Anti-quantum service quality evaluation method and system based on knowledge graph

Abstract

The invention discloses an anti-quantum service quality evaluation method based on a knowledge graph. The method comprises the following steps: initializing a knowledge network graph, weight parameters and public and private key pairs; collecting data of the service information; encrypting the collected data by fusing anti-quantum and homomorphic encryption technologies; constructing a coarse-grained knowledge graph based on the ciphertext service information book model; preliminarily evaluating and updating the knowledge graph; performing loop iteration to complete calculation of a service quality evaluation ciphertext; and the final service quality evaluation is obtained by using the private key for decryption. Through a lattice-based anti-quantum homomorphic cryptography mechanism, encryption processing of sensitive data is realized, information leakage and abuse are avoided, the method has the characteristics of privacy, security and the like, and compared with a traditional service quality evaluation model, the method can more effectively protect customer privacy and enhance user trust and satisfaction.

Description

Technical Field

[0001] The present invention relates to the field of information technology service quality, and in particular to a quantum-resistant service quality assessment method and system based on a knowledge graph. Background Art

[0002] With the development of economy and the improvement of people's living standards, the service industry has become an important part of the national economy. As an important management tool of the service industry, service quality assessment also plays an important role in improving the competitiveness of the service industry, improving service quality, and protecting consumer rights.

[0003] Existing service quality assessments often use methods based on expert experience, statistical modeling, and big data analysis, that is, based on practitioner experience, historical data trends, and big data, by selecting some service quality assessment support indicators and assigning weights to fill in the empirical model or analyzing and mining a large amount of service data to establish a service quality prediction model, and assess future service quality to help companies adjust service strategies and optimize resource allocation. However, the use of the above methods will not only lead to assessment errors due to the difficulty in adapting to new demand changes and uncertain factors, but also because the user's privacy is not protected accordingly during the assessment process, there is a risk of user privacy leakage. Summary of the invention

[0004] The purpose of the present invention is to provide a quantum-resistant service quality assessment method and system based on knowledge graph to solve the technical problems existing in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for evaluating quantum-resistant service quality based on a knowledge graph, including an authoritative platform and a service provider, characterized in that: a knowledge graph for evaluating service quality is included, specifically including the following steps:

[0007] S1. Initialize the knowledge graph, including the weight parameters of the service quality evaluation indicators therein, the authoritative platform selects a suitable lattice and modulus, generates a public key and private key pair based on the homomorphic encryption of the lattice cipher, and publishes the public key;

[0008] S2. Data for collecting service information, including relevant information of the service provider and consumer evaluation feedback information;

[0009] S3, using the data collected in steps S2 and S3 as plain text to encode the data into a polynomial, and encrypting it into ciphertext using the public key;

[0010] S4, semantically annotating the ciphertext, parsing out classes, instances, relationships and constraints, unifying the service information ontology model, and building the coarse-grained knowledge graph;

[0011] S5. According to the consumer's needs and the service information corresponding to the needs, a preliminary service quality evaluation is obtained through the knowledge graph matching calculation;

[0012] S6. Continuously integrating and updating the service information of the coarse-grained knowledge graph according to the preliminary service quality evaluation and the consumer evaluation feedback information to form a fine-grained knowledge graph;

[0013] S7, inferring the influence of different service quality evaluation indicators on the consumer evaluation feedback information through the fine-grained knowledge graph, and adjusting the weight parameter according to the influence;

[0014] S8, looping through steps S5 to S7 for several times, so that the knowledge graph becomes increasingly fine-grained, until the final encrypted result of the service quality evaluation is calculated;

[0015] S9. The service provider applies for a private key from the authoritative platform, decrypts the ciphertext of the service quality evaluation result, obtains the service quality evaluation result, and adjusts its service strategy.

[0016] Preferably, the public key and the private key both include a basis vector of the lattice and a random key vector.

[0017] Preferably, the service quality evaluation indicators include service reliability, service response speed, service security, product quality and service environment.

[0018] Preferably, the specific method of collecting consumer evaluation feedback information in step S2 includes: when the service provider receives a service request from a consumer, searching the consumer's consumption record in the latest knowledge graph, and when no consumption record of the consumer is detected, collecting the consumer's personal related information and initializing the consumer's evaluation feedback information; otherwise, collecting the consumer's evaluation feedback information and updating the corresponding knowledge graph content.

[0019] Preferably, the personal related information includes personal information and consumer needs, and the evaluation feedback information includes consumer grade, credit rating, satisfaction, and negative review rate.

