Machine learning-based home-based care service matching processing method and related equipment

Abstract

The invention belongs to the field of medical health, provides a home-based care service matching processing method and device based on machine learning and related equipment, and aims to solve the problem of low home-based care service matching efficiency in the prior art. Data fusion is carried out on health state information data, self-care ability information data, living state information data and psychological health information data of different dimensions, then a user portrait and a customer category corresponding to a home-based care customer are determined according to the fused data, and then a preset prediction model based on machine learning is used to predict the home-based care customer. And according to the user portrait and the customer category, predicting a home-based care service demand corresponding to the home-based care customer, and finally determining a target home-based care service item corresponding to the home-based care service demand, and matching the target home-based care service item to the home-based care customer, thereby realizing multi-dimensional information based on the multi-dimensional information of the home-based care customer. And based on prediction of machine learning, the accuracy of home-based care service matching is improved.

Description

Technical Field

[0001] The present application relates to the field of medical health, and in particular to a method, device, computer equipment and computer-readable storage medium for matching home-based elderly care services based on machine learning. Background Art

[0002] More and more elderly people are living at home, and the most prominent issue is how to match the supply of elderly care services with the diverse elderly care needs of the elderly population.

[0003] For the elderly, home-based care requires more than just basic living care, but also personalized care and respect. However, the reality is that there is a huge gap between service supply and demand. Elderly people in different situations have very different needs for elderly care. For example, some need professional medical care, some desire spiritual companionship, and some want to continue to participate in social activities. This diversity has brought considerable challenges to the provision of elderly care services.

[0004] In traditional technology, the personalized needs of the elderly are generally understood through manual follow-up. However, this way of understanding consumes huge manpower and material resources and is inefficient. There is a lack of effective technical means to automatically and accurately match the personalized needs of the elderly with service supply. This not only affects the quality of life of the elderly, but also restricts the effective development of the elderly care service industry.

[0005] Therefore, how to use technical means to accurately predict the personalized needs of the elderly and automatically provide precise matching of elderly care services has become a technical problem that the elderly care service industry urgently needs to solve. Summary of the invention

[0006] The present application provides a method, apparatus, computer device and computer-readable storage medium for matching home-based elderly care services based on machine learning, which can solve the technical problem of low efficiency in matching home-based elderly care services in traditional technologies.

[0007] In the first aspect, the present application provides a method for matching and processing home-based elderly care services based on machine learning, including: obtaining physical and mental health information data of home-based elderly care clients in different dimensions, the physical and mental health information data including health status information data, self-care ability information data, living status information data, and mental health information data; fusing the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data; determining a user portrait corresponding to the home-based elderly care client based on the fused data; determining a customer category corresponding to the home-based elderly care client based on the user portrait; predicting the home-based elderly care service demand corresponding to the home-based elderly care client based on a preset prediction model for machine learning and based on the user portrait and the customer category; matching the target home-based elderly care service project corresponding to the home-based elderly care service demand based on a preset home-based elderly care service project database, and matching the target home-based elderly care service project to the home-based elderly care client.

[0008] In the second aspect, the present application provides a home-based elderly care service matching and processing device based on machine learning, including: a first acquisition unit, used to acquire physical and mental health information data of different dimensions of home-based elderly care customers, the physical and mental health information data including health status information data, self-care ability information data, living status information data, and mental health information data; a data fusion unit, used to fuse the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data; a first determination unit, used to determine the user portrait corresponding to the home-based elderly care customer based on the fused data; a second determination unit, used to determine the customer category corresponding to the home-based elderly care customer based on the user portrait; a first prediction unit, used to predict the home-based elderly care service demand corresponding to the home-based elderly care customer based on a preset prediction model of machine learning and according to the user portrait and the customer category; a first matching unit, used to match the target home-based elderly care service project corresponding to the home-based elderly care service demand based on a preset home-based elderly care service project database, and match the target home-based elderly care service project to the home-based elderly care customer.

[0009] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the machine learning-based home-based elderly care service matching processing method.

[0010] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the machine learning-based home-based elderly care service matching processing method.

