Big data personalized service intelligent recommendation method and system based on privacy protection of old people

Abstract

The invention relates to the technical field of big data, in particular to a big data personalized service intelligent recommendation method and system based on old people privacy protection, and the method comprises the steps: introducing an old people user personal feature model based on an intelligent recommendation algorithm, and constructing a big data personalized online service intelligent recommendation algorithm based on random disturbance privacy protection; position information is introduced, and a big data personalized offline service intelligent recommendation algorithm based on position privacy protection is constructed; the method has the beneficial effects that big data personalized service intelligent recommendation algorithms and privacy and protection under different scenes and types are analyzed through a classification analysis method, and the corresponding big data personalized service intelligent recommendation method considering privacy protection is explored by combining privacy related problems of personalized service intelligent recommendation under different scenes and types; the privacy information of the elderly user is effectively protected, and good big data personalized service intelligent recommendation is provided.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and specifically to an intelligent recommendation method and system for big data personalized services based on the privacy protection of the elderly. Background Art

[0002] The all-round development and diversification of the elderly's pension service needs are presented. While enjoying the care and companionship in old age, they also hope to have fun and be useful in old age, and need to meet the service needs in various aspects such as life care, basic diet, daily companionship, cultural and entertainment, health guidance, and rehabilitation care. And during the process of the elderly enjoying services, the demand for personalization is also becoming stronger and stronger. For consumers, the elderly no longer want to enjoy the uniformly planned service form, but hope to have a service form that better meets their own needs and personalized preferences.

[0003] In the prior art, while bringing convenience to the elderly, a large number of services also bring difficulties in finding and selecting services. It is mainly manifested in the low efficiency of resource allocation, and the supply cannot meet the personalized and segmented pension service needs; there is an urgent need to enhance the accuracy, pertinence and effectiveness of the pension service system. On the other hand, personalized service intelligent recommendation needs to obtain the personal privacy big data of the elderly and provide recommendation services through big data analysis, which undoubtedly brings the risk of privacy leakage to elderly users. How to adopt reasonable methods to protect the privacy information of users while effectively realizing the intelligent recommendation of big data personalized services to meet the users' needs for obtaining personalized services is the focus of this research. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent recommendation method and system for big data personalized services based on the privacy protection of the elderly, so as to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: An intelligent recommendation method for big data personalized services based on the privacy protection of the elderly, the method includes the following steps:

[0006] Introduce the personal feature model of elderly users based on the intelligent recommendation algorithm, and construct an intelligent recommendation algorithm for big data personalized online services based on random perturbation privacy protection;

[0007] Introduce location information and construct an intelligent recommendation algorithm for big data personalized offline services based on location privacy protection.

[0008] Preferably, the specific operation of constructing an intelligent recommendation algorithm for big data personalized online services based on random perturbation privacy protection includes:

[0009] a) Establish a personal characteristic model for elderly users, which includes personal status, family status, and physical condition attributes, and determine the relevant user personal characteristic attributes in the model through qualitative or quantitative analysis methods;

[0010] b) Elderly users set privacy preference strategies before using the recommendation service to express their privacy protection degree requirements for data;

[0011] c) A trusted third party collects elderly user data and performs preprocessing, performs privacy protection processing on the elderly user data using different degrees of random perturbation, and stores it using a distributed cloud storage method;

[0012] d) The recommendation server uses the privacy-processed elderly user personal characteristic data and service rating data to establish an elderly user model, calculates user similarity, and finds similar elderly users;

[0013] e) Predict the service item set scores of similar elderly users, and return the TOP-N personalized online service recommendation results to the elderly users.

[0014] Preferably, it also includes preprocessing the elderly user data before the recommendation server requests to use the randomly perturbed privacy-processed elderly user personal characteristic data and service rating data to improve data quality and recommendation accuracy.

[0015] Preferably, the specific operations for constructing a big data personalized offline service intelligent recommendation algorithm based on location privacy protection include:

[0016] a) Introduce the real-time location information of elderly users as a constraint condition to construct a big data personalized offline service intelligent recommendation algorithm;

[0017] b) After the elderly user initiates a recommendation request, the location server anonymizes the real-time location of the elderly user;

[0018] c) A trusted third party collects data related to the personal characteristics and preferences of elderly users, and after data perturbation processing, stores it in the cloud using a distributed storage method;

[0019] d) The recommendation server applies to access the privacy-processed data, processes the elderly user personal characteristic and rating data, establishes an elderly user model, calculates the similarity degree of users and service prediction scores;

[0020] e) The recommendation server combines the anonymized real-time location of the user, calculates the distance cost between the service items of the recommended service set and the elderly user, adjusts the service prediction score, and obtains the final Top-N service item recommendation result;

[0021] f) Return the personalized service recommendation result to the trusted third party, and after sorting, return it to the target elderly user.

