Nursing home check-in demand prediction method, device and equipment and storage medium
By using multiple ARIMA models and integrated processing technologies in nursing home occupancy demand forecasting, combining historical occupancy data and exogenous variables of nursing home demand, the problems of prediction instability and inaccuracy in the existing technology are solved, achieving higher prediction accuracy and stability.
Patent Information
- Application Number
- CN202411734437.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-27
- Publication Date
- 2025-03-11
AI Technical Summary
When predicting nursing home accommodation demand, the existing technology relies on a single predictive model to focus only on time series trends and seasonal fluctuations in historical occupancy data, resulting in unstable prediction results and affecting accuracy.
By obtaining nursing home historical occupancy data and nursing home demand exogenous variables (such as population mortality data and characteristic labels for the elderly), at least two trained ARIMA models are used to predict occupancy demand and integrated processing of the prediction results to improve the accuracy and stability of the prediction.
By combining multi-dimensional historical occupancy data with exogenous variables of nursing home demand, multiple ARIMA models are used to predict and integrate processing, the accuracy and stability of nursing home occupancy demand prediction is significantly improved and the robustness of predicted results is enhanced.
Smart Images
Figure CN119599215A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the fields of medical technology and artificial intelligence technology, and in particular to a method, device, equipment and storage medium for predicting nursing home occupancy demand. Background Art
[0002] Elderly health service management is an important part of the medical technology field. With the increasing severity of the aging population, the forecast of nursing home admission demand (such as the forecast of nursing home bed demand) has become particularly important. Accurate admission demand forecasting can not only help nursing homes optimize resource allocation, but also improve the quality of elderly care services.
[0003] At present, predictions are usually made through a single forecasting model, mainly based on the historical occupancy data of nursing homes, and only focusing on the trends and seasonal fluctuations in the time series of historical occupancy data. Deviations or errors may occur, resulting in unstable forecast results and affecting the accuracy of occupancy demand forecasts.
[0004] Therefore, how to improve the accuracy of nursing home admission demand forecasting has become a technical problem that needs to be solved urgently. Summary of the invention
[0005] The main purpose of the embodiments of the present application is to propose a method, device, equipment and storage medium for predicting nursing home admission demand, aiming to improve the accuracy of nursing home admission demand prediction.
[0006] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a method for predicting nursing home admission demand, the method comprising:
[0007] Obtaining historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly;
[0008] Using at least two trained ARIMA models, forecasting nursing home occupancy demand based on the historical occupancy data and the nursing home demand exogenous variables, to obtain occupancy demand data for at least two nursing homes in the target city;
[0009] The admission demand data are integrated and processed to obtain the admission demand data of the target nursing home in the target city.
[0010] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application proposes a nursing home admission demand prediction device, the device comprising:
[0011] A data acquisition module, used to acquire historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly;
[0012] An occupancy demand forecasting module is used to forecast the occupancy demand of nursing homes according to the historical occupancy data and the exogenous variables of nursing home demand by using at least two trained ARIMA models, and obtain the occupancy demand data of at least two nursing homes in the target city;
[0013] The occupancy demand integration module is used to integrate the occupancy demand data of the nursing home to obtain the occupancy demand data of the target nursing home in the target city.
[0014] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes a computer device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the method described in the first aspect when executing the computer program.
[0015] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.
[0016] The present application proposes a method, device, equipment and storage medium for predicting the demand for nursing home admission. The method for predicting the demand for nursing home admission obtains the historical admission data of nursing homes in a target city, as well as exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly; predicts the demand for nursing home admission based on the historical admission data and exogenous variables of nursing home demand through at least two trained ARIMA models, and obtains at least two nursing home admission demand data for the target city; integrates the admission demand data to obtain target nursing home admission demand data for the target city. In this way, by combining multi-dimensional historical admission data with exogenous variables of nursing home demand, using at least two trained ARIMA models for prediction, and then obtaining the target nursing home admission demand data through integrated processing, the prediction accuracy can be improved, and the stability and robustness of the prediction results can be enhanced, thereby improving the accuracy of the prediction of nursing home admission demand. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 is a flow chart of a method for predicting nursing home occupancy demand provided in an embodiment of the present application;
[0018] Figure 2 This is another flow chart of the nursing home occupancy demand prediction method provided in the embodiment of the present application.
