Pension service quality intelligent evaluation method and system based on multi-dimensional data fusion

Abstract

The invention provides an old-age service quality intelligent evaluation method and system based on multi-dimensional data fusion, and relates to the technical field of old-age service data analysis. The method comprises the following steps: collecting historical old-age service record data of a plurality of old people, and generating a global service preference matrix and a local service fluctuation matrix of each old people; constructing a service association map of each elderly, calculating interaction influence parameters between any two elderly care services, and generating a historical interaction influence matrix; carrying out service track coding on the historical old-age service record data, and constructing a micro-service mode set of each old-age; and according to the micro-service mode set, calculating a plurality of individual behavior linkage weights of each elderly, calculating global preference parameters of the elderly about each endowment service, performing endowment service quality evaluation, and generating an endowment service quality evaluation result of each elderly. According to the invention, precise old-age care service quality evaluation based on individual demand characteristics is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of elderly care service data analysis, and particularly to an intelligent evaluation method and system for the quality of elderly care services based on multi-dimensional data fusion. Background Art

[0002] The intelligent evaluation of the quality of elderly care services is an important part of the construction of the elderly care service system. Some elderly care service quality evaluation methods use a unified standard or an expert scoring-based method to comprehensively score different types of elderly care service items such as medical care, social entertainment, and psychological counseling. However, in the actual process of elderly care services, due to the differences in individual needs and behavioral characteristics of different elderly people, there are significant differences in the focus, sensitivity, and actual experience of the same service item. For example, some elderly people may be more concerned about medical care services, while others may focus more on social activities or psychological support. And the elderly care service needs do not exist in isolation. There are complex complementary, substitution, and linkage relationships between different service items. For example, an increase in medical care needs is often accompanied by a decrease in social activities, and the application of intelligent assistive devices may reduce the need for manual care.

[0003] If a single scoring weight is obtained by ignoring individual differences and the same set of service evaluation indicators and weights are applied to all elderly people, it is easy to lack in-depth analysis of the individual preferences and behavioral differences of the elderly people, and it is impossible to achieve a truly elderly-centered personalized service quality evaluation, which may lead to a low reference value of the evaluation results of the quality of elderly care services. Summary of the Invention

[0004] To solve the above technical problems, the present invention proposes an intelligent evaluation method and system for the quality of elderly care services based on multi-dimensional data fusion. Through strategies such as service association network modeling and service behavior chain analysis, personalized tendency weights of each elderly person regarding different elderly care service items are dynamically generated, which can better achieve accurate evaluation of the quality of elderly care services based on individual demand characteristics.

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

[0006] As the first aspect of the present invention, there is provided an intelligent evaluation method for the quality of elderly care services based on multi-dimensional data fusion, including:

[0007] Collect historical elderly care service record data of multiple elderly people, construct the usage feature sequence of each elderly person regarding each elderly care service and perform window decomposition, and generate the global service preference matrix and local service fluctuation matrix of each elderly person;

[0008] Analyze the interactive effects of multiple elderly care services based on multiple usage feature sequences, construct a service association graph for each elderly person, calculate the interactive effect parameters between any two elderly care services based on multiple service association graphs, and generate a historical interactive effect matrix for multiple elderly care services;

[0009] Encode the service trajectories of historical elderly care service record data, construct an elderly care service link for each elderly person, perform service chain segmentation on each elderly care service link to generate multiple candidate service chains, and generate a microservice pattern set containing multiple target service chains for each elderly person;

[0010] Perform interactive processing on the microservice pattern set of each elderly person through the historical interactive effect matrix, and calculate the individual behavior linkage weights corresponding to multiple target service chains of each elderly person;

[0011] Generate the global preference parameters of the elderly person for each elderly care service according to the individual behavior linkage weights, the global service preference matrix, and the local service fluctuation matrix, and perform the quality evaluation of the elderly care service according to the global preference parameters to generate the quality evaluation results of the elderly care service for each elderly person.