[0020] Preferably, the relevant information of the service provider in step S2 includes multi-dimensional supporting data of the service itself and relevant data of the service provider. The multi-dimensional supporting data of the service itself includes service field, service quality, and cost-effectiveness. The relevant data of the service provider includes service scale, integrity record, and service method.

[0021] Preferably, the ciphertext of step S3 includes an encrypted polynomial and a noise polynomial.

[0022] Preferably, in step S4: the classes include service providers, consumers, service fields, service evaluations and service evaluation indicators; the instances include a service provider, a consumer, a service, a service evaluation and a service quality evaluation indicator; the relationships include service providers providing services, consumers selecting services, consumers evaluating services, service matching evaluation indicators and service evaluation indicator calculation evaluation scores; the constraint attributes include the geographical location of the service provider, the type of service, the quality of the service, the date of the service evaluation, the consumer's demand information, the consumer's level and the evaluation criteria of the service quality evaluation indicator.

[0023] Preferably, the looping steps in step S8 further include a collaborative evaluation mechanism sharing evaluation method, and the specific steps include the following:

[0024] H1. Access the latest knowledge graph. If there are multiple strongly related service quality evaluations, group and package the evaluations.

[0025] H2. If repeated evaluations are detected under similar service information, a collaborative evaluation mechanism is used to share the ciphertext of the service quality evaluation result. Otherwise, the ciphertext of the service quality evaluation result is obtained by iterating and looping through steps S5 to S7 for multiple times.

[0026] The present invention also provides a knowledge graph-based quantum-resistant service quality evaluation system for implementing any of the above methods, comprising the following modules:

[0027] Key management module: used to generate, manage and distribute public and private key pairs, and managed by the authoritative platform;

[0028] Data collection module: used to collect data of the service information;

[0029] Data encryption module: used to encrypt the data collected by the data collection module through homomorphic encryption based on the lattice cipher;

[0030] Parsing and building module: used to semantically annotate the ciphertext, parse out classes, instances, relationships and constraints, unify the service information ontology model, and build the coarse-grained knowledge graph;

[0031] Index management module: used to obtain corresponding service quality evaluation indicators through knowledge graph matching according to consumer needs and service information corresponding to the needs;

[0032] Service evaluation module: evaluates the service quality according to the service quality evaluation index obtained by the index management module and in combination with the consumer feedback information;

[0033] Reward feedback module: used to adjust the weight parameter according to the influence of different service quality evaluation indicators on the consumer evaluation feedback information;

[0034] Fusion and update module: based on the preliminary service quality evaluation and the consumer evaluation feedback information, the service information of the coarse-grained knowledge graph is continuously integrated and updated to form a fine-grained knowledge graph;

[0035] Data decryption module: used to send the ciphertext of the service quality evaluation result of the service evaluation module to the service provider, and use the private key to perform homomorphic decryption to obtain the final service quality evaluation result.

[0036] Beneficial effects of the present invention:

[0037] (1) The present invention proposes an efficient service quality assessment method driven by quantum-resistant homomorphic encryption and knowledge graph fusion, and uses a grouping and packaging mechanism for strongly correlated service requirements and policy information and a collaborative sharing mechanism for repeated assessment information to achieve high-performance interactive service quality assessment.

[0038] (2) Through the lattice-based quantum-resistant homomorphic cryptographic mechanism, encryption processing of sensitive data is achieved to avoid information leakage and abuse. This demonstrates that the present invention has the characteristics of privacy and security. It can more effectively protect customer privacy than traditional service quality evaluation models and enhance user trust and satisfaction.

[0039] (3) Apply deep learning to knowledge representation learning, explore the complex semantic associations between service requirements, service strategies, and service indicators, explore fine-grained service quality evaluation knowledge fusion reasoning technology, design a high-concurrency dynamic service quality evaluation system that supports multiple evaluation mechanisms, and realize adaptive service quality evaluation.

[0040] (4) The feedback mechanism is introduced to update the weight parameters in real time, ensuring the security and reliability of the knowledge graph network input data, and further ensuring the accuracy and timeliness of service quality evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flow chart of the steps of the service quality evaluation system of the present invention;

[0042] Figure 2This is a flow chart of the service quality evaluation method of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The embodiment of the present invention provides a quantum-resistant service quality assessment method and system based on knowledge graph. This embodiment takes the service quality assessment of chain restaurants as an example:

[0045] Assume that the third-party authoritative platform is a market supervision and management department, and the service provider is a chain brand with 10,000 restaurants. The total number of consumers who visit all restaurants under the chain brand in one day is 500,000. The service quality evaluation depends on the service attitude q m0 , service speed m1 , food quality m2 , food prices m3 and sanitary conditions m4 There are five evaluation indicators, where m is the number of iterations for updating the indicator weights, and m = 0, 1, 2, …, 99.