[0011] The present application provides a method, device, computer equipment and computer-readable storage medium for matching home-based elderly care services based on machine learning. The method obtains physical and mental health information data of home-based elderly care clients in different dimensions, and fuses the physical and mental health information data of different dimensions, and then determines the user portrait and customer category corresponding to the home-based elderly care client based on the fused data, and then predicts the home-based elderly care service needs corresponding to the home-based elderly care client based on a preset prediction model of machine learning and according to the user portrait and customer category, and finally determines the target home-based elderly care service items corresponding to the home-based elderly care service needs, and matches the target home-based elderly care service items to the home-based elderly care client, so as to classify the home-based elderly care clients according to their multi-dimensional information, and then realize personalized service recommendations for the home-based elderly care clients based on the user portrait and customer category corresponding to the home-based elderly care client and based on the preset prediction model of machine learning, thereby improving the accuracy of home-based elderly care service matching and improving the quality and efficiency of home-based elderly care services. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0013] Figure 1 A flowchart of a method for matching home-based elderly care services based on machine learning provided in an embodiment of the present application;

[0014] Figure 2 A schematic diagram of the first sub-process of the home-based elderly care service matching processing method based on machine learning provided in an embodiment of the present application;

[0015] Figure 3 A schematic diagram of a second sub-flow of the method for matching home-based elderly care services based on machine learning provided in an embodiment of the present application;

[0016] Figure 4 A schematic block diagram of a device for matching and processing home-based elderly care services based on machine learning provided in an embodiment of the present application;

[0017] Figure 5 A schematic block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0020] The embodiments of the present application provide a method for matching home-based elderly care services based on machine learning. The method can be applied to computer devices including but not limited to smart phones, tablet computers, laptop computers, desktop computers, servers, etc., and can be used to match elderly care services for the elderly in fields including but not limited to the medical and health fields.

[0021] In the face of the technical problem of low efficiency in matching home-based elderly care services in traditional technologies, the inventors proposed a method for matching home-based elderly care services based on machine learning in the embodiments of the present application. The core idea of ​​the embodiments of the present application is: using data fusion technology to comprehensively consider the five dimensions of age, health status, self-care ability, living status, and psychological emotions of the elderly living at home to form a comprehensive user portrait and classify the customers. Then, based on the machine learning prediction model, the service needs of the elderly living at home are predicted according to the customer categories, and the service needs are pushed to the elderly living at home through various media methods. The elderly living at home can be grouped according to the customer labels corresponding to the elderly living at home, to achieve personalized service recommendations, and to improve the accuracy of home-based elderly care matching, so as to improve the quality and efficiency of home-based elderly care services.

[0022] Some embodiments of the present application are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0023] See also Figure 1 , Figure 1 The flowchart of the method for matching home-based elderly care services based on machine learning provided in the embodiment of the present application is as follows. Figure 1 As shown, the method includes but is not limited to the following steps S11-S16:

[0024] S11. Obtaining physical and mental health information data of different dimensions of home-based elderly care customers, wherein the physical and mental health information data includes health status information data, self-care ability information data, living status information data, and mental health information data.

[0025] Explanatory, home-based elderly care clients have needs for both physical care and psychological and spiritual care. Therefore, it is necessary to assess the physical and mental health status of home-based elderly care clients, quantify the physical and mental health status into corresponding specific information, and obtain physical and mental health information data of home-based elderly care clients in different dimensions. The physical and mental health information data include health status information data, self-care ability information data, living status information data, and mental health information data. Among them, the health status information data indicates whether the home-based elderly care client has physiological or pathological diseases, the self-care ability information data indicates whether the home-based elderly care client can take care of himself, the living status information data indicates whether the home-based elderly care client lives alone or with his family, and the mental health information data indicates information on the psychological, mental, emotional and other states of the home-based elderly care client.

[0026] S12, fusing the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data.

