertising | |
CN109376901A (en) | A kind of service quality prediction technique based on decentralization matrix decomposition | |
Yin et al. | Modeling location-based user rating profiles for personalized recommendation | |
Yin et al. | LCARS: A spatial item recommender system | |
JP6013653B2 (en) | Network-oriented product rollout in online social networks | |
Quercia et al. | Friendsensing: recommending friends using mobile phones | |
US10373202B2 (en) | Selection of region-specific brand pages based on location information of social networking system users | |
CN105554704A (en) | Fake-locus-based location privacy protection method for use in recommendation system | |
Amintoosi et al. | A trust-based recruitment framework for multi-hop social participatory sensing | |
Gao et al. | Mining human mobility in location-based social networks | |
Wismans et al. | Improving a priori demand estimates transport models using mobile phone data: a Rotterdam-region case | |
US20130151950A1 (en) | System and method for web presence for one or more geographical locations | |
Dang et al. | Simulating the spatial diffusion of memes on social media networks | |
Huang et al. | Interest-driven outdoor advertising display location selection using mobile phone data | |
Xu et al. | Sume: Semantic-enhanced urban mobility network embedding for user demographic inference | |
Brown et al. | Group colocation behavior in technological social networks | |
Song et al. | Recovering individual’s commute routes based on mobile phone data | |
Gutowski et al. | Improving bandit-based recommendations with spatial context reasoning: An online evaluation | |
Talavera-Garcia et al. | Unveiling the pandemic's impact on visits to Madrid’s parks: insights from mobile phone data analysis | |
Yoon et al. | COVID-19 confines recreational gatherings in Seoul to familiar, less crowded, and neighboring urban areas | |
Chow | “We know who you are and we know where you live”: a research agenda for web demographics | |
Sinnott et al. | Privacy-preserving data linkage through blind geo-spatial data aggregation | |
Mezhuyev et al. | Evaluation of the likelihood of friend request acceptance in online social networks |
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 |