[0046] See also Figure 1 , an embodiment of the present invention provides a quantum-resistant service quality evaluation system based on knowledge graph, which specifically includes the following modules:

[0047] Key management module: used to generate, manage and distribute public and private key pairs, and managed by the market supervision and management department;

[0048] Data collection module: used to collect data on service information, including multi-dimensional supporting data of the service itself (such as the service field is a restaurant, service quality, cost performance, etc.), relevant data of the service provider (such as the service scale of 10,000, integrity record, service strategy, etc.) and relevant data of consumers (such as consumer level, consumer demand, credit, satisfaction, bad review rate, etc.);

[0049] Data encryption module: used to encrypt the data collected by the data collection module through homomorphic encryption based on the lattice cipher;

[0050] Parsing and construction module: used to semantically annotate the ciphertext, parse out classes, instances, relationships and constraints, unify the service information ontology model, and build the coarse-grained knowledge graph. In this embodiment, the classes include restaurant classR, consumer classC, service classS, service evaluation classE and service evaluation index classQ, where classQ has a subclass relationship with classS; the instance includes a restaurant r i , a consumer c j , a service j , a service evaluation j , some service quality evaluation indicator q m , i∈[0,10 4 -1],j∈[0,5*10 5 -1]; the relationship includes the restaurant providing service rService, the consumer selecting service cService, the consumer evaluating the service cEvaluation, the service selection evaluation index sQuota and the service evaluation index calculating the evaluation score qValue; the constraint attributes include the restaurant's geographical location rLocation, the service type sType, the service quality sQuality, the service evaluation date eTime, the consumer's demand information cRequirement, the consumer's level cLevel and the evaluation standard qStandard of the service quality evaluation index = (service attitude q m0 , service speed q m1 , food quality q m2 , food prices m3 , health status m4 );

[0051] Index management module: used to obtain corresponding service quality evaluation indicators through knowledge graph matching according to consumer needs and the service information corresponding to the needs. The evaluation criteria of service quality evaluation indicators are supported by multiple evaluation indicators, including service attitude, m0 , service speed m1 , food quality m2 , food prices m3 and sanitary conditions m4 Etc. Based on the fine-grained knowledge graph generated by the fusion update module, the knowledge representation learning based on deep learning is used to mine the complex semantic associations between service requirements, service strategies and service indicators. When evaluating service quality, the service quality evaluation indicators are quickly matched and adjusted according to the service field and consumer demand information.

[0052] Service evaluation module: evaluates the service quality according to the service quality evaluation index obtained by the index management module and the consumer feedback information. In this embodiment, the service evaluation module dynamically loads the service quality evaluation index, groups and packages the strongly related service requirements and policy information, and uses lightweight security evaluation technology to achieve quality evaluation of multiple service information at a time. At the same time, considering that there may be repeated evaluations of the same service requirements and policy information in similar service fields, a collaborative evaluation mechanism is used to share the evaluation results, thereby improving the service quality evaluation efficiency and concurrency capabilities.

[0053] Reward feedback module: used to adjust the weight parameters according to the influence of different service quality evaluation indicators on the consumer evaluation feedback information, so as to promote the security and accuracy of service quality evaluation and the fine-grained knowledge graph network, and improve the generalization ability of the evaluation model.

[0054] Fusion update module: Based on the preliminary service quality evaluation and the consumer evaluation feedback information, the service information of the coarse-grained knowledge graph is integrated and updated to form a fine-grained knowledge graph. It optimizes the indicator weight parameters of the reward feedback module and combines the service quality evaluation results generated by the service evaluation module, records the repeated evaluation data of the same needs and service information in similar service fields, and generates a finer-grained knowledge graph.

[0055] Data decryption module: It is used for restaurants to apply for private keys from the market supervision and management department, and when authorized, send the ciphertext of the service quality evaluation results of the service evaluation module to the service provider, and use the private key for homomorphic decryption to obtain the final service quality evaluation results, so that the restaurant can adjust its catering service mode.

[0056] Based on the above service quality evaluation system, see Figure 2 The embodiment of the present invention provides a method for evaluating quantum-resistant service quality based on a knowledge graph, including a knowledge graph for evaluating service quality, and specifically including the following steps:

[0057] Step S1, initializing the knowledge graph, including the weight parameters of the restaurant service quality evaluation indicators, where the service quality evaluation indicators include service attitude q m0 , service speed m1 , food quality m2 , food prices m3 and sanitary conditions m4 The control key management module of the authoritative platform (market supervision and management department) selects a suitable lattice and modulus, generates a homomorphically encrypted public key and private key pair based on the lattice cipher, and publishes the public key. Both the public key and private key include the basis vector of the lattice and a random key vector.