[0027] Explanatoryly, data fusion refers to converting data from different sources, formats, structures, and types into a unified format, structure, and type, and merging and mapping multiple data into a unified data set. Data fusion includes but is not limited to data preprocessing and data cleaning. Its purpose is to improve the uniformity, completeness, accuracy, and availability of data by integrating diverse data. Data fusion methods include but are not limited to ETL (extraction, transformation, loading) processes, data integration and federation, data mining, and machine learning. Among them, the ETL (extraction, transformation, loading) process is a process that includes extracting data from different data sources, converting the data into a unified format or structure, and finally loading it into the target system; data integration and federation is the integration of data from multiple data sources into a unified data model so that the data can be queried and analyzed together; data mining and machine learning technology uses data mining and machine learning algorithms to fuse data, identify patterns and relationships, and thus generate higher quality fused data.

[0028] Based on the above conception and description, the health status information data, self-care ability information data, living status information data, and mental health information data from different sources, different formats, different structures, and different dimensions are fused to obtain fused data, which facilitates comprehensive processing and analysis of the fused data, and obtains the most accurate possible physical and mental health information of home-based elderly care clients in different dimensions, thereby improving the accuracy of home-based elderly care service matching.

[0029] S13. Determine the user portrait corresponding to the home-based elderly care customer based on the fused data.

[0030] Explanatoryly, user profiling, also known as user persona or customer profiling, is the process of dividing users into different groups by collecting and analyzing personal information, behavioral data, etc. of different dimensions of users, and providing a detailed description and analysis of each group.

[0031] Based on the above ideas and descriptions, according to the fusion data, information from different dimensions is integrated to determine the user portrait corresponding to the home-based elderly care customer. The content of the user portrait includes but is not limited to the specific content of health status, self-care ability, living status, and mental health.

[0032] S14. Determine the customer category corresponding to the home-based elderly care customer based on the user portrait.

[0033] Explanatory wise, based on user portraits, unsupervised aggregation algorithms including but not limited to K-Means, Mini-Batch K-Means, Affinity Propagation, and Mean Shift are used to determine the customer categories corresponding to home-based elderly care customers. Customer categories include but are not limited to health categories, disability categories, and dementia categories. User portraits are multi-dimensional label representations of users, and customer categories are category representations of the comprehensive content of customers.

[0034] Exemplarily, when an unsupervised aggregation algorithm is used to determine the customer category corresponding to home-based elderly care customers, the portrait data is mapped to a high-dimensional coordinate system scatter plot based on the user portrait. A group of people who are close to each other are aggregated into one group. The density of core points in the group must reach the threshold set by the algorithm, and the iteration of the radius is calculated using the Euclidean distance to determine the customer category corresponding to the home-based elderly care customers.

[0035] S15. Based on a preset prediction model of machine learning, and according to the user portrait and the customer category, predict the home-based elderly care service needs corresponding to the home-based elderly care customers.

[0036] Explanatoryly, a pre-set prediction model, i.e., a preset prediction model, refers to a model that predicts the home-based elderly care service needs corresponding to home-based elderly care customers based on user portraits and customer categories. The preset prediction model includes but is not limited to a neural network model, such as a BP (back propagation) neural network.

[0037] According to the above ideas and settings, based on the preset prediction model of machine learning, and according to user portraits and customer categories, the home-based elderly care service needs corresponding to home-based elderly care customers are predicted. The home-based elderly care service needs include but are not limited to medical care, food, housing, transportation, health, care, entertainment, and nursing.

[0038] S16. Based on a preset home-based elderly care service project database, match the target home-based elderly care service project corresponding to the home-based elderly care service demand, and match the target home-based elderly care service project to the home-based elderly care customer.

[0039] Explanatoryly, based on a preset home-based elderly care service project database and according to the home-based elderly care service needs corresponding to the home-based elderly care customers, target home-based elderly care service projects corresponding to the home-based elderly care service needs are automatically matched. The target home-based elderly care service projects may be project plans, which represent a collection of service projects including but not limited to medical care, food, housing, transportation, health, nursing, entertainment, and care, so as to customize the elderly care services for home-based elderly care customers, and match the target home-based elderly care service projects to the home-based elderly care customers, thereby realizing personalized service recommendations for home-based elderly care customers.

