Cloud computer recommendation method and system applied to mobile office

Abstract

The application provides a cloud computer recommendation method and system applied to mobile office. A cloud server acquires historical record information of mobile device accessing a cloud computer, acquires a current geographic position of a user mobile device, acquires all available cloud computers existing in the geographic position according to geographic position analysis, calculates a maximum likelihood P1 of a cloud computer ID / name existing in the current geographic position being accessed by an outsider, statistically analyzes a probability P2 of each cloud computer ID / name accessed by the user using the mobile device, matches a corresponding weight, and obtains a ranking of the cloud computer ID / name most likely to be used by the user for selection and switching. Through the method, the cloud server can predict different scene cloud computers required to be accessed by the mobile device user through information analysis, thereby achieving active provision of use prediction and intelligent recommendation, switching guidance service for the user accessing the cloud computer.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of cloud computers, and more particularly relates to a cloud computer recommendation method and system applied to mobile office. BACKGROUND

[0002] The cloud computer technology provides virtual cloud desktop services for users based on virtualization technology and Clink protocol. Users can access cloud computers through various devices such as PCs, mobile phones, and cloud terminals, enter the corresponding cloud computer scene, and rely on the characteristics of low latency and high speed of 5G+ network to meet the needs of mobile office and leisure entertainment through cloud computers.

[0003] Nowadays, cloud computer technology is becoming more and more mature, and more and more scenarios can use cloud computers. Users using mobile devices (such as mobile phones, cloud terminals, etc.) can access many different scenarios of cloud computers. However, due to the characteristics of mobile office, users need to access different cloud computers in different geographical locations, or even access multiple different cloud computers. This requires users to operate mobile devices to find the cloud computer to be accessed (such as scanning the two-dimensional code of the cloud computer or connecting the website address of the cloud computer, etc.). If it is necessary to switch to another cloud computer, and the user does not have the two-dimensional code or website address of the cloud computer, the user needs to operate to find it, which is relatively cumbersome. There is no good way in the prior art to achieve quick cloud computer access and switching. SUMMARY

[0004] The present application provides a cloud computer recommendation method and system applied to mobile office. The cloud server predicts the different scenarios of cloud computers that the mobile device user needs to access through information analysis, thereby providing the user with access to cloud computers, use prediction, intelligent recommendation, and switching guidance services.

[0005] To achieve the above purpose, the technical solution of the present application is as follows:

[0006] A cloud computer recommendation method applied to mobile office, comprising:

[0007] S1, the mobile device of the user is installed with an APP for accessing cloud computers. After the mobile device is started and connected to the cloud server, the cloud server obtains the historical record information of the mobile device accessing the cloud computer. The historical record information includes the cloud computer ID / name, user IP information, and user geographical location information.

[0008] S2. Obtain the current geographical location of the user's mobile device, and analyze the geographical location to obtain all available cloud computers existing in that geographical location; for example, if it is determined that the user's mobile device is located in a corporate building, the cloud server can obtain the cloud computer ID / name of that corporate building; if it is determined that the user's mobile device is located in a school, the cloud server can obtain the cloud computer ID / name of that school; if it is determined that the user's mobile device is located in an industrial park, the cloud server can obtain the cloud computer ID / name of all the companies in that park.

[0009] S3. Based on the big data of cloud computer access records, the cloud server calculates the maximum likelihood P1 of the cloud computer ID / name existing in the current geographical location being accessed by outsiders.

[0010] S4. The cloud server statistically analyzes the probability P2 of each cloud computer ID / name accessed by the user using the mobile device based on the record information of the user's historical access to cloud computer ID / name.

[0011] S5. Based on the probabilities obtained in steps S3 and S4, match the corresponding weights to obtain the ranking of the cloud computer IDs / names most likely to be used by the user, and provide a list of cloud computer IDs / names on the user's mobile device according to the ranking for the user to select and switch.

[0012] F = Q1A + Q2B + Q3C + Q4D;

[0013] Where F is the ranking score; A is the probability that the cloud computer ID / name is accessed by outsiders; B is the probability that the cloud computer ID / name is accessed by outsiders at the current time; A and B are the maximum likelihood P1 of the cloud computer ID / name being accessed by outsiders at the current geographical location.

[0014] C represents the probability that the user accesses their frequently used cloud computer ID / name using a mobile device; D represents the probability that the user accesses the current cloud computer ID / name using a mobile device; C and D are derived from the probability P2 of each cloud computer ID / name that the user has accessed using a mobile device.

