Community elderly person portrait construction method integrating Delphi, LCA and ES

Abstract

The invention discloses a community elder figure portrait construction method integrating Delphi, LCA and ES. The method comprises the following steps: 1, quantitatively analyzing the importance of elder user description dimensions from the perspective of expert knowledge based on a Delphi method; 2, community elder figure portrait types are divided based on an LCA method; and 3, based on an ES method, determining optimal description semantic texts of all types of community elderly portraits. According to the method, the influence of expert knowledge and user characteristics is comprehensively considered, the constructed community elderly figure portraits have high representativeness, and the whole method is simple to operate, high in portability and easy to popularize and apply.

Description

Technical Field

[0001] The invention relates to a method for constructing community elderly person portraits by integrating Delphi, LCA and ES, and belongs to the technical field of service decision management. Background Art

[0002] With the development of economy and society, the requirements for community services to be aging-friendly are getting higher and higher. In the process of providing community services, the characteristics and needs of elderly users determine the content and form of community services, which means that service designers need to have a deep understanding of the characteristics and needs of elderly residents in the community. The key to efficiently analyzing the characteristics and needs of elderly users in the community lies in extracting key information such as the typical physical condition, daily behavior and activities of elderly users in the community. Existing technologies make it difficult to intuitively and accurately obtain and present key information of elderly residents in the community. Summary of the invention

[0003] Purpose of the invention: The present invention proposes a method for constructing community elderly portraits that integrates Delphi, LCA and ES. The description dimensions of elderly users are established through the Delphi method, and the Delphi and LCA methods are integrated. Starting from the key description dimensions identified by experts, LCA analysis is performed on key information such as user characteristics and needs fed back by community elderly people, and community elderly users are classified. The ES method is used to give appropriate descriptive semantic texts to various types of community elderly residents to complete the construction of community elderly portraits. A method for constructing community elderly portraits that takes expert knowledge and user feedback as the starting point and integrates Delphi, LCA and ES methods is proposed.

[0004] Technical solution, in order to solve the above technical problems, the present invention proposes a method for constructing community elderly portraits integrating Delphi, LCA and ES, the method comprising the following steps:

[0005] (1) Determination of the description dimensions of community elderly portraits

[0006] From the perspective of expert experience, the focus group method and Delphi method are integrated to conduct quantitative analysis on the description dimensions of elderly users. By calculating the authority of experts, the consistency of expert judgments and the importance of the description dimensions of elderly users, the key description dimensions of elderly users are determined.

[0007] (2) Determination of evaluation indicators for meeting the needs of community-based elderly-friendly services

[0008] From the perspective of user feedback, combined with literature analysis and focus group methods, we extracted the descriptive indicators of the community elderly portraits, used the LCA method to divide the community elderly portrait types, and obtained the proportion of the elderly portrait types in each community and their typical performance in descriptive indicators.

[0009] (3) Construction of an evaluation index system for community services suitable for the elderly

[0010] Literature analysis and focus group methods were used to extract the descriptive semantic texts of the community elderly portrait types described in step (2). The importance and relevance of each descriptive semantic text to the corresponding community elderly portrait type were analyzed based on the ES method to determine the best descriptive semantic text for each type of community elderly portrait.

[0011] Furthermore, the specific steps of step (1) are as follows:

[0012] (1-1), design the Dephi survey questionnaire;

[0013] Based on user surveys and focus group methods, we obtained the description dimensions of elderly users, formed a Delphi questionnaire, and collected expert evaluation data;

[0014] (1-2) Determine the importance of the descriptive dimensions, including the following:

[0015] (1-3), data analysis of the Delphi questionnaire;

[0016] The Delphi questionnaire was used to collect the evaluation data of experts on the description dimensions of the elderly portrait. n Importance Rating M in , Evaluation based on Ca i and indicator familiarity Cs i , quantitative analysis and description dimension D n The average importance of Expert's authority coefficient Cr i The consistency coefficient Cv of expert judgment n , where I represents the number of experts; that is:

[0017]

[0018]

[0019]

[0020] Furthermore, the specific steps of step (2) are as follows:

[0021] (2-1) Determination of descriptive indicators for community elderly portraits;

[0022] Based on the descriptive dimension results obtained in step (1), combined with literature analysis and focus group method, descriptive indicators of community elderly people's character portraits are extracted;

[0023] (2-2) Design LCA survey questionnaire;

[0024] Based on the summarized descriptive indicators, an LCA questionnaire was formed, which included three aspects: survey instructions and descriptive indicator self-assessment form;

[0025] Descriptive indicator self-assessment form: two-point scoring method, "yes" is assigned 1 point, "no" is assigned 0 points;

[0026] (2-3) LCA classification of community elderly portraits;

[0027] (2-4) Data analysis of LCA questionnaire

[0028] Based on the data collected by the LCA survey questionnaire, parameter estimation is performed according to the individual response pattern on the indicators, and the variance is decomposed into within-category and between-category. Within each category, the joint probability of multiple two-point scoring items is expressed as:

[0029]

[0030] In the above formula, y i represents individual i, y ij represents the score of individual i in the two options of indicator j, J represents the total number of indicators, c i is a latent category variable with k levels. When considering multiple category levels, formula (4) is expanded to:

[0031]

[0032] p(c i =k) ​​represents the ratio of a certain category group k to the total, also known as the latent category probability; according to the individual answer type, the Bayesian posterior probability is calculated as the classification basis, and the calculation formula is as follows:

[0033]

[0034] Calculate the reliability evaluation indicators AIC, BIC and aBIC of the LCA analysis results to test the rationality of the classification results. The calculation formula is as follows:

[0035]

[0036] In the formula, k is the number of levels, is the maximum likelihood estimate, m is the sample size;

[0037] Calculate the Entropy value E k Evaluation of classification accuracy, the calculation formula is as follows:

[0038]

[0039] Among them, P ikis the posterior probability that individual i belongs to category k, and its value range is 0 to 1. The closer it is to 1, the more accurate the classification is. K is the total number of potential category levels.

[0040] Furthermore, the specific steps of step (3) are as follows:

[0041] (3-1) Design ES survey questionnaire

[0042] Based on the classification results of the community elderly portrait types obtained in step (2), combined with literature analysis and focus group method, semantic texts describing the community elderly portrait types were generated, and based on this, an ES survey questionnaire was designed, which included three aspects: survey instructions, basic information of experts, semantic text relevance evaluation table, and semantic text importance evaluation table;

[0043] Survey description: including research purpose, background, significance and instructions for filling in;

[0044] Basic profile of the expert: including research purpose, background, work unit, position, technical title, years of work experience, highest education level, and work field;

[0045] Semantic text relevance evaluation table: using the Likert 10-point rating method, the range of relevant to irrelevant is 10 points to 1 point;

[0046] Semantic text importance evaluation table: using the Likert 10-point rating method, the value range from important to unimportant is 10 points to 1 point;

[0047] (3-2) Semantic judgment of the description type of community elderly portrait;

[0048] (3-3) Data analysis of ES questionnaire

[0049] Through the ES questionnaire collection, we obtain the evaluation data of experts on the description semantics; based on the information object-element concept of ES, we assume that the portrait type of the elderly in the community is object O, the description semantic text is feature c', the overall value of O about c' is v, and an ordered triple is constructed:

[0050] M=(O,c',v) (11)

[0051] When a character portrait type has many feature elements, it is represented by multi-dimensional object elements, namely:

[0052]

[0053] Among them, c' n is the nth element in c', v n is the nth element in v, c' n Corresponding to an old man portrait type O, v n Then O is about c'n The overall value of

[0054] Based on the collected semantic text relevance evaluation data, calculate the semantic text c' n Correlation with the elderly portrait type O v nx , the calculation formula is as follows:

[0055]

[0056] In the formula, y s is the number of scores s for each semantic text, S is the highest score, P is the number of evaluation experts, and v nx The larger the value, the more it describes the semantic text c' n The higher the correlation with the elderly person portrait type O;

[0057] Based on the recovered descriptive semantic text importance evaluation data, calculate the descriptive semantic text c' n Importance in the elderly portrait type O ny , the calculation formula is as follows:

[0058]

[0059] In the formula, I p is the importance score of expert p on the semantic text, P is the number of evaluation experts, and v ny The larger the value, the more it describes the semantic text c' n The higher the importance of type O for the elderly portrait; based on and Calculate O with respect to c' n The total value v n :

[0060]

[0061] Furthermore, the description dimensions of the portrait of the elderly in the community in step (1-1) include: pre-retirement occupation, health status, living ability, co-residents, medical activities, interests and hobbies, daily activities, community space and venues, usage of community equipment and facilities, and community service status.

[0062] Beneficial effects: Compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:

[0063] The present invention integrates the community elderly portrait construction method of Delphi, LCA and ES, and verifies the effectiveness and rationality of the method through the construction case of urban community elderly portrait. The method is extended to other service management decision-making fields, which can intuitively and accurately obtain and present the key information of community elderly residents, providing a reliable basis for service design. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The present invention is a flow chart of the method for constructing community elderly portraits by integrating Delphi, LCA and ES. DETAILED DESCRIPTION

[0065] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0066] like Figure 1 As shown, the present invention proposes a method for constructing a community elderly person portrait by integrating Delphi, LCA and ES, comprising the following steps:

[0067] Step 1: Determine the description dimensions of the community elderly portrait

[0068] (1) Design of Delphi survey questionnaire

[0069] The user survey and focus group method were used to obtain the descriptive dimensions of the elderly users and form a Delphi questionnaire, which mainly includes three aspects: survey instructions, basic profile of experts, and descriptive dimension importance evaluation table;

[0070] Survey description: including research purpose, background, significance and instructions for filling in;

[0071] Expert basic profile: including research purpose, background, work unit, position, technical support, years of work experience, highest education, work field, expert's familiarity with the content of this inquiry and the basis for selecting indicators; Descriptive dimension importance evaluation table: using Likert 5-level scoring method, the value range from important to unimportant is 5 points to 1 point, and the descriptive dimensions are expanded as shown in Table 1:

[0072] Table 1 Dimensions of persona descriptions derived from user surveys and focus groups

[0073]

[0074] (2) Determine the importance of descriptive dimensions

[0075] 1) Distribution and collection of Delphi questionnaires

[0076] After completing the design of the Delphi survey questionnaire, we contacted 7 experts in related fields to fill in the questionnaire online and collect it;

[0077] 2) Data analysis of Delphi questionnaire

[0078] The Delphi questionnaire was used to collect the evaluation data of experts on the description dimensions of the elderly portrait. n Importance Rating M in , Evaluation based on Ca i and indicator familiarity Cs i , quantitative analysis and description dimension D n The average importance of Expert's authority coefficient Cr i The consistency coefficient Cv of expert judgment n (where I represents the number of experts), that is:

[0079]

[0080]

[0081]

[0082] Based on the expert evaluation data collected from the Delphi questionnaire, Ca i and indicator familiarity Cs i , use formula (1) to calculate the authority coefficient Cr of each expert i , taking the calculation of an expert's authority coefficient as an example,

[0083]

[0084] After completing the authority coefficient Cr of each expert i After calculation, the authority coefficient Cr is retained i >7 Delphi questionnaire data, with a total of 6 experts with authority coefficients Cr i >7;

[0085] Based on the authority coefficient Cr i >7 Delphi questionnaire data, calculate the experts' descriptive dimension D n The average importance score and the consistency coefficient of expert judgment Cv n , based on pre-retirement occupation D 1 For example,