[0040] The embodiment of the present application obtains physical and mental health information data of different dimensions of home-based elderly care clients, and fuses the health status information data, self-care ability information data, living status information data, and mental health information data included in the physical and mental health information data, and then determines the user portrait and customer category corresponding to the home-based elderly care client based on the fused data, and then predicts the home-based elderly care service needs corresponding to the home-based elderly care client based on a preset prediction model of machine learning and according to the user portrait and customer category, and finally determines the target home-based elderly care service items corresponding to the home-based elderly care service needs, and matches the target home-based elderly care service items to the home-based elderly care client, so as to classify the home-based elderly care clients according to their multi-dimensional information, and then realize personalized service recommendations for the home-based elderly care clients based on the user portrait and customer category corresponding to the home-based elderly care clients and based on the preset prediction model of machine learning, thereby improving the accuracy of home-based elderly care service matching and improving the quality and efficiency of home-based elderly care services.

[0041] In one embodiment, the physical and mental health information data of home-based elderly care clients in different dimensions are obtained, including:

[0042] Determine the medical information of home care clients;

[0043] Based on a preset wearable smart device, obtaining health information of the home-based elderly care customer;

[0044] The health status of the home-based elderly care client is classified according to the medical information and the health information to obtain health status information data of the home-based elderly care client.

[0045] Explanatory, with the permission of relevant institutions or personnel, obtain the medical information of home-based elderly care clients, the medical information includes but is not limited to the physical examination reports, test reports, prescriptions, medications, and medical records of home-based elderly care clients, and based on preset wearable smart devices, the wearable smart devices include but are not limited to smart watches, smart gloves, and smart bracelets, collect and obtain the health information of home-based elderly care clients, the health information includes but is not limited to heart rate, blood pressure, blood sugar, uric acid, blood oxygen saturation, body temperature, stress, mood, sleep, exercise, diet and other health indicators, and then automatically and comprehensively determine the health status of home-based elderly care clients based on machine learning based on the medical information and health information, thereby classifying the health status of home-based elderly care clients and obtaining the health status information data of home-based elderly care clients, the health status information includes but is not limited to health, minor illness, serious illness, and disability.

[0046] The embodiment of the present application determines the medical information and health information of the home-based elderly care clients, and classifies the health status of the home-based elderly care clients according to the medical information and health information. It can evaluate the health status of the home-based elderly care clients based on information from multiple dimensions, thereby obtaining more accurate health status information data, and then improve the accuracy of home-based elderly care service matching based on the multi-dimensional information of the home-based elderly care clients, thereby improving the quality and efficiency of home-based elderly care services.

[0047] In one embodiment, determining medical information of a home-based elderly care client includes:

[0048] Obtain medical image data for home-based elderly care clients;

[0049] Based on a preset OCR recognition method, the medical image data is recognized to obtain the medical text data of the home-based elderly care customer;

[0050] The medical text data is structured and preset target medical content is extracted to obtain the medical information of the home-based elderly care client.

[0051] Explanatory speaking, medical information refers to the information on medical treatment, testing and inspection of home-based elderly care clients during physical examinations, medical consultations and other medical treatments, which is in image format in many cases. Thus, the medical image data of home-based elderly care clients are obtained, and based on a preset OCR recognition method, the medical image data are identified, thereby converting the image into text to obtain the medical text data of the home-based elderly care clients, and then the various medical text data with inconsistent formats are uniformly structured, and then the preset target medical content is extracted accordingly, such as extracting related attributes such as diseases, medications, abnormal indicators, surgeries, onset time, current control status, etc., to obtain the medical information of home-based elderly care clients.

[0052] The embodiment of the present application identifies medical image data based on a preset OCR recognition method and structures the medical text data, thereby obtaining the medical information of home-based elderly care clients based on a standardized data format, and can combine a variety of medical data to obtain multi-dimensional medical information of home-based elderly care clients, obtain more accurate medical status of home-based elderly care clients, and then evaluate the health status of home-based elderly care clients based on multi-dimensional information, improve the accuracy of home-based elderly care service matching, and improve the quality and efficiency of home-based elderly care services.