[0015] Q1, Q2, Q3, and Q4 are the weights set for A, B, C, and D, respectively.

[0016] S6. Store the user's current cloud computer usage information in the user's history record and the cloud server's big data record of cloud computer usage scenarios.

[0017] Furthermore, the calculation process of the maximum likelihood P1 in step S3 includes:

[0018] According to the access record of the cloud computer ID / name existing in the geographical position in the big data, the maximum likelihood function of accessing each cloud computer ID / name in the current geographical position is calculated as follows:

[0019] P1=f D (x1,x2,…,x n |θ);

[0020] Wherein P1 is the likelihood; f D is a probability distribution function;

[0021] x1,x2,…,x n is the access probability of the cloud computer existing in the geographical position; n is the number of the cloud computers existing in the geographical position, and the calculation method is as follows:

[0022]

[0023] Wherein, m i is the access times of the i-th cloud computer by the outsiders;

[0024] θ is a distribution parameter, also called a data fluctuation parameter, which is calculated by using an L2 regular formula, and specifically is as follows:

[0025]

[0026] Wherein W is the weight vector of the cloud computer existing in the geographical position.

[0027] Further, the statistical analysis step in step S4 comprises:

[0028] The normal distribution algorithm is used, and the cloud computer ID / name and the corresponding access user IP information and access user geographical position information in the record information are taken as the parameters of the normal distribution algorithm to obtain a normal curve, and the probability of the user using the mobile device to access the cloud computer existing in other geographical positions is obtained according to the normal curve; the specific process is as follows:

[0029] The normal distribution curve is constructed, and the distribution of parameter 1: the cloud computer ID / name and the access user IP information and parameter 2: the cloud computer ID / name and the access user geographical position information is respectively represented, then the two normal distribution curves are two-dimensionally fitted to obtain the final probability P2.

[0030] The application further provides a cloud computer recommendation system applied to mobile office, comprising:

[0031] The first obtaining module is configured to obtain, after a mobile device of a user for accessing a cloud computer is connected to a cloud server, record information of the user of the mobile device historically accessing the cloud computer, wherein the record information comprises a cloud computer ID / name, user IP information, and user geographic location information.

[0032] The second obtaining module is configured to obtain a current geographic location of the mobile device of the user, and analyze all available cloud computers existing in the geographic location according to the geographic location.

[0033] The first analysis module is configured to calculate, by the cloud server, a maximum likelihood P1 of the cloud computer ID / name existing in the current geographic location being accessed by a stranger according to big data accessed by the cloud computer ID / name.

[0034] The second analysis module is configured to statistically analyze, by the cloud server, a probability P2 of the user using the mobile device to access a cloud computer existing in another geographic location according to the record information of the user of the mobile device historically accessing the cloud computer.

[0035] The sorting and displaying module is configured to obtain a sorting of the cloud computer ID / name most likely to be used by the user according to the probabilities obtained by the first analysis module and the second analysis module, and provide a list of the cloud computer ID / name on the mobile device of the user according to the sorting, so that the user can select and switch.

[0036] F=Q1A+Q2B+Q3C+Q4D.

[0037] Wherein F is a sorting score; A is a probability of the cloud computer ID / name being accessed by a stranger; B is a probability of the cloud computer ID / name not being accessed by a stranger; A and B are from the maximum likelihood P1 of the cloud computer ID / name existing in the current geographic location being accessed by a stranger;

[0038] C is a probability of the user using the mobile device to access a cloud computer ID / name frequently used by the user; D is a probability of the user using the mobile device to access another cloud computer ID / name; C and D are from the probability P2 of each cloud computer ID / name accessed by the user using the mobile device;

[0039] Q1, Q2, Q3, and Q4 are weights set for A, B, C, and D respectively.

[0040] The data recording module is configured to store record information of the user using the cloud computer this time into the historical record of the user and the big data record of the cloud computer use scenario of the cloud server.

[0041] Further, the first analysis module comprises:

[0042] The computing unit: according to the access record of the cloud computer ID / name existing in the geographical position in the big data, the maximum likelihood function of accessing each cloud computer ID / name in the current geographical position is calculated as follows:

[0043] P1=f D (x1,x2,…,x n |θ);

[0044] Wherein P1 is the likelihood; f D is a probability distribution function;

[0045] x1,x2,…,x n is the access probability of the cloud computer existing in the geographical position; n is the number of cloud computers existing in the geographical position, and the calculation method is as follows:

[0046]

[0047] Wherein, m i is the access frequency of the i-th cloud computer by the outsider;

[0048] θ is a distribution parameter, also known as a data fluctuation parameter, which is calculated by using an L2 regular formula, and specifically is:

[0049]

[0050] Wherein W is the weight vector of the cloud computer existing in the geographical position.