[0086] Formula (2) is used to calculate the pre-retirement occupation D 1 The average importance of

[0087]

[0088] Using formula (3), calculate the pre-retirement occupation D 1 The expert judgment consistency coefficient Cv 1 :

[0089]

[0090] Complete the average importance of all descriptive dimensions Coefficient of consistency with expert judgment Cv 1 , the results are shown in Table 2;

[0091] Table 2 Results of average importance of descriptive dimensions and expert judgment consistency coefficient

[0092]

[0093] When describing the average importance of a dimension Coefficient of variation Cv n When <0.25, the dimension is retained, otherwise it is eliminated. From Table 2, we can see that the average importance and coefficient of variation of all descriptive dimensions meet the standards.

[0094] Step 2: Classification of community elderly portrait types

[0095] (1) Determination of description indicators for community elderly portraits

[0096] Based on the descriptive dimension results obtained in step 1, combined with literature analysis and focus group method, the descriptive indicators of community elderly people’s portraits were extracted. 7 Taking this as an example, a total of 40 daily behavior activity dimension description indicators of each community scene are extracted under this dimension, as shown in Table 3;

[0097] Table 3 Descriptive indicators of daily behavior activities compiled by literature analysis and focus group method

[0098]

[0099]

[0100] (2) Design of LCA survey questionnaire

[0101] Based on the summarized descriptive indicators, the LCA questionnaire was formed, which mainly includes three aspects: survey instructions and descriptive indicator self-assessment form;

[0102] Descriptive indicator self-assessment form: two-point scoring method, "the phenomenon exists" is assigned 1 point, and "the phenomenon does not exist" is assigned 0 points;

[0103] (3) LCA classification of community elderly portraits

[0104] 1) Distribution and collection of LCA questionnaires

[0105] After completing the design of the LCA survey questionnaire, 162 elderly people aged 60 and above living in the community were recruited to fill in the questionnaire and collected;

[0106] 2) Data analysis of LCA questionnaire

[0107] Based on the data collected from the LCA survey questionnaire, parameter estimation is performed based on the individual response patterns on the indicators. The variance is decomposed into within-category and between-category. Within each category, the joint probability of multiple two-point scoring items can be expressed as:

[0108]

[0109] In the above formula, y i represents individual i, y ij represents the score of individual i in the two options of indicator j. J represents the total number of indicators, c is a latent class variable with k levels. When considering multiple class levels at the same time, formula (4) is expanded to:

[0110]

[0111] p(c i =k) ​​represents the ratio of a certain category group k to the total, also known as the latent category probability. According to the individual answer type, the Bayesian posterior probability is calculated as the classification basis. The calculation formula is as follows:

[0112]

[0113] Calculate the reliability evaluation indicators AIC, BIC and aBIC of the LCA analysis results to test the rationality of the classification results. The calculation formula is as follows:

[0114]

[0115]

[0116]

[0117] In the formula, k is the number of levels, is the maximum likelihood estimate, and m is the sample size. Calculate the Entropy value E k To evaluate the classification accuracy, the Entropy value is calculated as follows:

[0118]

[0119] Among them, P ik is the posterior probability that individual i belongs to category k, and its value range is 0 to 1. The closer it is to 1, the more accurate the classification.

[0120] Mplus software was used to conduct LCA analysis and establish a character portrait of the elderly in the community based on their daily behavior and activities. 7 Taking the dimension as an example, the classification results of the daily behavior and activity description dimension of the community elderly character portrait are shown in Table 4;

[0121] Table 4 Classification results of daily behavior and activity description dimensions of community elderly portraits

[0122]

[0123] Then, based on the classification results, the specific performance of the descriptive indicators of the daily behavior activity dimensions of the portraits of the elderly in each category of the community was determined, taking the C1 category as an example, as shown in Table 5.