[0012] Preferably, generating a microservice pattern set containing multiple target service chains for each elderly person includes:

[0013] Determine multiple service behaviors of the elderly person in the historical elderly care service record data, and construct an elderly care service link of the elderly person for multiple elderly care services based on the time stamps of the service behaviors and the corresponding service items in chronological order;

[0014] Perform service chain segmentation on the elderly care service link based on a preset sequence length to generate multiple candidate service chains, count the distribution frequencies of each candidate service chain under the elderly care service link, construct a pattern distribution sequence of each candidate service chain according to the multiple time stamps of the candidate service chains, perform behavioral regularity analysis on each candidate service chain according to the pattern distribution sequence, calculate the behavioral regularity index of each candidate service chain, and determine multiple target service chains from the multiple candidate service chains according to the behavioral regularity index, and construct a microservice pattern set for each elderly person.

[0015] Preferably, performing interactive processing on the microservice pattern set of each elderly person through the historical interactive effect matrix, and calculating the individual behavior linkage weights corresponding to multiple target service chains of each elderly person includes:

[0016] Determine the elderly care service combination corresponding to each target service chain according to the start and end points of the target service chain, extract the interactive effect parameter of each target service chain from the historical interactive effect matrix according to the elderly care service combination, and calculate the individual behavior linkage weight of the target service chain based on the distribution frequency and the interactive effect parameter of the target service chain.

[0017] Preferably, according to the individual behavior linkage weight, the global service preference matrix, and the local service fluctuation matrix, generate the global preference parameters of the elderly for each elderly care service, including:

[0018] Modify the historical interaction influence matrix according to the individual behavior linkage weights of multiple target service chains to generate a dynamic interaction influence matrix of the historical interaction influence matrix. Perform interaction processing on the micro-service mode set of each elderly according to the dynamic interaction influence matrix, and calculate the target behavior linkage weights corresponding to multiple target service chains of each elderly.

[0019] Determine multiple target service chains associated with each elderly care service for the elderly. Generate a chain tendency factor of the elderly for each elderly care service according to the target behavior linkage weights of the target service chains. Calculate the long-term and short-term preference tendency factors of each elderly care service for the elderly according to the global service preference matrix and the local service fluctuation matrix. Generate the global preference parameters of the elderly for each elderly care service according to the chain tendency factor and the long-term and short-term preference tendency factors.

[0020] Preferably, for the historical interaction influence matrix, it is characterized in that

[0021] Calculate the initial correlation coefficient between the usage feature sequences of any two elderly care services. Construct a service association graph of the elderly according to the initial correlation coefficient and the preset correlation threshold. Determine the association degree of each elderly for each elderly care service according to the service association graph. Modify the initial correlation coefficient between any two elderly care services for the elderly based on the association degree of the elderly care service to generate the target correlation coefficient between any two elderly care services, and construct the local interaction matrix of each elderly.

[0022] Fuse multiple local interaction matrices, including determining multiple target correlation coefficients corresponding to any two elderly care services according to multiple local interaction matrices, calculating the interaction influence parameters between any two elderly care services, and generating a historical interaction influence matrix according to multiple interaction influence parameters.

[0023] Preferably, perform behavioral regularity analysis on each candidate service chain according to the mode distribution sequence, and calculate the behavioral regularity index of each candidate service chain, including:

[0024] Statistically analyze the time interval between any two adjacent candidate service chains in the mode distribution sequence, calculate the standard deviation of multiple time intervals, and use the ratio between the distribution frequency of the candidate service chain and the standard deviation of multiple time intervals as the behavioral regularity index.

[0025] As a second aspect of the present invention, there is provided an intelligent evaluation system for the quality of elderly care services based on multi-dimensional data fusion, which is used to implement the above-mentioned intelligent evaluation method for the quality of elderly care services based on multi-dimensional data fusion, including:

[0026] A service usage feature analysis module, which is used to collect historical elderly care service record data of multiple elderly people, construct a usage feature sequence for each elderly person regarding each elderly care service and perform window decomposition, and generate a global service preference matrix and a local service fluctuation matrix for each elderly person;

[0027] A service interaction impact analysis module, which is used to perform interaction impact analysis on multiple elderly care services according to multiple usage feature sequences, construct a service association graph for each elderly person, calculate the interaction impact parameters between any two elderly care services according to multiple service association graphs, and generate a historical interaction impact matrix regarding multiple elderly care services;