[0051] Further, the second analysis module comprises:

[0052] A normal computing unit is configured to use a normal distribution algorithm, and take the cloud computer ID / name in the record information and the corresponding user IP information when accessing and the user geographical position information when accessing as parameters of the normal distribution algorithm to obtain a normal curve, and obtain the probability of the user using the mobile device to access the cloud computer existing in other geographical positions according to the normal curve; and specifically as follows:

[0053] A normal distribution curve is constructed to represent the distribution of parameter 1: the cloud computer ID / name and the user IP information when accessing, and parameter 2: the cloud computer ID / name and the user geographical position information when accessing, and then two-dimensional fitting is performed on the above two normal distribution curves to obtain the final probability P2.

[0054] Compared with the prior art, the present application has the following beneficial effects:

[0055] Through the method of the present application, the cloud server can predict the cloud computer of different scenes required to be accessed by the mobile device user through information analysis, so as to realize the active provision of the use prediction and intelligent recommendation, switching guidance service of accessing the cloud computer for the user. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 is a flowchart of the present application. DETAILED DESCRIPTION

[0057] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0058] The cloud computer recommendation method and system applied to mobile office provided by the present application are suitable for the case that a user uses a mobile device to switch to access different cloud computer IDs / names, and the switching service is intelligently guided by a cloud server.

[0059] In the present embodiment, it is assumed that a certain user is on a business trip in a certain industrial park, and the address is a certain enterprise in the industrial park. According to the method of the present application, the specific process of the cloud server providing intelligent guidance switching service is as shown in Figure 1 , which includes:

[0060] Step 1: After the user arrives at the destination, the mobile device is started. The mobile device is pre-installed with an APP for accessing a cloud computer. When the mobile device is started, it is first connected to the cloud server. After the connection, the cloud server obtains the record information of the historical access of the cloud computer by the mobile device. The record information includes the cloud computer ID / name, the user IP information, and the user geographic location information.

[0061] It should be noted that the cloud server stores the big data of the access records of all users to different cloud computer IDs / names. The main contents of the data records include: the user, the access time, the accessed cloud computer ID / name, the network IP at the time of access, the geographic location information at the time of access, etc. The geographic location information is mainly the longitude and latitude information provided by the mobile device.

[0062] Step 2: The cloud server obtains the current geographic location of the mobile device through the current longitude and latitude information provided by the mobile device. According to the geographic location, the cloud server queries the database to find that the geographic location is a certain industrial park. If the current geographic location (longitude and latitude information) can be accurately positioned to a certain enterprise in the park, only the cloud computer ID / name of the enterprise is obtained. If it cannot be accurately positioned to a certain enterprise, the cloud computer ID / name of all enterprises in the park is obtained.

[0063] Step 3: The cloud server calculates the probability of the cloud computer ID / name existing in the current geographic location being accessed by outsiders according to the big data of the access records of the cloud computer ID / name stored in the record.

[0064] The calculation method includes:

[0065] 1. According to the access record of the cloud computer ID / name existing in the geographical position in the big data, a maximum likelihood function of accessing the cloud computer ID / name in the current geographical position is calculated;

[0066] The maximum likelihood function is as follows:

[0067] P1=f D (x1,x2,…,x n |θ);

[0068] Wherein P1 is the likelihood; f D is the probability distribution function; θ is the distribution parameter, also known as the data fluctuation parameter;

[0069] x1,x2,…,x n is the access probability of the cloud computer existing in the geographical position; n is the number of cloud computers existing in the geographical position;

[0070] The calculation method of the access probability of the cloud computer is as follows:

[0071]

[0072] Wherein, m i is the number of times of accessing the i-th cloud computer by the outsider;

[0073] 2. The regular distribution algorithm is used as the data fluctuation parameter of the outsider accessing the cloud computer ID / name; the regular distribution algorithm uses the L2 regular formula, which is specifically as follows:

[0074]

[0075] Wherein W is the weight vector of the cloud computer existing in the geographical position;

[0076] 3. The maximum likelihood P1 is obtained by calculating according to the above function and the fluctuation parameter.