[0124] Table 5 Specific performance results of the descriptive indicators of the daily behavior and activities dimension of the community elderly portraits in category C1

[0125]

[0126]

[0127] Then, the reliability evaluation indicators AIC, BIC and aBIC of the LCA analysis results were calculated to test the rationality of the classification results; the Entropy value E was calculated. k Evaluate classification accuracy. The calculation results show that AIC = 2954.978, BIC = 3836.135, aBIC = 2675.587, all less than 4000; E k =0.993, which is close to 1, indicating that the classification results of the model are reasonable and the classification accuracy is good.

[0128] Step 3: Determine semantics of description of community elderly household portrait type

[0129] (1) Design ES survey questionnaire

[0130] Based on the classification results of the community elderly portrait types obtained in step 2, combined with literature analysis and focus group method, semantic text describing the community elderly portrait types was generated, and the ES survey questionnaire was designed based on this, which mainly includes three aspects: survey instructions, basic information of experts, semantic text relevance evaluation table and semantic text importance evaluation table.

[0131] Survey description: including research purpose, background, significance and instructions for filling in;

[0132] Basic profile of the expert: including research purpose, background, work unit, position, technical title, years of work experience, highest education level, and work field;

[0133] Semantic text relevance evaluation table: using the Likert 10-point rating method, the range of relevant to irrelevant is 10 points to 1 point;

[0134] Semantic text importance evaluation table: using the Likert 10-point rating method, the value range from important to unimportant is 10 points to 1 point;

[0135] (2) Semantic judgment of the description type of community elderly portrait

[0136] 1) Distribution and collection of ES questionnaires

[0137] After completing the design of the ES survey questionnaire, we contacted 24 experts in related fields to fill out the questionnaire online and collect it;

[0138] 2) Data analysis of ES questionnaire

[0139] Through the ES questionnaire, we collected experts’ evaluation data on the semantics of descriptions;

[0140] Based on the concept of information object element of ES, let the portrait type of the elderly in the community be object O, the descriptive semantic text be feature c', the overall value of O with respect to c' be v, and construct an ordered triple:

[0141] M=(O,c',v) (11)

[0142] When a character portrait type has many feature elements, it is represented by multidimensional object elements, that is,

[0143]

[0144] Among them, c' n is the nth element in c', v n is the nth element in v, c' n Corresponding to an old man portrait type O, v n Then O is about c' n The overall value of

[0145] Based on the collected semantic text relevance evaluation data, calculate the semantic text c' n Correlation with the elderly portrait type O v nx , the calculation formula is as follows:

[0146]

[0147] In the formula, y sis the number of scores s for each semantic text, S is the highest score, P is the number of evaluation experts, and v nx The larger the value, the more it describes the semantic text c' n The higher the correlation with the elderly person portrait type O;

[0148] Based on the recovered descriptive semantic text importance evaluation data, calculate the descriptive semantic text c' n Importance in the elderly portrait type O ny , the calculation formula is as follows:

[0149]

[0150] In the formula, I p is the importance score of expert p on the semantic text, P is the number of evaluation experts, and v ny The larger the value, the more it describes the semantic text c' n The higher the importance of type O for the elderly portrait;

[0151] based on Calculate O with respect to c' n The total value v n :

[0152]

[0153] In the formula, v n The larger the value, the more it describes the semantic text c' n The more accurate the description of the elderly character portrait type O is.

[0154] Taking the C1 category of daily behavior activities as an example, literature analysis and focus group method were used to generate semantic texts describing the character portraits of community elderly people in category C1, including: unstable health status, mobility difficulties, long-term illness, social isolation, sensory failure, mental depression, long-term bed rest, high curiosity, active daily mental state, three generations living together, and colorful life, and expert evaluation data were collected;

[0155] Based on the concept of ES information object element, let the C1 community elderly portrait type be object O 1 , describing the semantic text as feature c', the overall value of O with respect to c' is v, and the ordered triples are constructed using formula (12):

[0156]