[0019] Figure 3 It is a schematic diagram of a training architecture for training multiple ARIMA models provided in an embodiment of the present application;
[0020] Figure 4 is a schematic block diagram of a nursing home occupancy demand prediction device provided in an embodiment of the present application;
[0021] Figure 5 It is a schematic block diagram of the structure of the computer device provided in the embodiment of the present application. DETAILED DESCRIPTION
[0022] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0023] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.
[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0025] First, some nouns involved in this application are analyzed:
[0026] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
[0027] Autoregressive Integrated Moving Average Model (ARIMA): ARIMA is a statistical model widely used in time series analysis, especially for predicting and modeling time-dependent numerical data. The ARIMA model mainly consists of three core components: Autoregressive (AR), Integrated (I), and Moving Average (MA).
[0028] The AR part represents the linear relationship between the current value of the time series and its past values. In simple terms, the current value is the weighted sum of the previous values. The AR part can be represented by the symbol AR(p), where p represents the number of autoregressive terms, which can be understood as the lag order of the AR part (that is, the number of past values considered).
[0029] The I part is used to convert a non-stationary time series into a stationary time series. Usually, differencing is done by calculating the difference between the current value and the previous value. The purpose of differencing is to remove the trend or seasonal component in the data and make the data stationary. The I part can be represented by the symbol I(d), where d represents the differencing order (i.e. the number of differences required to make the time series stationary).
[0030] The MA part represents the relationship between the current value and the past errors. The error here refers to the difference between the predicted value of the ARIMA model and the actual value. The MA part can be represented by the symbol MA(q), where q represents the number of moving average terms, which can be understood as the lag order of the MA part (that is, the number of past errors considered).
[0031] Therefore, the ARIMA model can be expressed as ARIMA(p, d, q), where p, d, and q are the parameters of the ARIMA model.
[0032] Based on this, the embodiments of the present application provide a method, device, equipment and storage medium for predicting nursing home admission demand, aiming to improve the accuracy of nursing home admission demand prediction.
[0033] The nursing home admission demand prediction method, device, equipment and storage medium provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the nursing home admission demand prediction method in the embodiments of the present application is described.
[0034] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
[0035] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.
[0036] The method for predicting the demand for admission to a nursing home provided in the embodiment of the present application relates to the fields of medical technology and artificial intelligence technology. The method for predicting the demand for admission to a nursing home provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a vehicle-mounted terminal, a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the method for predicting the demand for admission to a nursing home, etc., but is not limited to the above forms.
[0037] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer computer devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.
[0038] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.
[0039] Figure 1 is an optional flow chart of the nursing home occupancy demand prediction method provided in the embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S103.
[0040] Step S101, obtaining historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly.
[0041] Target cities are cities with forecasted demand for nursing home admissions.
[0042] Obtain the historical occupancy data of nursing homes in the target city. The historical occupancy data of nursing homes reflects the past occupancy demand pattern and is a key factor affecting the occupancy demand of nursing homes.
[0043] Exemplarily, the historical admission data may include the historical number of residents, admission time, length of stay, discharge time, etc. of the nursing home.
[0044] Obtain exogenous variables of nursing home demand in the target city. Exogenous variables of nursing home demand refer to additional key factors that affect nursing home occupancy demand in addition to the historical occupancy data of nursing homes. Exogenous variables of nursing home demand mainly include population mortality data of the target city and characteristic labels of the elderly, and other factors closely related to nursing home occupancy demand.
[0045] For example, considering that the mortality rate reflects the changing trend of the elderly population, a higher mortality rate may lead to a decrease in the number of elderly people, thereby affecting the demand for elderly care services. Therefore, the historical mortality rate statistics by month or quarter can be used as the mortality rate data.
[0046] For example, considering that the needs of elderly people of different age groups for nursing homes vary significantly, the characteristic label of the elderly may include age stage, such as 65 years old and below, 66-70 years old, 71-75 years old, 76-80 years old, 81 years old and above.
[0047] For example, considering that elderly people with different living conditions have significant differences in their needs for nursing homes, the characteristic labels of the elderly may include living conditions, such as living with family members or living alone.
[0048] For example, considering that the health status of the elderly directly affects the demand for admission to nursing homes, the characteristic tags of the elderly may include health status. The tag values of health status are, for example, healthy vitality, chronic disease, and serious disease.