[0022] Preferably, the location server anonymizes the real-time location of elderly users by using a location privacy protection method based on collaborative user mean K-anonymity to protect the real-time location privacy of elderly users.

[0023] A big data personalized service intelligent recommendation system based on elderly privacy protection is applied to a big data personalized service intelligent recommendation method based on elderly privacy protection. The system includes:

[0024] An online recommendation module that introduces a personal characteristic model of elderly users based on an intelligent recommendation algorithm and constructs a big data personalized online service intelligent recommendation algorithm based on random perturbation privacy protection;

[0025] An offline recommendation module that introduces location information and constructs a big data personalized offline service intelligent recommendation algorithm based on location privacy protection.

[0026] Preferably, the specific operations of the online recommendation module include:

[0027] a) Establish a personal characteristic model of elderly users, which includes personal status, family status, and physical status attributes, and determine the relevant user personal characteristic attributes in the model through qualitative or quantitative analysis methods;

[0028] b) Elderly users set privacy preference policies before using the recommendation service to express their privacy protection degree requirements for data;

[0029] c) A trusted third party collects elderly user data and performs preprocessing, performs privacy protection processing on the elderly user data using different degrees of random perturbation, and stores it using a distributed cloud storage method;

[0030] d) The recommendation server uses the privacy-processed personal characteristic data and service score data of elderly users to establish an elderly user model, calculates user similarity, and finds similar elderly users;

[0031] e) Predict the service item set scores of similar elderly users and return the TOP-N personalized online service recommendation results to elderly users.

[0032] Preferably, the online recommendation module also includes preprocessing the elderly user data before the recommendation server requests to use the personal characteristic data and service score data of elderly users after random perturbation privacy processing to improve data quality and recommendation accuracy.

[0033] Preferably, the specific operations of the offline recommendation module include:

[0034] a) Introduce the real-time location information of elderly users as a constraint condition and construct a big data personalized offline service intelligent recommendation algorithm;

[0035] b) After the elderly user initiates a recommendation request, the location server anonymizes the real-time location of the elderly user;

[0036] c) The trusted third party collects data related to the personal characteristics and preferences of the elderly user. After data perturbation processing, it is stored in the cloud using a distributed storage method;

[0037] d) The recommendation server applies for access to the privacy-processed data, processes the personal characteristics and rating data of the elderly user, establishes an elderly user model, and calculates the similarity degree of users and service prediction scores;

[0038] e) The recommendation server combines the anonymized real-time location of the user, calculates the distance cost between the service items in the recommended service set and the elderly user, adjusts the service prediction score, and obtains the final Top-N service item recommendation result;

[0039] f) Return the personalized service recommendation result to the trusted third party, and after sorting, return it to the target elderly user.

[0040] Preferably, the location server in the offline recommendation module anonymizes the real-time location of the elderly user using a location privacy protection method based on collaborative user mean K-anonymity to protect the real-time location privacy of the elderly user.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] The big data personalized service intelligent recommendation method and system based on the privacy protection of the elderly proposed by the present invention, through big data analysis technology, identify and process a large amount of information, learn and understand the needs and preferences of the elderly users, mine the potential interest preferences of the elderly users based on collaborative filtering technology, and based on the big data analysis results, recommend service types that meet their service needs and personalized needs to the elderly users, helping the elderly users make quick decisions and also helping service providers achieve more effective marketing. Through classification analysis, the big data personalized service intelligent recommendation algorithm and privacy and protection in different scenarios and types are analyzed, combined with the privacy-related issues of personalized service intelligent recommendation in different scenarios and categories, and the corresponding big data personalized service intelligent recommendation methods considering privacy protection are explored to achieve both effective protection of the privacy information of the elderly users and the provision of good big data personalized service intelligent recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is the flowchart of the method of the present invention;

[0044] Figure 2 is the flowchart of the online service intelligent recommendation algorithm of the present invention;

[0045] Figure 3 is the architecture diagram of the online recommendation module of the present invention;

[0046] Figure 4 This is the flowchart of the intelligent recommendation algorithm for offline services of the present invention. Detailed implementation manners

[0047] In order to clearly and completely describe the objectives, technical solutions of the present invention and make the advantages more clearly understood, the following further details the embodiments of the present invention with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are some but not all of the embodiments of the present invention, and are only used to explain the embodiments of the present invention, rather than limiting the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0048] Embodiment 1, please refer to Figures 1 to 3 The present invention provides a technical solution: intelligent recommendation for personalized online services of big data based on random perturbation privacy protection

[0049] a) Intelligent recommendation algorithm for personalized online services of big data