[0028] A microservice link analysis module, which is used to perform service trajectory encoding on historical elderly care service record data, construct an elderly care service link for each elderly person, perform service chain segmentation on each elderly care service link to generate multiple candidate service chains, and generate a microservice mode set containing multiple target service chains for each elderly person;

[0029] An individual behavior linkage analysis module, which is used to perform interaction processing on the microservice mode set of each elderly person respectively through the historical interaction impact matrix, and calculate the individual behavior linkage weights corresponding to multiple target service chains of each elderly person;

[0030] An elderly care service quality evaluation and management module, which is used to generate global preference parameters for each elderly person regarding each elderly care service according to the individual behavior linkage weights, the global service preference matrix and the local service fluctuation matrix, perform elderly care service quality evaluation according to the global preference parameters, and generate an elderly care service quality evaluation result for each elderly person.

[0031] The present invention has the following beneficial effects:

[0032] By dynamically analyzing the historical elderly care service record data, this invention explores the dynamic change relationship between the long-term and short-term service needs of the elderly regarding elderly care services, generates the global service preference matrix and local service fluctuation matrix for each elderly person, analyzes the interactive correlation characteristics between different elderly care services, obtains the historical interactive influence matrix for multiple elderly care services, and further conducts individual behavior chain analysis for each elderly person. Through the extraction of microservice mode sets and the adjustment of the correlation relationship between services in combination with the dynamic interactive influence matrix, it dynamically identifies the personalized tendencies of each elderly person towards different elderly care service items, generates personalized service weights that conform to the actual demand characteristics of the elderly, realizes the personalized and accurate evaluation of the quality of elderly care services, makes the evaluation results of the quality of elderly care services more in line with the actual needs and feelings of the elderly, and improves the intelligent level of the evaluation of the quality of elderly care services. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic flowchart of a method for intelligent evaluation of the quality of elderly care services based on multi-dimensional data fusion provided by an embodiment of the present invention.

[0034] Figure 2 It is a schematic structural diagram of a system for intelligent evaluation of the quality of elderly care services based on multi-dimensional data fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to enable those skilled in the art of this technology to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0036] Please refer to Figure 1 , a method for intelligent evaluation of the quality of elderly care services based on multi-dimensional data fusion provided by an embodiment of the present invention includes the following steps:

[0037] Step S1: Collect the historical elderly care service record data of multiple elderly people, construct the usage feature sequence of each elderly person for each elderly care service and perform window decomposition, and generate the global service preference matrix and local service fluctuation matrix for each elderly person.

[0038] Specifically, collect the elderly care service record data of multiple elderly people over a period of historical time. The collected data includes, but is not limited to, the usage time, usage frequency, service type, etc. of various elderly care services. For example, the relevant service record data of the elderly for items such as medical care, social entertainment, and psychological counseling. For each elderly person, for all elderly care service items, construct the usage feature sequence corresponding to each elderly care service in chronological order, and use the window decomposition technology to divide each usage feature sequence into multiple time windows, and extract the mean feature of the usage frequency of each elderly care service within each window, which represents a long-term stable preference of the elderly within a specific period. At the same time, extract the fluctuation feature of each elderly care service within each window, such as the standard deviation of the usage frequency, etc., which represents the short-term fluctuation of the elderly, such as the short-term fluctuation of the elderly care service selection tendency caused by special events such as illness. Thus, a global service preference matrix is generated to characterize the overall service preference and long-term dependence relationship of the elderly, and a local service fluctuation matrix is constructed to characterize the volatility and sensitivity of the short-term service demand of the elderly.

[0039] Step S2: Conduct an interactive influence analysis on multiple elderly care services based on multiple usage feature sequences, construct a service association graph for each elderly person, calculate the interactive influence parameters between any two elderly care services based on multiple service association graphs, and generate a historical interactive influence matrix for multiple elderly care services.