[0049] For example, considering that the self-care ability of the elderly can reflect their dependence on nursing home services, the characteristic label of the elderly can include self-care ability. The label value of self-care ability can be self-care, semi-self-care, disability, and dementia.
[0050] This provides a multi-dimensional basis for predicting the demand for nursing home admission.
[0051] Step S102, using at least two trained ARIMA models, forecasting nursing home occupancy demand based on historical occupancy data and exogenous variables of nursing home demand, and obtaining occupancy demand data of at least two nursing homes in the target city.
[0052] Based on the ensemble learning strategy, at least two trained ARIMA models are used to forecast nursing home occupancy demand based on historical occupancy data and exogenous variables of nursing home demand using their respective parameters to obtain at least two nursing home occupancy demand data for the target city.
[0053] Exemplarily, each nursing home occupancy demand data may be the nursing home bed demand at a future time point t predicted by each trained ARIMA model, such as the nursing home bed demand {10, 12, 14} in the next three months.
[0054] It is understandable that the parameters of any two trained ARIMA models are different, which also means that any two nursing home admission demand data are different.
[0055] In this way, on the one hand, since the historical occupancy data reflects the past occupancy demand pattern, the trained ARIMA model is good at capturing the trend, seasonality, cyclical and other characteristics in these time series data; on the other hand, since the exogenous variables of nursing home demand can not only consider the internal changes of time series data, but also integrate the changes of the external environment, by introducing the exogenous variables of nursing home demand into multiple trained ARIMA models, it can help each trained ARIMA model to fully capture the various factors affecting the occupancy demand of nursing homes, and enhance the trained ARIMA model's ability to predict future changes. As a result, the prediction results of each trained ARIMA model are more accurate, comprehensive and reliable.
[0056] Step S103, integrating and processing the admission demand data to obtain the admission demand data of the target nursing home in the target city.
[0057] Based on the integrated learning strategy, the admission demand data of at least two nursing homes in the target city are integrated to obtain the admission demand data of the target nursing home in the target city.
[0058] In this way, by integrating multiple trained ARIMA models for demand forecasting, we can make full use of the advantages of different ARIMA models and improve the accuracy of forecasting. Each trained ARIMA model may perform differently under different data patterns or trends. Integrated processing can reduce the errors or accidental fluctuations that may occur in a single model, thereby generating more stable and accurate forecasting results. Especially in the face of high uncertainty or incomplete data quality, integrated forecasting can significantly reduce errors and improve the credibility of forecasts.
[0059] As a result, personalized predictions are achieved at the city level, providing a detailed scientific basis for the allocation of nursing home resources at the city level, which will help various nursing homes, governments and relevant decision makers to optimize the allocation of nursing resources (such as planning nursing home bed resources in advance), improve the quality of nursing services and better respond to future challenges in elderly care needs.
[0060] The method for predicting nursing home admission demand provided in the above embodiment obtains historical admission data of nursing homes in a target city, as well as exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly; predicts nursing home admission demand based on historical admission data and exogenous variables of nursing home demand through at least two trained ARIMA models, and obtains at least two nursing home admission demand data in the target city; integrates the admission demand data to obtain target nursing home admission demand data in the target city. In this way, by combining multi-dimensional historical admission data with exogenous variables of nursing home demand, using at least two trained ARIMA models for prediction, and then obtaining target nursing home admission demand data through integrated processing, the prediction accuracy can be improved, and the stability and robustness of the prediction results can be enhanced, thereby improving the accuracy of nursing home admission demand prediction.
[0061] In step S101 of some embodiments, the original historical occupancy data of nursing homes in the target city and the original nursing home demand exogenous variables of the target city can be obtained; the original historical occupancy data are cleaned to obtain the cleaned original historical occupancy data; the original nursing home demand exogenous variables are cleaned to obtain the cleaned original nursing home demand exogenous variables; the cleaned original historical occupancy data and the cleaned original nursing home demand exogenous variables are time-aligned to obtain the historical occupancy data and the nursing home demand exogenous variables.
[0062] In order to predict the nursing home occupancy demand in the target city, it is necessary to collect the original historical occupancy data of the nursing homes in the target city, as well as the original nursing home demand exogenous variables in the target city.