[0058] Step S2: collect service information data through the data collection module, including relevant information of the service provider and consumer evaluation feedback information. The relevant information of the service provider includes multi-dimensional supporting data of the service itself and relevant data of the service provider. The multi-dimensional supporting data of the service itself includes service field (catering), service quality (service speed, etc. ... m1 ), cost performance (food quality m2 With food prices m3 ), the relevant data of the service provider include service scale (10,000), integrity record, and service method.

[0059] The specific method of collecting consumer evaluation feedback information in step S2 includes: when the service provider receives a service request from a consumer, searching the consumer's consumption record in the latest knowledge graph, and when no consumption record of the consumer is detected, collecting the consumer's personal related information (including personal information, consumer needs), and initializing the consumer's evaluation feedback information (consumer level is primary, credit is general, satisfaction is 0, and negative review rate is 0), otherwise collecting the consumer's evaluation feedback information and updating the corresponding knowledge graph content.

[0060] Step S3, integrating quantum resistance and homomorphic encryption technology to encrypt the collected data, using the data encryption module to encode the data collected in steps S2 and S3 as plain text into a polynomial, and encrypting it into ciphertext using the public key:

[0061] Enc(m)=(c1, c2)=(m·pk+e1, r·pk+e2);

[0062] Where m is the plaintext, pk is the public key vector, r is a random polynomial, e1 and e2 are noise polynomials, and the ciphertext includes the encrypted polynomial and the noise polynomial.

[0063] Step S4: semantically annotate the ciphertext through the above-mentioned parsing construction module, parse out the classes, instances, relationships and constraints (refer to the above-mentioned parsing construction module for details, which will not be repeated here), unify the service information ontology model, and build the coarse-grained knowledge graph;

[0064] Step S5: According to the consumer's needs and the service information corresponding to the needs, the indicator management module matches the knowledge graph, and then calculates the preliminary service quality evaluation through the service evaluation module in combination with the consumer's corresponding satisfaction.

[0065] Step S6: The fusion and update module continuously fuses and updates the service information of the coarse-grained knowledge graph according to the preliminary service quality evaluation and the consumer evaluation feedback information to form a fine-grained knowledge graph.

[0066] Step S7: The reward feedback module infers the influence of the different service quality evaluation indicators on the consumer evaluation feedback information through the fine-grained knowledge graph, and adjusts the weight parameter according to the influence.

[0067] Step S8: The service quality evaluation method is connected to the latest knowledge graph. If there are multiple strongly related service quality evaluations, the evaluations are grouped and packaged.

[0068] Step S9: If repeated evaluations are detected under similar service information, the collaborative evaluation mechanism is used to share the ciphertext of the service quality evaluation result, otherwise the ciphertext of the service quality evaluation result is obtained by iterating and looping through steps S5 to S7 multiple times. In steps S8 to S9, lightweight security evaluation technology is used to implement quality evaluation of multiple service information at a time. At the same time, considering that there may be repeated evaluations of the same service requirements and policy information under similar service fields, a collaborative evaluation mechanism is used to share the evaluation results, thereby improving the efficiency and concurrency of service quality evaluation.

[0069] Step S10, looping through steps S5 to S9, with the number of iterations m=100 in this embodiment, so that the knowledge graph becomes more and more fine-grained, until the final result ciphertext of the service quality evaluation is calculated;