[0040] Specifically, for the multiple usage feature sequences of each elderly person, conduct an in-depth analysis of the linkage relationship between elderly care service items, identify the interactive influence features of different service items in usage behavior, and construct a service association graph for each elderly person by analyzing the correlation between the frequencies of the elderly using different elderly care services. And conduct a global analysis based on multiple encouraging association graphs, evaluate the overall interactive influence features between different elderly care services from an overall perspective, calculate the interactive influence parameters between any two elderly care services, and finally construct a historical interactive influence matrix corresponding to the overall multiple elderly care services.

[0041] In an alternative embodiment, for the construction of the historical interaction influence matrix, first calculate the initial correlation coefficient between the usage feature sequences of any two elderly care services. In this embodiment, mutual information is taken as an example, and the mutual information between the two sequences is used as the initial correlation parameter. According to the initial correlation coefficient and the preset correlation threshold, the service association graph of the elderly is constructed. The service association graph takes elderly care services as nodes and the interaction correlation between elderly care services as edges, and the edge weight represents the correlation interaction intensity between service items, that is, the initial correlation coefficient. To improve the representativeness of the service association graph, only when the initial correlation parameter of two services is higher than the preset correlation threshold, it is considered that there is an effective association between the two services, so as to form a dynamic service relationship network reflecting the individual service behavior characteristics of the elderly. And determine the association degree of each elderly person with respect to each elderly care service according to the service association graph, specifically the total number of graph edges of each elderly care service. The stronger the connection between a certain service and multiple other services, that is, the higher the number of retained edges, the higher the overall importance of the service.

[0042] At the same time, in order to avoid the problem that the initial correlation coefficient is too high or too low due to some high-frequency but low-dependency services, the initial correlation coefficient between any two elderly care services of the elderly is corrected by combining the association degree of the elderly care services. For example, after normalizing the overall association degree, the sum of the association degrees corresponding to the two elderly care services is used as the correction weight. The initial correlation coefficient of the combination with a high association degree is retained as much as possible, and the coefficient of the combination with a low association degree is weakened, so as to generate the target correlation coefficient between any two elderly care services, and construct the local interaction matrix of each elderly person.

[0043] Fuse the local interaction matrices corresponding to multiple elderly people. For the multiple target correlation coefficients corresponding to any two elderly care services, in this embodiment, they are accumulated and averaged and then used as the interaction influence parameter between the two elderly care services to realize the fusion of multiple local interaction matrices, and construct a historical interaction influence matrix representing the overall association characteristics of different elderly care services, reflecting the comprehensive interaction relationship between all elderly care services under the actual behavior background of the elderly.

[0044] Step S3: Perform service trajectory coding on the historical elderly care service record data, construct the elderly care service link of each elderly person, perform service chain segmentation on each elderly care service link to generate multiple candidate service chains, and generate a micro-service mode set containing multiple target service chains for each elderly person.

[0045] Specifically, for multiple groups of historical elderly care service record data, the usage trajectories of elderly care services for each elderly person are encoded to form a complete elderly care service link. This link reflects the process of the elderly person sequentially using different elderly care service items in a time series manner, revealing the internal logic of the elderly person's service combination and behavior path. Based on the elderly care service link, chain segmentation is performed, and the complete behavior link is segmented to generate multiple candidate service chains. Then, representative ones are extracted from them, such as candidate service chains with a higher frequency of occurrence and stronger stability in the elderly care service link, to form the micro-service mode set of each elderly person. This set can reflect the combined needs and behavior habits of the elderly person for multiple services, and reflect the true usage relationship and behavior chain characteristics between services.

[0046] In an alternative implementation, for the construction process of the micro-service mode set containing multiple target service chains for each elderly person, multiple service behaviors of the elderly person in the historical elderly care service record data are determined. Based on the timestamps of the service behaviors and the corresponding service items, an elderly care service link for the elderly person regarding multiple elderly care services is constructed in chronological order, reflecting the true path and process of the elderly person's use of elderly care services.