[0063] Taking into account the possible quality defects of the original historical admission data and the original nursing home demand exogenous variables, the original historical admission data can be cleaned to obtain the cleaned original historical admission data, and the original nursing home demand exogenous variables can be cleaned to obtain the cleaned original nursing home demand exogenous variables.
[0064] In order to facilitate the prediction of the trained ARIMA model, the cleaned original historical occupancy data and the cleaned original nursing home demand exogenous variable can be time-aligned to ensure that the timestamps (such as date or time) of the historical occupancy data and the nursing home demand exogenous variable are consistent to avoid analysis errors caused by time misalignment. For example, the population mortality rate and the number of occupants should have the same timestamp:
[0065] date Number of guests mortality rate January xx a 5％ February b 5％ March xx c 5％ April xx d 5％
[0066] In this way, the quality and availability of historical occupancy data and exogenous variables of nursing home demand can be ensured, facilitating the prediction of the trained ARIMA model.
[0067] In some embodiments, the original historical check-in data is cleaned to obtain the cleaned original historical check-in data, and the missing value filling process is performed on the original historical check-in data to obtain the filled original historical check-in data; the outlier correction process is performed on the filled original historical check-in data to obtain the corrected original historical check-in data; the corrected original historical check-in data is standardized to obtain the cleaned original historical check-in data.
[0068] Exemplarily, the cleaning process mainly includes missing value filling process, outlier correction process and standardization process.
[0069] The original historical occupancy data may be processed for missing values, for example, by using mean, median filling, or interpolation to fill in missing values, and obtain the filled original historical occupancy data.
[0070] The original historical occupancy data after filling can be corrected for outliers, for example, outliers in the original historical occupancy data after filling can be detected by means of box plots, standard deviations, etc., and the outliers can be adjusted to a normal range to obtain corrected original historical occupancy data. The outliers can also be directly deleted.
[0071] The corrected original historical occupancy data may be standardized, for example, the corrected original historical occupancy data may be normalized to obtain cleaned original historical occupancy data.
[0072] In this way, through the cleaning operation, the quality of the cleaned original historical occupancy data is improved.
[0073] Similarly, the original nursing home demand exogenous variables are cleaned to obtain the cleaned original nursing home demand exogenous variables, thereby improving the quality of the cleaned original nursing home demand exogenous variables.
[0074] In step S103 of some embodiments, the nursing home admission demand data may be weighted averaged to obtain the target nursing home admission demand data.
[0075] The admission demand data of multiple nursing homes can be processed by weighted average to achieve integration of the admission demand data of multiple nursing homes, so as to obtain more robust admission demand data of target nursing homes.
[0076] For example, a weight may be set for each trained ARIMA model according to the training effect and verification effect of each ARIMA model, so as to perform weighted averaging processing on the admission demand data of multiple nursing homes.
[0077] In this way, by combining the results of multiple models, it is possible to make up for the errors or deviations that may exist in a single model, thereby reducing the errors or deviations of a single model. This weighted average method can also make full use of the advantages of multiple models, which will improve the accuracy and stability of the overall prediction, and can flexibly adjust the weights according to the performance of different models to further optimize the prediction effect.
[0078] See also Figure 2 In some embodiments, before step S101, including but not limited to steps S201 to S202.
[0079] Step S201, obtaining historical occupancy data samples of nursing homes in at least two cities, and exogenous variable samples of nursing home demand in at least two cities, wherein the exogenous variable samples of nursing home demand include population mortality data samples and feature label samples of the elderly;
[0080] Step S202: training at least two ARIMA models according to the historical admission data samples and the nursing home demand exogenous variable samples to obtain at least two trained ARIMA models, wherein the original parameters of any two ARIMA models are different.
[0081] See also Figure 3 , Figure 3 The following is a diagram of the training architecture of multiple ARIMA models (such as ARIMA1, ARIMA2, and ARIMA3). Figure 3 As shown, in order to train multiple ARIMA models, it is necessary to collect original historical admission data samples of nursing homes in multiple cities, as well as original nursing home demand exogenous variable samples in multiple cities.
[0082] Similarly, the original historical admission data samples are cleaned, the original nursing home demand exogenous variable samples are cleaned, and the cleaned original historical admission data samples and the cleaned nursing home demand exogenous variable samples are time-aligned to obtain historical admission data samples and nursing home demand exogenous variable samples, thereby ensuring the quality and availability of historical admission data samples and nursing home demand exogenous variable samples.