[0053] See also Figure 2 , Figure 2 This is a schematic diagram of the first sub-flow of the home-based elderly care service matching processing method based on machine learning provided in the embodiment of the present application. Figure 2 As shown, in this embodiment, the physical and mental health information data of home-based elderly care customers in different dimensions are obtained, including:

[0054] S21, collecting behavioral images of home-based elderly care customers based on preset home cameras;

[0055] S22, acquiring the behavior image based on Internet of Things communication, and identifying the behavior image based on a preset cross-modal learning model to obtain a behavior information text;

[0056] S23, based on a preset intelligent client terminal, obtaining a call voice corresponding to a call with the home-based elderly care client;

[0057] S24, identifying the call text corresponding to the call voice based on a preset automatic speech recognition model;

[0058] S25. Determine the self-care ability information data of the home-based elderly care client based on the medical text data, the behavior information text and the call text.

[0059] Explanativiously, a cross-modal learning model is pre-set, i.e., a preset cross-modal learning model. The preset cross-modal learning model represents a model that learns from data of one modality and applies its knowledge to data of another modality. Cross-modal learning includes but is not limited to learning from image data for text generation. The preset cross-modal learning model includes but is not limited to a Transformer model, a combination of a convolutional neural network (CNN) and a recurrent neural network (RNN), and cross-modal contrastive learning.

[0060] The automatic speech recognition model is pre-set, that is, the automatic speech recognition model is preset. The automatic speech recognition model (Automatic Speech Recognition, referred to as ASR) is a technology for converting human speech into text.

[0061] According to the above concept and setting, with the popularization and development of the Internet of Things and video surveillance, it is also convenient to collect and obtain relevant information of home-based elderly care customers through video surveillance with the permission of relevant institutions or personnel, and then use the relevant information to analyze the status of home-based elderly care customers and provide corresponding services for them in a targeted manner. Therefore, based on the preset home camera, the behavior image of the home-based elderly care customer is collected, and based on the Internet of Things communication, the behavior image is obtained, and then based on cross-modal learning, the behavior image is recognized to obtain the behavior information text, and the behavior information text represents the text information describing the behavior content of the home-based elderly care customer based on the behavior image. At the same time, based on the preset smart customer terminal, the preset smart customer terminal includes but is not limited to video surveillance, smart phones, smart bracelets, smart watches, smart speakers, and smart home devices, the call voice corresponding to the call with the home-based elderly care customer is obtained, and based on the preset automatic speech recognition model, the call text corresponding to the call voice is recognized, and then according to the above medical text data, behavior information text and call text, the home-based elderly care customer's daily living activity ability information, social participation ability information, basic motor ability information, cognitive ability information are extracted to obtain the self-care ability information data of the home-based elderly care customer.

[0062] The embodiment of the present application collects behavioral images and call voices of home-based elderly care clients, and based on the Internet of Things communication, obtains the behavioral images and call voices, and then obtains behavioral information text and call text, and then obtains self-care ability information data of home-based elderly care clients based on medical text data, behavioral information text and call text, and then combines the self-care ability information data, can combine Internet of Things devices and Internet of Things technology, and based on ASR technology, monitor the behavior of home-based elderly care clients in real time, realize real-time evaluation of the health status of home-based elderly care clients based on multi-dimensional information, and adjust the home-based elderly care service matching, further improve the accuracy of home-based elderly care service matching, so as to improve the quality and efficiency of home-based elderly care services.

[0063] See also Figure 3 , Figure 3 This is a schematic diagram of the second sub-flow of the home-based elderly care service matching processing method based on machine learning provided in the embodiment of the present application. Figure 3 As shown, in this embodiment, the physical and mental health information data of home-based elderly care customers in different dimensions are obtained, including:

[0064] S31, based on a preset home camera, collecting facial expression images of home-based elderly care customers;

[0065] S32, acquiring the expression image based on Internet of Things communication;

[0066] S33. Based on a preset artificial intelligence recognition model, the facial expression image is recognized to obtain the mental health information data of the home-based elderly care client.