[0047] For the complete elderly care service link, service chain segmentation is performed on the elderly care service link based on a preset sequence length to generate multiple candidate service chains. Among them, the preset sequence length can be, for example, a combination of 2 items, 3 items, or 4 items of consecutive services. In this embodiment, taking the combination of two consecutive elderly care services as an example, a sliding window method is adopted to sequentially extract multiple micro-chains of a fixed length to obtain multiple candidate service chains for each elderly person. Further, the distribution frequency of each candidate service chain under the elderly care service link is statistically analyzed, and a pattern distribution sequence of each candidate service chain is constructed based on the multiple timestamps of the candidate service chain. Then, behavioral regularity analysis is performed on each candidate service chain according to the pattern distribution sequence, and a behavioral regularity index of each candidate service chain is calculated. For example, the time interval between any two adjacent candidate service chains in the pattern distribution sequence is statistically analyzed, the standard deviation of multiple time intervals is calculated, and the ratio between the distribution frequency of the candidate service chain and the standard deviation of multiple time intervals is used as the behavioral regularity index to comprehensively reflect its behavioral stability and regularity. Finally, based on the behavioral regularity index, multiple representative target service chains are determined from multiple candidate service chains, and the micro-service mode set of each elderly person is constructed to comprehensively and meticulously reflect the stable and preferred service link existing in the actual service behavior of the elderly person.

[0048] Step S4: Perform interaction processing on the micro-service mode set of each elderly person through the historical interaction influence matrix, and calculate the individual behavior linkage weights corresponding to multiple target service chains of each elderly person.

[0049] Specifically, based on the previously extracted microservice mode set and combined with the historical interaction influence matrix, for each microservice chain of each elderly person, dynamic coupling of interaction weights is carried out, comprehensively considering the frequency of the individual behavior chain and the service interaction relationship at the historical global level, to form the individual behavior linkage weight.

[0050] In this embodiment, the elderly care service combination corresponding to each target service chain is determined according to the starting point and the ending point of the target service chain, that is, the elderly care services corresponding to the starting point and the ending point of the target service chain respectively. The interaction influence parameter corresponding to this combination is extracted from the historical interaction influence matrix as the interaction influence parameter of each target service chain. Finally, the individual behavior linkage weight of the target service chain is calculated based on the distribution frequency and the interaction influence parameter of the target service chain. That is, for each target service chain of each elderly person, according to the interaction influence degree between the service items in this chain, comprehensively considering the relative importance and frequency of this service chain in the individual behavior of the elderly person, for example, the fusion weight is determined according to the distribution frequency, and the interaction influence parameter is corrected and fused to obtain the individual behavior linkage weight of the target service chain. Through this way of dynamic fusion of the behavior chain and service interaction, it can better represent the service linkage mode existing in the actual life of the elderly, avoid the interference of single service behavior or occasional behavior on the overall model, and thus more accurately capture the real service usage logic of the elderly.

[0051] Step S5: Generate the global preference parameter of each elderly person for each elderly care service according to the individual behavior linkage weight, the global service preference matrix and the local service fluctuation matrix, and conduct the quality evaluation of the elderly care service according to the global preference parameter to generate the quality evaluation result of the elderly care service for each elderly person.

[0052] Specifically, according to the individual behavior linkage weight obtained in the previous step, as well as the global service preference matrix and the local service fluctuation matrix, multi-dimensional fusion analysis is carried out on the personalized tendency of each elderly person for each elderly care service to generate the global preference parameter that comprehensively reflects the personal service needs, so as to comprehensively reflect multi-dimensional factors such as the long-term dependence, recent demand fluctuation, and service linkage of the elderly, and effectively depict the real sensitivity and preference weight of the elderly for different elderly care service items.

[0053] Finally, according to the global preference parameter, combined with the preset quality scoring standard of each elderly care service, for example, the scores of different elderly care services obtained through expert evaluation, the service quality evaluation based on personalized weights is carried out for all elderly care services to generate the personalized elderly care service quality evaluation result for each elderly person. Compared with some evaluation methods with single weights, this method can dynamically adjust the evaluation weights of each service according to the actual service needs of each elderly person, so as to realize more accurate intelligent evaluation of the elderly care service quality and improve the personalized and dynamic level of the elderly care service management.