[0077] Fourth step: According to the record information of the mobile device user accessing the cloud computer ID / name in the history, the cloud server statistically analyzes the probability P2 of each cloud computer ID / name accessed by the user using the mobile device;

[0078] The statistical analysis step includes:

[0079] The normal distribution algorithm is used, and the cloud computer ID / name in the record information and the corresponding user IP information and user geographical position information at the time of access are taken as the parameters of the normal distribution algorithm to obtain the normal curve, and the probability of the user using the mobile device to access the cloud computer existing in other geographical positions is obtained according to the normal curve; specifically as follows:

[0080] Constructing normal distribution curves respectively representing parameter 1: cloud computer ID / name and user IP information at the time of access; and parameter 2: cloud computer ID / name and user geographic location information at the time of access, and then performing two-dimensional fitting on the two normal distribution curves to obtain the final probability P2;

[0081] Fifth step: according to the probabilities obtained in steps S3 and S4, matching corresponding weights to obtain the ranking of cloud computer ID / name most likely to be used by the user, and providing a cloud computer ID / name list on the mobile device of the user according to the ranking for the user to switch cloud computer ID / name;

[0082] F=Q1A+Q2B+Q3C+Q4D;

[0083] Wherein F is the ranking score; A is the probability of the cloud computer ID / name being accessed by outsiders; B is the probability of the cloud computer ID / name being accessed by outsiders at the current time (A and B are from the maximum likelihood P1); C is the probability of the user using the mobile device to access the cloud computer ID / name commonly used by the user; D is the probability of the user using the mobile device to access the current cloud computer ID / name (C and D are from the probability P2 of each cloud computer ID / name accessed by the user using the mobile device); Q1, Q2, Q3, and Q4 are weights set for A, B, C, and D respectively. The weights are set according to the working nature of the user (such as frequently traveling or occasionally traveling, dependence on cloud computers for work, etc.).

[0084] The various cloud computer ID / names existing in the industrial park and the cloud computer ID / name commonly used by the user are ranked according to the ranking score for the user to select or switch.

[0085] Sixth step: storing the record information of the user using the cloud computer this time into the history record of the user and the big data record of the cloud computer use scenario of the cloud server.

[0086] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A cloud computer recommendation method applied to mobile office, characterized in that, The application relates to a cloud computer access method and device. S1, an APP for accessing a cloud computer is installed on a user's mobile device, after the mobile device is started and connected to a cloud server, the cloud server obtains historical record information of the mobile device accessing the cloud computer; the historical record information comprises a cloud computer ID / name, user IP information and user geographical position information; S2, the current geographical position of the user's mobile device is obtained, and all available cloud computers existing in the geographical position are obtained according to geographical position analysis; S3, the cloud server calculates the maximum likelihood P1 of the cloud computer ID / name existing in the current geographical position being accessed by an outsider according to big data of cloud computer access records; S4, the cloud server statistically analyzes the probability P2 of each cloud computer ID / name accessed by the user using the mobile device according to record information of the cloud computer ID / name accessed by the user using the mobile device; S5, the probability obtained in steps S3 and S4 is matched with corresponding weights to obtain the ranking of the cloud computer ID / name most likely to be used by the user, and a cloud computer ID / name list is provided on the mobile device of the user according to the ranking, so that the user can select and switch; F=Q1A+Q2B+Q3C+Q4D; Wherein F is a ranking score; A is the probability of the cloud computer ID / name being accessed by an outsider; B is the probability of the cloud computer ID / name being accessed by an outsider at the current time; A and B are from the maximum likelihood P1 of the cloud computer ID / name existing in the current geographical position being accessed by an outsider; C is the probability of the cloud computer ID / name being accessed by the user using the mobile device; D is the probability of the cloud computer ID / name being accessed by the user using the mobile device; C and D are from the probability P2 of each cloud computer ID / name accessed by the user using the mobile device; Q1, Q2, Q3 and Q4 are weights set for A, B, C and D respectively; S6, record information of the user using the cloud computer this time is stored in the historical record of the user and the big data record of the cloud computer use scene of the cloud server; The calculation process of the maximum likelihood P1 in step S3 comprises: According to the access record of the cloud computer ID / name existing in the geographical position in big data, the maximum likelihood function of accessing each cloud computer ID / name in the current geographical position is calculated as follows: P1 = f D (x1,x2,…,x n |θ) where P1is the likelihood ratio; f D is a probability distribution function; x1, x2,..., x n The access probability of the cloud computer existing in the geographic location; n is the number of cloud computers existing in the geographic location, and the calculation method is: wherein m i is the number of times the ith cloud computer is accessed by a foreigner; Theta is a distribution parameter, also known as a data fluctuation parameter, which is calculated by using an L2 regular formula, and the specific formula is as follows: Wherein W is a weight vector of the cloud computer existing in the geographical position; The statistical analysis step in step S4 comprises: A normal distribution algorithm is used, the cloud computer ID / name and corresponding user IP information and user geographical position information in the record information are taken as parameters of the normal distribution algorithm, a normal curve is obtained, and the probability of the user using the mobile device to access the cloud computer existing in other geographical positions is obtained according to the normal curve; the specific process is as follows: Constructing normal distribution curves respectively representing parameters 1: cloud computer ID / name and user IP information at the time of access; and parameters 2: cloud computer ID / name and user geographic location information at the time of access, and then performing two-dimensional fitting on the above two normal distribution curves to obtain the final probability P2.