[0067] Explanatory wise, a pre-set artificial intelligence recognition model, i.e., a preset artificial intelligence recognition model, also known as AI intelligent recognition, represents an artificial intelligence model for recognizing the relationship between facial expressions and mental health status. The preset artificial intelligence recognition model includes but is not limited to DeepFace, VGGFace, OpenFace and other models.

[0068] According to the above-mentioned preset home camera, in addition to being able to obtain behavioral information text, it is also possible to collect facial expression images of home-based elderly care customers based on the preset home camera, and obtain facial expression images based on the Internet of Things communication. Since facial expressions are related to human psychology, spirit, emotions, etc., facial expression images are recognized based on the preset artificial intelligence recognition model to obtain the mental health information data of home-based elderly care customers. The mental health information data includes but is not limited to information data such as mental health status, mental health status, emotional status, etc.

[0069] The embodiment of the present application collects facial expression images of home-based elderly care clients based on a preset home camera, obtains facial expression images based on Internet of Things communication, and then recognizes facial expression images based on a preset artificial intelligence recognition model to obtain mental health information data of home-based elderly care clients. The Internet of Things and artificial intelligence (AI) can be combined to monitor the mental health status of home-based elderly care clients in real time, and the health status of home-based elderly care clients can be evaluated in real time based on multi-dimensional information, and the matching of home-based elderly care services can be adjusted to further improve the accuracy of the matching of home-based elderly care services, thereby improving the quality and efficiency of home-based elderly care services.

[0070] In one embodiment, the user portrait includes several characteristics of the physical and mental health information of the customer; and according to the user portrait, determining the customer category corresponding to the home-based elderly care customer includes:

[0071] Based on a preset Multiple Hot encoding method, all the customer's physical and mental health information features are encoded into the same feature to obtain a unified encoding feature;

[0072] Based on a preset feature embedding method, mapping the unified coding feature from high-dimensional data to a low-dimensional space to obtain an embedding vector;

[0073] Based on the preset Wide & Deep model and according to the embedding vector, the customer category corresponding to the home-based elderly care customer is determined.

[0074] Explanatoryly, the pre-setting of the Multiple Hot encoding method, that is, the preset Multiple Hot encoding method, the preset Multiple Hot encoding method, also known as Multi-hot encoding, is a method of encoding multiple attributes into one feature at the same time.

[0075] Pre-setting feature embedding method, that is, preset feature embedding method, also known as Embedding, is a technology that maps high-dimensional data to low-dimensional space. The purpose is to convert discrete and sparse data into continuous and dense vector representations, so that these data can be better processed and understood by machine learning or deep learning models. Embedding is usually used to represent elements in fields such as text, images, and graphic nodes. Embedding is essentially a way to represent complex objects (such as words, phrases, users, products, etc.) with a real number vector. Each dimension in the vector corresponds to a potential feature of the object. This representation can capture the similarity or correlation between objects.

[0076] The Wide&Deep model is pre-set, that is, the preset Wide&Deep model. The Wide&Deep model is the Wide anddeep model. The preset Wide&Deep model is the model used by TensorFlow for classification and regression. The preset Wide&Deep model combines the memorization ability (Memorization) of the linear model and the generalization ability (Generalization) of the DNN model. During the training process, the parameters of the two models are optimized at the same time, so as to achieve the optimal prediction ability of the overall model.

[0077] Based on the above concept and setting, when the user portrait contains several customer physical and mental health information features of different dimensions, when determining the customer category corresponding to the home-based elderly care customer according to the user portrait, firstly, based on the preset Multiple Hot encoding method, all customer physical and mental health information features are encoded into the same feature to obtain a unified encoding feature, which is a sequence representation containing several information granularities, thereby encoding the multi-dimensional attributes of the home-based elderly care customer into one feature at the same time, so as to facilitate the comprehensive analysis of information of different dimensions to improve the accuracy of the analysis of home-based elderly care customers. Then, based on the preset feature embedding method, the unified encoding feature is mapped from high-dimensional data to low-dimensional space to obtain an embedded vector, thereby converting the discrete and sparse unified encoding feature into a continuous and dense vector representation, and can capture the similarity or correlation between the above-mentioned unified encoding feature objects of different granularities. Finally, based on the preset Wide&Deep model, we make full use of the memory ability (Memorization) of the linear model combined with the Wide&Deep model and the generalization ability (Generalization) of the DNN model, and according to the embedding vector, we comprehensively analyze the multi-dimensional information of the home-based elderly care customers represented by the embedding vector, determine the customer category corresponding to the home-based elderly care customers, and achieve the most accurate prediction possible for the customer category corresponding to the home-based elderly care customers.