[0054] In an alternative embodiment, global preference parameters of the elderly for each elderly care service are generated according to individual behavior linkage weights, a global service preference matrix, and a local service fluctuation matrix, including:

[0055] The historical interaction influence matrix is corrected according to the individual behavior linkage weights of multiple target service chains to generate a dynamic interaction influence matrix of the historical interaction influence matrix. In this process, for any elderly care service combination in the historical interaction influence matrix, based on the interaction influence parameter of this elderly care service combination, considering multiple target service chains involving this elderly care service combination, the average value of the individual behavior linkage weights of the multiple target service chains is used as the correction weight, and the interaction influence parameter of the elderly care service combination is dynamically adjusted to generate a more individualized dynamic interaction influence matrix to reflect the influence of individual behavior characteristics on the interaction relationship between services.

[0056] The microservice mode sets of each elderly person are interactively processed according to the dynamic interaction influence matrix, and the target behavior linkage weights corresponding to the multiple target service chains of each elderly person are calculated. That is, in the same way as the calculation method of the individual behavior linkage weight, the dynamically adjusted dynamic interaction influence matrix replaces the original historical interaction influence matrix based on historical overall rules.

[0057] For multiple target service chains associated with each elderly care service of the elderly, specifically, the target service chains with this elderly care service as the starting point or the end point are considered to be relevant. According to the target behavior linkage weights corresponding to the relevant multiple target service chains, the sum is generated as the chain tendency factor of the elderly for each elderly care service, reflecting the actual behavior importance of this elderly care service as the core node of the chain, and can accurately reflect the comprehensive behavior performance of the dependence and linkage characteristics between a certain elderly care service and other services in the actual life of the elderly.

[0058] Furthermore, the long- and short-term preference tendency factors for each elderly care service are calculated based on the global service preference matrix and the local service fluctuation matrix. In this process, considering that the elderly's preference tendencies for elderly care services may change in different periods, the long- and short-term preference tendency factors within a specific period can be extracted from the global service preference matrix and the local service fluctuation matrix. For example, the ratio of the mean value to the standard deviation of the usage frequency of a certain elderly care service within a specific time is used as the long- and short-term preference tendency factor for that period. And based on the chain tendency factor and the long- and short-term preference tendency factor, after weighted fusion, the global preference parameter of the elderly for this elderly care service within a specific period is obtained. The weights for weighted fusion can be set according to the overall proportion of the chain tendency factor and the long- and short-term preference tendency factor, or the weights of the chain tendency factor and the long- and short-term preference tendency factor can be quantified based on expert experience, and reasonably set by measuring the importance of elderly care services between combined behaviors and long-term dependence behaviors.

[0059] It should be noted that the needs of each elderly person for different elderly care services may have individual differences to varying degrees. For some elderly care service quality evaluation methods that use the same evaluation weight for each elderly care service for different elderly people, it is difficult to reflect individual differences, and it may only have a high degree of matching for some elderly groups. For some groups with certain differences between individual needs and group needs, using this method cannot well evaluate the matching degree of the current elderly care services for them, and it is easy to lead to unrepresentative final elderly care service evaluation results in the case of large individual preference differences. The global preference parameter constructed by the present invention comprehensively considers the service dependence and linkage relationship of the elderly in the behavior chain, as well as the long-term rigid demand and short-term temporary demand for services, and can dynamically and individually depict the actual attention of different elderly people to each elderly care service, reflecting the people-centered service concept, and ensuring that the evaluation results are closer to the real feelings of the elderly. For example, for the elderly who are sensitive to medical care, the scoring weight of medical services is high, and for the elderly who are not interested in social activities, the social weight is lower. The final comprehensive evaluation result is more in line with the actual experience, providing a good data basis for constructing a personalized elderly care service system.