[0070] Step S11, the restaurant applies for a private key to the market supervision and management department, and provides the corresponding identity certificate and authorization credentials. After the market supervision and management department completes the identity authentication and authorization approval, it provides the corresponding private key according to the authority and needs. The data decryption module provides the restaurant with the service evaluation result ciphertext, and uses the homomorphically encrypted decryption private key to decrypt the ciphertext service quality evaluation result to obtain the final evaluation result, and gives suggestions to adjust the catering service mode.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent substitution or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A quantum-resistant service quality assessment method based on knowledge graph, including an authoritative platform and a service provider, characterized in that: Including a knowledge graph for service quality evaluation, specifically including the following steps: S1. Initialize the knowledge graph, including the weight parameters of the service quality evaluation indicators therein, the authoritative platform selects a suitable lattice and modulus, generates a public key and private key pair based on the homomorphic encryption of the lattice cipher, and publishes the public key; S2. Data for collecting service information, including relevant information of the service provider and consumer evaluation feedback information; S3, using the data collected in steps S2 and S3 as plain text to encode the data into a polynomial, and encrypting it into ciphertext using the public key; S4, semantically annotating the ciphertext, parsing out classes, instances, relationships and constraints, unifying the service information ontology model, and building the coarse-grained knowledge graph; S5. According to the consumer's needs and the service information corresponding to the needs, a preliminary service quality evaluation is obtained through the knowledge graph matching calculation; S6. Continuously integrating and updating the service information of the coarse-grained knowledge graph according to the preliminary service quality evaluation and the consumer evaluation feedback information to form a fine-grained knowledge graph; S7, inferring the influence of different service quality evaluation indicators on the consumer evaluation feedback information through the fine-grained knowledge graph, and adjusting the weight parameter according to the influence; S8, looping through steps S5 to S7 for several times, so that the knowledge graph becomes increasingly fine-grained, until the final encrypted result of the service quality evaluation is calculated; S9. The service provider applies for a private key from the authoritative platform, decrypts the ciphertext of the service quality evaluation result, obtains the service quality evaluation result, and adjusts its service strategy.

2. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: The public key and the private key both include the basis vector of the lattice and a random key vector.

3. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: The service quality evaluation indicators include service reliability, service response speed, service security, product quality and service environment.

4. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: The specific method of collecting consumer evaluation feedback information in step S2 includes: when the service provider receives a service request from a consumer, searching the consumer's consumption record in the latest knowledge graph. If no consumption record of the consumer is detected, the consumer's personal related information is collected and the consumer's evaluation feedback information is initialized. Otherwise, the consumer's evaluation feedback information is collected and the corresponding knowledge graph content is updated.

5. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 4, characterized in that: The personal related information includes personal information and consumer needs, and the evaluation feedback information includes consumer level, credit rating, satisfaction, and negative review rate.

6. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: The relevant information of the service provider in step S2 includes multi-dimensional supporting data of the service itself and relevant data of the service provider. The multi-dimensional supporting data of the service itself includes service field, service quality, and cost performance. The relevant data of the service provider includes service scale, integrity record, and service method.

7. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: The ciphertext of step S3 includes the encrypted polynomial and the noise polynomial.

8. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: In step S4: the classes include service providers, consumers, service areas, service evaluations and service evaluation indicators; the instances include a service provider, a consumer, a service, a service evaluation and a service quality evaluation indicator; the relationships include service providers providing services, consumers selecting services, consumers evaluating services, service matching evaluation indicators and service evaluation indicator calculation evaluation scores; the constraint attributes include the geographical location of the service provider, the type of service, the quality of the service, the date of the service evaluation, the consumer's demand information, the consumer's level and the evaluation criteria of the service quality evaluation indicator.

9. A method for evaluating quantum-resistant service quality based on knowledge graph as claimed in claim 1, characterized in that: The steps in the loop in step S8 also include a collaborative evaluation mechanism sharing evaluation method, and the specific steps include the following: H1. Access the latest knowledge graph. If there are multiple strongly related service quality evaluations, group and package the evaluations. H2. If repeated evaluations are detected under similar service information, a collaborative evaluation mechanism is used to share the ciphertext of the service quality evaluation result. Otherwise, the ciphertext of the service quality evaluation result is obtained by iterating and looping through steps S5 to S7 for multiple times.

10. A quantum-resistant service quality evaluation system based on knowledge graph, characterized by: For implementing any one of claims 1 to 9, comprising the following modules: Key management module: used to generate, manage and distribute public and private key pairs, and managed by the authoritative platform; Data collection module: used to collect data of the service information; Data encryption module: used to encrypt the data collected by the data collection module through homomorphic encryption based on the lattice cipher; Parsing and building module: used to semantically annotate the ciphertext, parse out classes, instances, relationships and constraints, unify the service information ontology model, and build the coarse-grained knowledge graph; Index management module: used to obtain corresponding service quality evaluation indicators through knowledge graph matching according to consumer needs and service information corresponding to the needs; Service evaluation module: evaluates the service quality according to the service quality evaluation index obtained by the index management module and in combination with the consumer feedback information; Reward feedback module: used to adjust the weight parameter according to the influence of different service quality evaluation indicators on the consumer evaluation feedback information; Fusion and update module: based on the preliminary service quality evaluation and the consumer evaluation feedback information, the service information of the coarse-grained knowledge graph is continuously integrated and updated to form a fine-grained knowledge graph; Data decryption module: used to send the ciphertext of the service quality evaluation result of the service evaluation module to the service provider, and use the private key to perform homomorphic decryption to obtain the final service quality evaluation result.

Images