2. A cloud computer recommendation system applied to mobile office, characterized in that, Comprise: The first acquisition module: for the user's mobile device for accessing the cloud computer to connect the cloud server after starting, the cloud server acquires the record information of the user's history of accessing the cloud computer; the record information includes cloud computer ID / name, user IP information, user geographic location information; The second acquisition module: for acquiring the current geographic location of the user's mobile device, and analyzing and acquiring all available cloud computers existing in the geographic location according to the geographic location; The first analysis module: for the cloud server to calculate the maximum likelihood rate P1 of the cloud computer ID / name accessed by the cloud server according to the big data of the cloud computer ID / name; The second analysis module: for the cloud server to statistically analyze the probability P2 of the user using the mobile device to access the cloud computer existing in other geographic locations according to the record information of the user's history of accessing the cloud computer; The sorting display module: according to the probability obtained by the first analysis module and the second analysis module, matching the corresponding weight, obtaining the sorting of the cloud computer ID / name most likely to be used by the user, and providing the cloud computer ID / name list on the user's mobile device according to the sorting, so that the user can select and switch; F=Q1A+Q2B+Q3C+Q4D; Wherein F is the sorting score; A is the probability of the cloud computer ID / name being accessed by outsiders; B is the probability of the cloud computer ID / name not being accessed by outsiders; A and B come from the maximum likelihood rate P1 of the cloud computer ID / name existing in the current geographic location being accessed by outsiders; C is the probability of the user using the mobile device to access the cloud computer ID / name he often uses; D is the probability of the user using the mobile device to access other cloud computer ID / name; C and D come from the probability P2 of each cloud computer ID / name accessed by the user using the mobile device; Q1, Q2, Q3, Q4 are respectively the weights set for A, B, C, D; The data recording module: stores the record information of the user's current use of the cloud computer into the user's history record and the big data record of the cloud computer use scene of the cloud server; The first analysis module comprises: The calculation unit: according to the access record of the cloud computer ID / name existing in the geographic location in the big data, the maximum likelihood rate function of accessing each cloud computer ID / name in the current geographic location is calculated as follows: P1 = f D (x1,x2,…,x n |θ) where P1is the likelihood ratio; f D is a probability distribution function; x1, x2,..., x n the access probability of the cloud computer existing in the geographic location; n is the number of cloud computers existing in the geographic location, and the calculation method is: wherein m i is the number of times the ith cloud computer is accessed by a foreigner; θ is a distribution parameter, also known as a data fluctuation parameter, which is calculated by using an L2 regular formula, specifically: Wherein W is the weight vector of the cloud computer existing in the geographic location; The second analysis module comprises: A normal calculation unit is configured to employ a normal distribution algorithm, take the cloud computer ID / name in the record information and the corresponding user IP information at the time of access and user geographic location information at the time of access as parameters of the normal distribution algorithm, obtain a normal curve, and acquire a probability of the user using a mobile device to access a cloud computer existing in another geographic location according to the normal curve; specifically as follows: A normal distribution curve is constructed to represent the distribution of parameter 1: the cloud computer ID / name and the user IP information at the time of access, and parameter 2: the cloud computer ID / name and the user geographic location information at the time of access, and then two normal distribution curves are fitted in two dimensions to obtain the final probability P2.

Images