[0060] Please refer to Figure 2 , an intelligent evaluation system for the quality of elderly care services based on multi-dimensional data fusion provided by an embodiment of the present invention, is specifically based on the concept of the above-mentioned intelligent evaluation method for the quality of elderly care services based on multi-dimensional data fusion, and includes:

[0061] A service usage feature analysis module, configured to collect historical elderly care service record data of multiple elderly people, construct a usage feature sequence for each elderly person for each elderly care service and perform window decomposition, and generate a global service preference matrix and a local service fluctuation matrix for each elderly person;

[0062] A service interaction impact analysis module, which is used to conduct interaction impact analysis on multiple elderly care services according to multiple usage feature sequences, construct a service association graph for each elderly person, calculate the interaction impact parameters between any two elderly care services based on multiple service association graphs, and generate a historical interaction impact matrix for multiple elderly care services;

[0063] A microservice link analysis module, which is used to perform service trajectory encoding on historical elderly care service record data, construct an elderly care service link for each elderly person, perform service chain segmentation on each elderly care service link to generate multiple candidate service chains, and generate a microservice pattern set containing multiple target service chains for each elderly person;

[0064] An individual behavior linkage analysis module, which is used to perform interaction processing on the microservice pattern set of each elderly person respectively through the historical interaction impact matrix, and calculate the individual behavior linkage weights corresponding to multiple target service chains of each elderly person;

[0065] An elderly care service quality evaluation and management module, which is used to generate global preference parameters of the elderly person for each elderly care service according to the individual behavior linkage weights, the global service preference matrix, and the local service fluctuation matrix, conduct elderly care service quality evaluation according to the global preference parameters, and generate the elderly care service quality evaluation results of each elderly person.

[0066] The above are only specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The parts not described in detail in this specification belong to the prior art well-known to those skilled in the art.

Claims

1. An intelligent evaluation method for elderly care service quality based on multi-dimensional data fusion, characterized in that: include: Collect historical elderly care service record data of multiple elderly people, construct the usage feature sequence of each elderly person for each elderly person and perform window decomposition to generate the global service preference matrix and local service fluctuation matrix for each elderly person; Conduct interactive impact analysis on multiple elderly care services based on multiple usage feature sequences, construct a service association map for each elderly person, calculate the interactive impact parameters between any two elderly care services based on multiple service association maps, and generate a historical interactive impact matrix for multiple elderly care services; Encode the service trajectory of historical elderly care service record data, build the elderly care service chain for each elderly person, split each elderly care service chain into multiple candidate service chains, and generate a microservice model set containing multiple target service chains for each elderly person; Through the historical interaction influence matrix, the microservice pattern set of each elderly person is interactively processed, and the individual behavior linkage weights corresponding to multiple target service chains of each elderly person are calculated; According to the individual behavior linkage weights, the global service preference matrix and the local service fluctuation matrix, the global preference parameters of the elderly for each elderly care service are generated. The elderly care service quality is evaluated based on the global preference parameters, and the elderly care service quality evaluation results for each elderly person are generated.

2. According to claim 1, the intelligent evaluation method for elderly care service quality based on multi-dimensional data fusion is characterized in that: Generate a set of microservice patterns containing multiple target service chains for each elderly person, including: Determine multiple service behaviors of the elderly in the historical elderly care service record data, and build elderly care service links for multiple elderly care services based on the time sequence according to the timestamps of the service behaviors and the corresponding service items; Based on the preset sequence length, the service chain of the elderly care service chain is segmented to generate multiple candidate service chains. The distribution frequency of each candidate service chain under the elderly care service chain is counted, and the pattern distribution sequence of each candidate service chain is constructed according to multiple timestamps of the candidate service chain. According to the pattern distribution sequence, the behavioral regularity analysis of each candidate service chain is performed, and the behavioral regularity index of each candidate service chain is calculated. According to the behavioral regularity index, multiple target service chains are determined from the multiple candidate service chains to construct a set of microservice patterns for each elderly person.

3. According to claim 2, the intelligent evaluation method for elderly care service quality based on multi-dimensional data fusion is characterized in that: Through the historical interaction influence matrix, the microservice mode set of each elderly person is interactively processed, and the individual behavior linkage weights corresponding to multiple target service chains of each elderly person are calculated, including: According to the starting point and end point of the target service chain, the corresponding elderly care service combination of each target service chain is determined. According to the elderly care service combination, the interaction influence parameter of each target service chain is extracted from the historical interaction influence matrix. The individual behavior linkage weight of the target service chain is calculated based on the distribution frequency and interaction influence parameter of the target service chain.