[0083] The historical admission data samples and the nursing home demand exogenous variable samples can be used as training sets to train multiple ARIMA models to obtain at least two trained ARIMA models, wherein the original parameters of any two ARIMA models are different.
[0084] In this way, on the one hand, by using data samples from multiple cities, the adaptability of each ARIMA model to the elderly care needs of different regions can be increased, and each trained ARIMA model can have better generalization ability when facing new cities or regions, thereby improving the robustness of each trained ARIMA model; and by using data samples from multiple cities, the prediction bias caused by the particularity of a single city can be reduced, making the model more stable and reliable when facing the needs of different cities or regions. In addition, by incorporating the exogenous variable samples of nursing home demand into the training of each ARIMA model, it can also help each ARIMA model capture the various factors that affect the demand for nursing home admission, and better understand the various factors that affect the demand for admission, so that each trained ARIMA model is more sensitive to changes in the external environment, so that it can maintain a high prediction accuracy. In this way, the stability, anti-interference and prediction accuracy of each trained ARIMA model are improved.
[0085] On the other hand, by training multiple ARIMA models with different parameters, the advantages of each ARIMA model can be fully utilized (for example, one ARIMA model may perform better in dealing with seasonal fluctuations, while another ARIMA model may be more suitable for capturing trend changes), so that various change patterns can be better handled, the risk of overfitting or underfitting can be reduced, and the prediction accuracy can be improved.
[0086] After step S201 in some embodiments, original parameters may be set according to the autocorrelation function graph and the partial autocorrelation function graph, wherein the original parameters include the difference order, the number of autoregressive terms, and the number of moving average terms.
[0087] The original parameters of each ARIMA model can be determined through the autocorrelation function (ACF) plot and the partial autocorrelation function (PACF) plot.
[0088] Among them, the autocorrelation function (ACF) represents the correlation between the series and its own different lag periods. The number of moving average items can be determined through the ACF graph.
[0089] The partial autocorrelation function (PACF) represents the correlation between the series and itself at different lags, but eliminates the influence of the intermediate lags. The number of autoregressive terms can be determined through the PACF plot.
[0090] The trend of the time series is analyzed through the ACF graph and PACF graph to determine whether the time series needs to be differentiated, so as to select the appropriate difference order.
[0091] In this way, different original parameters are set for different ARIMA models, so that different ARIMA models can better capture different time series characteristics in the historical admission data samples and the nursing home demand exogenous variable samples.
[0092] In step S202 of some embodiments, historical admission data samples and nursing home demand exogenous variable samples can be input into at least two ARIMA models for iterative training to adjust the difference order, autoregressive terms, and moving average terms of each ARIMA model until the number of iterations is greater than a preset threshold, and the training is stopped to obtain at least two trained ARIMA models.
[0093] Historical admission data samples and nursing home demand exogenous variable samples can be input into multiple ARIMA models for iterative training to adjust the difference order, autoregressive terms and moving average terms of each ARIMA model, so that each ARIMA model can effectively learn to accurately predict future time series until the number of iterations exceeds the preset threshold, and the training is stopped to obtain at least two trained ARIMA models.
[0094] In this way, by training multiple ARIMA models with different parameter configurations through historical nursing home admission data samples and exogenous variable samples in multiple cities, the robustness, stability and prediction accuracy of each trained ARIMA model can be enhanced, and it can better adapt to the changing trends of needs in different cities.
[0095] After obtaining multiple trained ARIMA models, they can be applied to the task of predicting nursing home admission demand in the target city.
[0096] Among them, before each trained ARIMA model is applied to the nursing home admission demand prediction task of the target city, the prediction accuracy of each trained ARIMA model can also be evaluated, for example, by calculating the mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE) and other indicators to evaluate the prediction accuracy. Only when the prediction accuracy is greater than a preset threshold, each trained ARIMA model is applied to the nursing home admission demand prediction task of the target city.
[0097] It is understandable that new city historical occupancy data samples and exogenous variables (such as the elderly’s acceptance of nursing homes) can be continuously added to the ARIMA model, so that the ARIMA model can be continuously optimized and adjusted to cope with new challenges and changes.