[0078] In the embodiment of the present application, by performing Multiple Hot encoding and vectorizing the encoding Embedding, and then determining the customer category corresponding to the home-based elderly care customer based on the Wide & Deep model, the multi-dimensional information of the home-based elderly care customer can be uniformly embedded and processed, and the customer category of the home-based elderly care customer can be determined based on the multi-dimensional information of the home-based elderly care customer, which can improve the classification accuracy of the home-based elderly care customers, and then based on the accurate classification of the home-based elderly care customers, accurate personalized service recommendations for the home-based elderly care customers can be achieved, and the accuracy of home-based elderly care service matching can be improved, thereby improving the quality and efficiency of home-based elderly care services.

[0079] In one embodiment, after matching the target home-based elderly care service item to the home-based elderly care client, the method further includes:

[0080] Based on the preset Internet of Things communication method, the target home-based elderly care service item is pushed to the preset Internet of Things terminal corresponding to the home-based elderly care customer.

[0081] Explanatory, after matching the target home-based elderly care service items to the home-based elderly care customers, based on the preset IoT communication method and in accordance with the recommendation system technology, the target home-based elderly care service items are pushed to the preset IoT terminal corresponding to the home-based elderly care customers. For example, personalized home-based elderly care service packages are recommended to home-based elderly care customers through smart home devices such as front-end speakers, WeChat service accounts, related APPs and other channels. At the same time, behavioral data such as clicks, browsing, collections, purchases, and satisfaction of home-based elderly care service packages by home-based elderly care customers are collected, and the home-based elderly care service matching model based on machine learning is optimized to update the home-based elderly care service matching model based on machine learning, thereby further improving the accuracy and quality of home-based elderly care service matching.

[0082] The embodiment of the present application recommends target home-based elderly care service items to home-based elderly care customers through the Internet of Things communication method, and can realize intelligent home-based elderly care services based on the Internet of Things. It can not only realize personalized service recommendations for home-based elderly care customers to improve the accuracy and quality of home-based elderly care service matching, but also improve the efficiency and intelligence level of home-based elderly care services, thereby improving the intelligence level of home-based elderly care based on the Internet of Things.

[0083] It should be noted that the machine learning-based home-based elderly care service matching and processing methods described in the above embodiments can recombine the technical features contained in different embodiments as needed to obtain a combined implementation plan, but they are all within the scope of protection required by this application.

[0084] See also Figure 4 , Figure 4 A schematic block diagram of a machine learning-based home-based elderly care service matching processing device provided in an embodiment of the present application. Corresponding to the above-mentioned machine learning-based home-based elderly care service matching processing method, an embodiment of the present application also provides a machine learning-based home-based elderly care service matching processing device. Figure 4 As shown, the machine learning-based home-based elderly care service matching processing device includes a unit for executing the machine learning-based home-based elderly care service matching processing method described above, and the machine learning-based home-based elderly care service matching processing device can be configured in a computer device. Specifically, please refer to Figure 4 The machine learning-based home-based elderly care service matching processing device 40 includes a first acquisition unit 41, a data fusion unit 42, a first determination unit 43, a second determination unit 44, a first prediction unit 45 and a first matching unit 46.