4. According to claim 3, the intelligent evaluation method for elderly care service quality based on multi-dimensional data fusion is characterized in that: According to the individual behavior linkage weights, the global service preference matrix and the local service fluctuation matrix, the global preference parameters of the elderly for each elderly care service are generated, including: According to the individual behavior linkage weights of multiple target service chains, the historical interaction influence matrix is ​​modified to generate a dynamic interaction influence matrix of the historical interaction influence matrix. According to the dynamic interaction influence matrix, the microservice pattern set of each elderly person is interactively processed to calculate the target behavior linkage weights corresponding to the multiple target service chains of each elderly person. Determine the multiple target service chains associated with each elderly care service for the elderly, generate the chain tendency factors for each elderly care service for the elderly according to the target behavior linkage weights of the target service chains, calculate the long-term and short-term preference tendency factors for each elderly care service for the elderly according to the global service preference matrix and the local service fluctuation matrix, and generate the global preference parameters for each elderly care service for the elderly according to the chain tendency factors and the long-term and short-term preference tendency factors.

5. According to the method for intelligent evaluation of elderly care service quality based on multi-dimensional data fusion according to claim 1, for the historical interaction influence matrix, it is characterized in that: Calculate the initial correlation coefficient between the usage feature sequences of any two elderly care services, construct a service correlation map for the elderly based on the initial correlation coefficient and a preset correlation threshold, determine the correlation degree of each elderly person with respect to each elderly care service based on the service correlation map, modify the initial correlation coefficient between any two elderly care services for the elderly based on the correlation degree of the elderly care services, generate a target correlation coefficient between any two elderly care services, and construct a local interaction matrix for each elderly person; The multiple local interaction matrices are fused, including determining multiple target correlation coefficients corresponding to any two elderly care services according to the multiple local interaction matrices, calculating the interaction influence parameters between any two elderly care services, and generating a historical interaction influence matrix according to the multiple interaction influence parameters.

6. According to claim 2, the intelligent evaluation method for elderly care service quality based on multi-dimensional data fusion is characterized in that: According to the pattern distribution sequence, the behavior regularity analysis is performed on each candidate service chain, and the behavior regularity index of each candidate service chain is calculated, including: The time interval between any two adjacent candidate service chains in the statistical pattern distribution sequence is calculated, the standard deviation of multiple time intervals is calculated, and the ratio between the distribution frequency of the candidate service chain and the standard deviation of multiple time intervals is used as the behavior regularity index.

7. An intelligent evaluation system for elderly care service quality based on multi-dimensional data fusion, characterized in that: The system is used to implement the intelligent evaluation method for elderly care service quality based on multi-dimensional data fusion as described in any one of claims 1 to 6, including: The service usage feature analysis module is used to collect historical elderly care service record data of multiple elderly people, construct the usage feature sequence of each elderly person for each elderly person and perform window decomposition to generate the global service preference matrix and local service fluctuation matrix for each elderly person; The service interaction impact analysis module is used to perform interaction impact analysis on multiple elderly care services based on multiple usage feature sequences, construct a service association map for each elderly person, calculate the interaction impact parameters between any two elderly care services based on multiple service association maps, and generate a historical interaction impact matrix for multiple elderly care services; The microservice link analysis module is used to encode the service trajectory of historical elderly care service record data, build the elderly care service link for each elderly person, split the service chain of each elderly care service link to generate multiple candidate service chains, and generate a microservice pattern set containing multiple target service chains for each elderly person; The individual behavior linkage analysis module is used to interactively process the microservice pattern set of each elderly person through the historical interaction influence matrix, and calculate the individual behavior linkage weights corresponding to multiple target service chains of each elderly person; The elderly care service quality evaluation management module is used to generate the global preference parameters of the elderly for each elderly care service based on the individual behavior linkage weights, the global service preference matrix and the local service fluctuation matrix, evaluate the elderly care service quality based on the global preference parameters, and generate the elderly care service quality evaluation results for each elderly person.

Images