[0098] In summary, the nursing home admission demand prediction method provided in the above embodiment can significantly improve the efficiency, accuracy and reliability of nursing home admission demand prediction, promote the intelligence and informatization of elderly care services, and improve the overall quality of elderly care services.
[0099] See also Figure 4 , Figure 4It is a schematic block diagram of a nursing home admission demand prediction device provided in an embodiment of the present application. The nursing home admission demand prediction device can be configured in a computer device to execute the aforementioned nursing home admission demand prediction method.
[0100] like Figure 4 As shown, the nursing home occupancy demand prediction device 400 includes: a data acquisition module 401, an occupancy demand prediction module 402 and an occupancy demand integration module 403.
[0101] A data acquisition module 401 is used to acquire historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly;
[0102] The occupancy demand forecasting module 402 is used to forecast the occupancy demand of nursing homes according to the historical occupancy data and the exogenous variables of nursing home demand by using at least two trained ARIMA models, and obtain the occupancy demand data of at least two nursing homes in the target city;
[0103] The occupancy demand integration module 403 is used to integrate the occupancy demand data of the nursing homes to obtain the occupancy demand data of the target nursing homes in the target city.
[0104] In one embodiment, the occupancy demand integration module 403 is further used to:
[0105] The nursing home admission demand data is weighted averaged to obtain the target nursing home admission demand data.
[0106] In one embodiment, the nursing home admission demand prediction device 400 further includes a model training module, which is used to:
[0107] Obtain historical occupancy data samples of nursing homes in at least two cities, and exogenous variable samples of nursing home demand in at least two of the cities, wherein the exogenous variable samples of nursing home demand include population mortality data samples and feature label samples of the elderly;
[0108] According to the historical occupancy data samples and the nursing home demand exogenous variable samples, at least two ARIMA models are trained to obtain at least two trained ARIMA models, wherein original parameters of any two ARIMA models are different.
[0109] In one embodiment, the nursing home occupancy demand prediction device 400 further includes a parameter setting module, and the parameter setting module is used to:
[0110] The original parameters are set according to the autocorrelation function graph and the partial autocorrelation function graph, wherein the original parameters include the difference order, the number of autoregressive terms and the number of moving average terms.
[0111] In one embodiment, the model training module is further used to:
[0112] The historical occupancy data samples and the nursing home demand exogenous variable samples are input into at least two of the ARIMA models for iterative training to adjust the difference order, the number of autoregressive terms and the number of moving average terms of each ARIMA model until the number of iterations is greater than a preset threshold, and the training is stopped to obtain at least two trained ARIMA models.
[0113] In one embodiment, the data acquisition module 401 is further used for:
[0114] Obtaining original historical occupancy data of nursing homes in the target city and original nursing home demand exogenous variables in the target city;
[0115] Cleaning the original historical occupancy data to obtain cleaned original historical occupancy data;
[0116] Cleaning the original nursing home demand exogenous variable to obtain a cleaned original nursing home demand exogenous variable;
[0117] The cleaned original historical occupancy data and the cleaned original nursing home demand exogenous variables are time aligned to obtain the historical occupancy data and the nursing home demand exogenous variables.
[0118] In one embodiment, the data acquisition module 401 is further used for:
[0119] Performing missing value filling processing on the original historical occupancy data to obtain filled original historical occupancy data;
[0120] Performing outlier correction processing on the padded original historical occupancy data to obtain corrected original historical occupancy data;
[0121] The corrected original historical occupancy data is standardized to obtain the cleaned original historical occupancy data.
[0122] Among them, each module in the above-mentioned nursing home occupancy demand prediction device 400 corresponds to each step in the above-mentioned nursing home occupancy demand prediction method embodiment, and its functions and implementation processes will not be repeated here one by one.
[0123] The nursing home admission demand prediction device 400 can execute the nursing home admission demand prediction method provided in the embodiment of the present application, and therefore can achieve the beneficial effects that can be achieved by the nursing home admission demand prediction method provided in the embodiment of the present application. Please see the previous embodiments for details and will not be repeated here.
[0124] The method and apparatus of the present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.
[0125] Exemplarily, the above method and apparatus may be implemented in the form of a computer program. The computer program may be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.
[0126] See also Figure 5 , Figure 5 It is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application.