[0157] Where c' 1 Due to unstable health, c' 2 For people with disabilities, 3 For long-term illness, 4 For social isolation,5 For sensor failure, c' 6 For depression, c' 7 For long-term bed rest, 8 For a higher desire for knowledge, c' 9 For daily mental state active, c' 10 For three generations living together and c' 11 To enrich life;

[0158] Based on the collected expert evaluation data, calculate the descriptive semantic text c' n Correlation with the elderly portrait type O v nx , the importance of the descriptive semantic text c'n in the elderly person portrait type O and the elderly person portrait type O on the descriptive semantic text c' n The total value v n ;

[0159] The recycled object O 1 Daily mental state activity in relevant expert evaluation data c' 9 Take the relevant data as an example;

[0160] Using formula (13), calculate the daily mental state activity c' 9 With object O 1 The correlation v 9x :

[0161]

[0162] Formula (14) is used to calculate the daily mental state activity c' 9 The importance v of object O1 9y :

[0163]

[0164] Formula (15) is used to calculate the daily mental state activity c' of object O1 9 The total value v 9 :

[0165] v 9 =0.115151515·6.541667=0.753283

[0166] Complete all description semantic text c' n Relative to the total value v of object O n The calculation and sorting results are as follows:

[0167]

[0168] Where c' 1Due to unstable health, c' 2 For people with disabilities, 3 For long-term illness, 4 For social isolation, 5 For sensor failure, c' 6 For depression, c' 7 For long-term bed rest, 8 For a higher desire for knowledge, c' 9 For daily mental state active, c' 10 For three generations living together and c' 11 To make life colorful.

[0169] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Two modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles or novel features disclosed herein.

Claims

1. A method for constructing community elderly portraits integrating Delphi, LCA and ES, characterized in that: The method comprises the following steps: (1) Determination of the description dimensions of community elderly portraits From the perspective of expert experience, the focus group method and Delphi method are integrated to conduct quantitative analysis on the description dimensions of elderly users. By calculating the authority of experts, the consistency of expert judgments and the importance of the description dimensions of elderly users, the key description dimensions of elderly users are determined. (2) Determination of evaluation indicators for meeting the needs of community-based elderly-friendly services From the perspective of user feedback, combined with literature analysis and focus group methods, we extracted the descriptive indicators of the community elderly portraits, used the LCA method to divide the community elderly portrait types, and obtained the proportion of the elderly portrait types in each community and their typical performance in descriptive indicators. (3) Construction of an evaluation index system for community services suitable for the elderly Literature analysis and focus group methods were used to extract the descriptive semantic texts of the community elderly portrait types described in step (2). The importance and relevance of each descriptive semantic text to the corresponding community elderly portrait type were analyzed based on the ES method to determine the best descriptive semantic text for each type of community elderly portrait.

2. A method for constructing community elderly portraits integrating Delphi, LCA and ES according to claim 1, characterized in that: The specific steps of step (1) are as follows: (1-1), design the Dephi survey questionnaire; Based on user surveys and focus group methods, we obtained the description dimensions of elderly users, formed a Delphi questionnaire, and collected expert evaluation data; (1-2) Determine the importance of the descriptive dimensions, including the following: (1-3), data analysis of the Delphi questionnaire; The Delphi questionnaire was used to collect the evaluation data of experts on the description dimensions of the elderly portrait. n Importance Rating M in , Evaluation based on Ca i and indicator familiarity Cs i , quantitative analysis and description dimension D n The average importance of Expert's authority coefficient Cr i The consistency coefficient Cv of expert judgment n , where I represents the number of experts; that is:

3. A method for constructing community elderly portraits integrating Delphi, LCA and ES according to claim 2, characterized in that: The specific steps of step (2) are as follows: (2-1) Determination of descriptive indicators for community elderly portraits; Based on the descriptive dimension results obtained in step (1), combined with literature analysis and focus group method, descriptive indicators of community elderly people's character portraits are extracted; (2-2) Design LCA survey questionnaire; Based on the summarized descriptive indicators, an LCA questionnaire was formed, which included three aspects: survey instructions and descriptive indicator self-assessment form; Descriptive indicator self-assessment form: two-point scoring method, "yes" is assigned 1 point, "no" is assigned 0 points; (2-3) LCA classification of community elderly portraits; (2-4) Data analysis of LCA questionnaire Based on the data collected by the LCA survey questionnaire, parameter estimation is performed according to the individual response pattern on the indicators, and the variance is decomposed into within-category and between-category. Within each category, the joint probability of multiple two-point scoring items is expressed as: In the above formula, y i represents individual i, y ij represents the score of individual i in the two options of indicator j, J represents the total number of indicators, c i is a latent category variable with k levels. When considering multiple category levels, formula (4) is expanded to: p(c i =k) ​​represents the ratio of a certain category group k to the total, also known as the latent category probability; according to the individual answer type, the Bayesian posterior probability is calculated as the classification basis, and the calculation formula is as follows: Calculate the reliability evaluation indicators AIC, BIC and aBIC of the LCA analysis results to test the rationality of the classification results. The calculation formula is as follows: In the formula, k is the number of levels, is the maximum likelihood estimate, m is the sample size; Calculate the Entropy value E k Evaluation of classification accuracy, the calculation formula is as follows: Among them, P ik is the posterior probability that individual i belongs to category k, and its value range is 0 to 1, and K is the total number of potential category levels.

4. A method for constructing community elderly portraits integrating Delphi, LCA and ES according to claim 3, characterized in that: The specific steps of step (3) are as follows: (3-1) Design ES survey questionnaire Based on the classification results of the community elderly portrait types obtained in step (2), combined with literature analysis and focus group method, semantic texts describing the community elderly portrait types were generated, and based on this, an ES survey questionnaire was designed, which included three aspects: survey instructions, basic information of experts, semantic text relevance evaluation table, and semantic text importance evaluation table; Survey description: including research purpose, background, significance and instructions for filling in; Basic profile of the expert: including research purpose, background, work unit, position, technical title, years of work experience, highest education level, and work field; Semantic text relevance evaluation table: using the Likert 10-point rating method, the range of relevant to irrelevant is 10 points to 1 point; Semantic text importance evaluation table: using the Likert 10-point rating method, the value range from important to unimportant is 10 points to 1 point; (3-2) Semantic judgment of the description type of community elderly portrait; (3-3) Data analysis of ES questionnaire Through the ES questionnaire collection, we obtain the evaluation data of experts on the description semantics; based on the information object-element concept of ES, we assume that the portrait type of the elderly in the community is object O, the description semantic text is feature c', the overall value of O about c' is v, and an ordered triple is constructed: M=(O,c',v) (11) When a character portrait type has many feature elements, it is represented by multi-dimensional object elements, namely: Among them, c' n is the nth element in c', v n is the nth element in v, c' n Corresponding to an old man portrait type O, v n Then O is about c' n The overall value of Based on the collected semantic text relevance evaluation data, calculate the semantic text c' n Correlation with the elderly portrait type O v nx , the calculation formula is as follows: In the formula, y s The number of scores s for each semantic text, S is the highest score, P is the number of evaluation experts, and the descriptive semantic text c' is calculated based on the importance evaluation data of the recovered descriptive semantic texts. n Importance in the elderly portrait type O ny , the calculation formula is as follows: In the formula, I p Score the importance of the semantic text described by expert p; based on and Calculate O with respect to c' n The total value v n :

5. A method for constructing community elderly portraits integrating Delphi, LCA and ES according to claim 4, characterized in that: The description dimensions of the community elderly person portrait in step (1-1) include: pre-retirement occupation, health status, living ability, co-residents, medical activities, interests and hobbies, daily activities, community space and place, use of community equipment and facilities, and community service status.

Images