[0050] The intelligent recommendation for personalized online services of big data in this patent fully considers big data analysis and improves the collaborative filtering recommendation based on elderly users by combining the personal characteristic model of elderly users. The personal characteristics of elderly users refer to the attributes related to the elderly users themselves, including personal conditions (such as gender, age, etc.), family conditions (such as spouse status, degree of care by children, living style, financial savings of the elderly, total family income level, etc.), and physical conditions (such as degree of self-care, self-evaluated physical condition, total number of diseases, number of somatic symptoms, etc.). Different service types are related to different personal characteristics of elderly users, and relevant user personal characteristic attributes in the model are determined by qualitative or quantitative analysis methods. The intelligent recommendation algorithm for personalized online services of big data combined with the user personal characteristic model can calculate the user similarity more fully, and when there is no historical evaluation information of the user for online services, the user similarity can also be calculated based on the user personal characteristics for intelligent recommendation of online services, which can solve the data sparsity and cold start problems to a certain extent.

[0051] b) Intelligent recommendation method for personalized online services of big data considering privacy protection

[0052] First, each elderly user needs to set a privacy preference strategy before using the recommendation to express the privacy protection requirements for the data. Then, a trusted third party is used to collect elderly user data, pre-process the big data, and use different degrees of random perturbations to protect the privacy of the elderly user data. Distributed cloud storage is used on the big data storage to further prevent attackers from attacking. After that, the recommendation server requests to use the elderly user's personal feature data and service rating data after random perturbation privacy processing to establish an elderly user model, calculate the comprehensive elderly user similarity, and find similar elderly users. Finally, the service item set of similar elderly users is rated and predicted, and the TOP-N personalized online service recommendation results are returned to the elderly user.

[0053] Embodiment 2, based on embodiment 1, refers to the attached Figure 4 , proposed: Intelligent recommendation of personalized offline services based on big data and location privacy protection.

[0054] a) Big data personalized offline service intelligent recommendation algorithm

[0055] The big data personalized online service intelligent recommendation algorithm studied in this technology introduces location information as a constraint condition so that the recommendation results meet the location scenario needs of elderly users.

[0056] b) Intelligent recommendation algorithm for personalized offline services based on big data with privacy protection in mind

[0057] Based on the big data personalized online service intelligent recommendation method based on random perturbation privacy protection, the location privacy protection method is further introduced into the big data personalized offline service intelligent recommendation, and the big data personalized offline service intelligent recommendation method considering location privacy protection is constructed, which can satisfy the elderly users to obtain personalized offline service intelligent recommendation while protecting their real-time location and other privacy information of the elderly users. After the elderly user initiates the recommendation request, the location server first anonymizes the real-time location of the elderly user and sends it to the recommendation server. At the same time, the trusted third party collects the user's personal characteristics and preference related data and stores them in the cloud in a distributed storage manner after data perturbation processing. Then, the recommendation server applies for access to the privacy-processed data of the trusted third party, performs big data processing on the elderly user's personal characteristics and rating data, builds the elderly user model, calculates the similar user degree, and then calculates the service prediction score. The recommendation server further combines the user's anonymized real-time location, calculates the distance cost between the service items in the recommended service set and the elderly user, adjusts the service prediction score, and obtains the final Top-N service items as the final recommendation result. Finally, the personalized service recommendation results are returned to the trusted third party and returned to the target elderly user after sorting.

[0058] Solve the problem of the lack of effectiveness and pertinence in the supply of elderly care services and the imbalance between supply and demand. Since the personalized recommendation tools provided based on big data analysis in the past involved users' privacy information, bringing the risk of privacy leakage, and there was no special service recommendation for elderly users. This patent focuses on the privacy issues of intelligent recommendation of big data personalized services, analyzes the intelligent recommendation algorithms for personalized elderly users for online and offline services respectively, and combines privacy protection technologies to protect the privacy of elderly users, aiming to change the past extensive way of elderly care service supply, establish a precision elderly care service system, and fully meet the elderly care needs of the elderly.

[0059] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A big data personalized service intelligent recommendation method based on privacy protection for the elderly, characterized by: The method comprises the following steps: Based on the intelligent recommendation algorithm, the personal characteristic model of elderly users is introduced to build an intelligent recommendation algorithm for big data personalized online services based on random perturbation privacy protection; Introduce location information and build a big data personalized offline service intelligent recommendation algorithm based on location privacy protection.

2. The big data personalized service intelligent recommendation method based on privacy protection for the elderly according to claim 1 is characterized by: The specific operations of building a big data personalized online service intelligent recommendation algorithm based on random perturbation privacy protection include: a) Establishing a personal characteristic model for elderly users, which includes personal status, family status and physical status attributes, and determining the relevant personal characteristic attributes of users in the model through qualitative or quantitative analysis methods; b) Elderly users set privacy preference strategies before using recommendation services and express their demand for data privacy protection; c) A trusted third party collects and pre-processes the data of elderly users, uses random perturbations of varying degrees to protect the privacy of the data, and stores it in a distributed cloud storage manner; d) The recommendation server uses the privacy-processed personal feature data and service rating data of elderly users to build an elderly user model, calculate user similarity, and find similar elderly users; e) Predict the ratings of the service item sets of similar elderly users and return the TOP-N personalized online service recommendation results to the elderly users.