[0127] See also Figure 5 The computer device includes a processor and a memory connected via a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.
[0128] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.
[0129] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any one of the nursing home admission demand prediction methods.
[0130] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0131] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:
[0132] Obtaining historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly;
[0133] Using at least two trained ARIMA models, forecasting nursing home occupancy demand based on the historical occupancy data and the nursing home demand exogenous variables, to obtain occupancy demand data for at least two nursing homes in the target city;
[0134] The admission demand data are integrated and processed to obtain the admission demand data of the target nursing home in the target city.
[0135] In one embodiment, when the processor implements the integrated processing of the admission demand data to obtain the admission demand data of the target nursing home in the target city, it is used to implement:
[0136] The nursing home admission demand data is weighted averaged to obtain the target nursing home admission demand data.
[0137] In one embodiment, before obtaining the historical occupancy data of nursing homes in the target city and the exogenous variables of nursing home demand in the target city, the processor is further configured to implement:
[0138] Obtain historical occupancy data samples of nursing homes in at least two cities, and exogenous variable samples of nursing home demand in at least two of the cities, wherein the exogenous variable samples of nursing home demand include population mortality data samples and feature label samples of the elderly;
[0139] According to the historical occupancy data samples and the nursing home demand exogenous variable samples, at least two ARIMA models are trained to obtain at least two trained ARIMA models, wherein original parameters of any two ARIMA models are different.
[0140] In one embodiment, after obtaining the historical occupancy data samples of nursing homes in at least two cities and the exogenous variable samples of nursing home demand in at least two of the cities, the processor is further configured to implement:
[0141] The original parameters are set according to the autocorrelation function graph and the partial autocorrelation function graph, wherein the original parameters include the difference order, the number of autoregressive terms and the number of moving average terms.
[0142] In one embodiment, when the processor implements the training of at least two ARIMA models according to the historical occupancy data samples and the nursing home demand exogenous variable samples to obtain at least two trained ARIMA models, it is used to implement:
[0143] The historical occupancy data samples and the nursing home demand exogenous variable samples are input into at least two of the ARIMA models for iterative training to adjust the difference order, the number of autoregressive terms and the number of moving average terms of each ARIMA model until the number of iterations is greater than a preset threshold, and the training is stopped to obtain at least two trained ARIMA models.
[0144] In one embodiment, when the processor implements the acquisition of the historical occupancy data of the nursing homes in the target city and the exogenous variables of the nursing home demand in the target city, it is used to implement:
[0145] Obtaining original historical occupancy data of nursing homes in the target city and original nursing home demand exogenous variables in the target city;
[0146] Cleaning the original historical occupancy data to obtain cleaned original historical occupancy data;
[0147] Cleaning the original nursing home demand exogenous variable to obtain a cleaned original nursing home demand exogenous variable;
[0148] The cleaned original historical occupancy data and the cleaned original nursing home demand exogenous variables are time aligned to obtain the historical occupancy data and the nursing home demand exogenous variables.
[0149] In one embodiment, when the processor implements the cleaning process on the original historical occupancy data to obtain the cleaned original historical occupancy data, it is used to implement:
[0150] Performing missing value filling processing on the original historical occupancy data to obtain filled original historical occupancy data;
[0151] Performing outlier correction processing on the padded original historical occupancy data to obtain corrected original historical occupancy data;
[0152] The corrected original historical occupancy data is standardized to obtain the cleaned original historical occupancy data.
[0153] The computer device can execute the nursing home admission demand prediction method provided in the embodiment of the present application, and therefore can achieve the beneficial effects that can be achieved by the nursing home admission demand prediction method provided in the embodiment of the present application. Please refer to the previous embodiments for details and will not be repeated here.
[0154] An embodiment of the present application also provides a computer-readable storage medium.
[0155] A computer program is stored on the computer-readable storage medium of the present application, and when the computer program is executed by a processor, the steps of the above-mentioned method for predicting nursing home admission demand are implemented.
[0156] The computer-readable storage medium may be an internal storage unit of the nursing home admission demand prediction device or computer device described in the aforementioned embodiment, such as a hard disk or memory of the nursing home admission demand prediction device or computer device. The computer-readable storage medium may also be an external storage device of the nursing home admission demand prediction device or computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital card (Secure Digital Card, SD Card), a flash card (Flash Card), etc., equipped on the nursing home admission demand prediction device or computer device.
[0157] Since the computer program stored in the computer-readable storage medium can execute any one of the nursing home admission demand prediction methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the nursing home admission demand prediction methods provided in the embodiments of the present application can be achieved. Please see the previous embodiments for details and will not be repeated here.
[0158] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.
[0159] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.
[0160] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.
[0161] The above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and these modifications or substitutions should be included in the protection scope of the present application.
Claims
1. A method for predicting nursing home occupancy demand, characterized in that: The method comprises: Obtaining historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly; Using at least two trained ARIMA models, forecasting nursing home occupancy demand based on the historical occupancy data and the nursing home demand exogenous variables, to obtain occupancy demand data for at least two nursing homes in the target city; The admission demand data are integrated and processed to obtain the admission demand data of the target nursing home in the target city.
2. The method according to claim 1, characterized in that The step of integrating the admission demand data to obtain the admission demand data of the target nursing home in the target city includes: The nursing home admission demand data is weighted averaged to obtain the target nursing home admission demand data.
3. The method according to claim 1, characterized in that Before obtaining the historical occupancy data of nursing homes in the target city and the exogenous variables of nursing home demand in the target city, the method further includes: Obtain historical occupancy data samples of nursing homes in at least two cities, and exogenous variable samples of nursing home demand in at least two of the cities, wherein the exogenous variable samples of nursing home demand include population mortality data samples and feature label samples of the elderly; According to the historical occupancy data samples and the nursing home demand exogenous variable samples, at least two ARIMA models are trained to obtain at least two trained ARIMA models, wherein original parameters of any two ARIMA models are different.
4. The method according to claim 3, characterized in that After obtaining historical nursing home occupancy data samples of at least two cities and nursing home demand exogenous variable samples of at least two cities, the method further includes: The original parameters are set according to the autocorrelation function graph and the partial autocorrelation function graph, wherein the original parameters include the difference order, the number of autoregressive terms and the number of moving average terms.
5. The method according to claim 4, characterized in that The at least two ARIMA models are trained according to the historical admission data samples and the nursing home demand exogenous variable samples to obtain at least two trained ARIMA models, including: The historical occupancy data samples and the nursing home demand exogenous variable samples are input into at least two of the ARIMA models for iterative training to adjust the difference order, the number of autoregressive terms and the number of moving average terms of each ARIMA model until the number of iterations is greater than a preset threshold, and the training is stopped to obtain at least two trained ARIMA models.
6. The method according to claim 1, characterized in that The acquisition of historical occupancy data of nursing homes in the target city and exogenous variables of nursing home demand in the target city include: Obtaining original historical occupancy data of nursing homes in the target city and original nursing home demand exogenous variables in the target city; Cleaning the original historical occupancy data to obtain cleaned original historical occupancy data; Cleaning the original nursing home demand exogenous variable to obtain a cleaned original nursing home demand exogenous variable; The cleaned original historical occupancy data and the cleaned original nursing home demand exogenous variables are time aligned to obtain the historical occupancy data and the nursing home demand exogenous variables.
7. The method according to claim 6, characterized in that The cleaning process of the original historical check-in data to obtain the cleaned original historical check-in data includes: Performing missing value filling processing on the original historical occupancy data to obtain filled original historical occupancy data; Performing outlier correction processing on the padded original historical occupancy data to obtain corrected original historical occupancy data; The corrected original historical occupancy data is standardized to obtain the cleaned original historical occupancy data.
8. A device for predicting the demand for admission to a nursing home, characterized in that: The device comprises: A data acquisition module, used to acquire historical occupancy data of nursing homes in a target city and exogenous variables of nursing home demand in the target city, wherein the exogenous variables of nursing home demand include population mortality data and characteristic labels of the elderly; An occupancy demand forecasting module is used to forecast the occupancy demand of nursing homes according to the historical occupancy data and the exogenous variables of nursing home demand by using at least two trained ARIMA models, and obtain the occupancy demand data of at least two nursing homes in the target city; The occupancy demand integration module is used to integrate the occupancy demand data of the nursing home to obtain the occupancy demand data of the target nursing home in the target city.
9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the nursing home occupancy demand prediction method as described in any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for predicting nursing home occupancy demand according to any one of claims 1 to 7 is implemented.