3. The big data personalized service intelligent recommendation method based on privacy protection for the elderly according to claim 2 is characterized by: It also includes preprocessing the elderly user data before the recommendation server requests the elderly user's personal feature data and service rating data after random perturbation privacy processing to improve data quality and recommendation accuracy.

4. The big data personalized service intelligent recommendation method based on privacy protection for the elderly according to claim 1 is characterized by: The specific operations of building a big data personalized offline service intelligent recommendation algorithm based on location privacy protection include: a) Introducing the real-time location information of elderly users as a constraint condition to build a big data personalized offline service intelligent recommendation algorithm; b) After the elderly user initiates a recommendation request, the location server hides the elderly user's real-time location; c) A trusted third party collects data related to the personal characteristics and preferences of elderly users, and after data disturbance processing, stores it in the cloud using a distributed storage method; d) The recommendation server applies for access to privacy-processed data, processes the personal characteristics and rating data of elderly users, builds an elderly user model, and calculates similar user degrees and service prediction ratings; e) The recommendation server combines the user's hidden real-time location, calculates the distance cost between the service items in the recommended service set and the elderly user, adjusts the service prediction score, and obtains the final Top-N service item recommendation results; f) The personalized service recommendation results are returned to the trusted third party, and then returned to the target elderly users after sorting.

5. The big data personalized service intelligent recommendation method based on privacy protection for the elderly according to claim 4 is characterized by: The location server conceals the real-time location of elderly users and adopts a location privacy protection method based on collaborative user mean K-anonymity to protect the real-time location privacy of elderly users.

6. A big data personalized service intelligent recommendation system based on privacy protection for the elderly, applied to the big data personalized service intelligent recommendation method based on privacy protection for the elderly as described in any one of claims 1 to 5, characterized in that: The system comprises: The online recommendation module introduces the personal feature model of elderly users based on the intelligent recommendation algorithm, and constructs an intelligent recommendation algorithm for big data personalized online services based on random perturbation privacy protection; The offline recommendation module introduces location information and builds a big data personalized offline service intelligent recommendation algorithm based on location privacy protection.

7. The big data personalized service intelligent recommendation system based on privacy protection for the elderly according to claim 6 is characterized by: The specific operations of the online recommendation module include: a) Establishing a personal characteristic model for elderly users, which includes personal status, family status and physical status attributes, and determining the relevant personal characteristic attributes of users in the model through qualitative or quantitative analysis methods; b) Elderly users set privacy preference strategies before using recommendation services and express their demand for data privacy protection; c) A trusted third party collects and pre-processes the data of elderly users, uses random perturbations of varying degrees to protect the privacy of the data, and stores it in a distributed cloud storage manner; d) The recommendation server uses the privacy-processed personal feature data and service rating data of elderly users to build an elderly user model, calculate user similarity, and find similar elderly users; e) Predict the ratings of the service item sets of similar elderly users and return the TOP-N personalized online service recommendation results to the elderly users.

8. The big data personalized service intelligent recommendation system based on privacy protection for the elderly according to claim 7 is characterized by: The online recommendation module also includes pre-processing the elderly user data before the recommendation server requests the elderly user's personal feature data and service rating data after random perturbation privacy processing to improve data quality and recommendation accuracy.

9. The big data personalized service intelligent recommendation system based on privacy protection for the elderly according to claim 6 is characterized by: The offline recommendation module specifically operates as follows: a) Introducing the real-time location information of elderly users as a constraint condition to build a big data personalized offline service intelligent recommendation algorithm; b) After the elderly user initiates a recommendation request, the location server hides the elderly user's real-time location; c) A trusted third party collects data related to the personal characteristics and preferences of elderly users, and after data disturbance processing, stores it in the cloud using a distributed storage method; d) The recommendation server applies for access to privacy-processed data, processes the personal characteristics and rating data of elderly users, builds an elderly user model, and calculates similar user degrees and service prediction ratings; e) The recommendation server combines the user's hidden real-time location, calculates the distance cost between the service items in the recommended service set and the elderly user, adjusts the service prediction score, and obtains the final Top-N service item recommendation results; f) The personalized service recommendation results are returned to the trusted third party, and then returned to the target elderly users after sorting.

10. The big data personalized service intelligent recommendation system based on privacy protection for the elderly according to claim 9 is characterized by: The location server in the offline recommendation module conceals the real-time location of the elderly user and adopts a location privacy protection method based on collaborative user mean K-anonymity to protect the real-time location privacy of the elderly user.

Images