[0085] The first acquisition unit 41 is used to acquire physical and mental health information data of different dimensions of home-based elderly care customers, wherein the physical and mental health information data includes health status information data, self-care ability information data, living status information data, and mental health information data;

[0086] A data fusion unit 42 is used to fuse the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data;

[0087] A first determining unit 43 is used to determine a user portrait corresponding to the home-based elderly care customer according to the fusion data;

[0088] A second determining unit 44 is used to determine the customer category corresponding to the home-based elderly care customer according to the user portrait;

[0089] A first prediction unit 45 is used to predict the home-based elderly care service demand corresponding to the home-based elderly care customer based on a preset prediction model of machine learning and according to the user portrait and the customer category;

[0090] The first matching unit 46 is used to match the target home-based elderly care service item corresponding to the home-based elderly care service demand based on a preset home-based elderly care service item database, and match the target home-based elderly care service item to the home-based elderly care client.

[0091] In one embodiment, the first acquiring unit 41 includes:

[0092] The first determination subunit is used to determine the medical information of the home-based elderly care client;

[0093] A first acquisition subunit is used to acquire the health information of the home-based elderly care client based on a preset wearable smart device;

[0094] The first classification subunit is used to classify the health status of the home-based elderly care client according to the medical information and the health information, and obtain the health status information data of the home-based elderly care client.

[0095] In one embodiment, the first determining subunit includes:

[0096] The second acquisition subunit is used to acquire medical image data of home-based elderly care customers;

[0097] A first recognition subunit is used to recognize the medical image data based on a preset OCR recognition method to obtain the medical text data of the home-based elderly care customer;

[0098] The first processing sub-unit is used to perform structured processing on the medical text data and extract preset target medical content to obtain the medical information of the home-based elderly care client.

[0099] In one embodiment, the first acquiring unit 41 includes:

[0100] The first acquisition subunit is used to collect behavior images of home-based elderly care customers based on a preset home camera;

[0101] A second recognition subunit is used to obtain the behavior image based on Internet of Things communication, and recognize the behavior image based on a preset cross-modal learning model to obtain the behavior information text;

[0102] A third acquisition subunit is used to acquire the call voice corresponding to the call with the home-based elderly care client based on a preset smart client terminal;

[0103] A third recognition subunit, configured to recognize a call text corresponding to the call voice based on a preset automatic speech recognition model;

[0104] The second determining subunit is used to determine the self-care ability information data of the home-based elderly care client according to the medical text data, the behavior information text and the call text.

[0105] In one embodiment, the first obtaining unit 41 includes:

[0106] The second acquisition subunit is used to acquire facial expression images of home-based elderly care customers based on a preset home camera;

[0107] A fourth acquisition subunit, configured to acquire the expression image based on Internet of Things communication;

[0108] The fourth recognition subunit is used to recognize the expression image based on a preset artificial intelligence recognition model to obtain the mental health information data of the home-based elderly care client.

[0109] In one embodiment, the user portrait includes several characteristics of the customer's physical and mental health information; the second determining unit 44 includes:

[0110] The encoding subunit is used to encode all the customer's physical and mental health information features into the same feature based on a preset Multiple Hot encoding method to obtain a unified encoding feature;

[0111] A mapping subunit, used to map the unified coding feature from high-dimensional data to a low-dimensional space based on a preset feature embedding method to obtain an embedding vector;

[0112] The third determination subunit is used to determine the customer category corresponding to the home-based elderly care customer based on a preset Wide & Deep model and according to the embedding vector.

[0113] In one embodiment, the machine learning-based home-based elderly care service matching processing device 40 further includes:

[0114] The push unit is used to push the target home-based elderly care service item to the preset Internet of Things terminal corresponding to the home-based elderly care customer based on the preset Internet of Things communication method.

[0115] It should be noted that technical personnel in the relevant field can clearly understand that the specific implementation process of the above-mentioned machine learning-based home-based elderly care service matching processing device and each unit can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and conciseness of the description, it will not be repeated here.

[0116] At the same time, the division and connection method of each unit in the above-mentioned home-based elderly care service matching processing device based on machine learning is only for illustrative purposes. In other embodiments, the home-based elderly care service matching processing device based on machine learning can be divided into different units as needed, and each unit in the home-based elderly care service matching processing device based on machine learning can also adopt a different connection order and method to complete all or part of the functions of the above-mentioned home-based elderly care service matching processing device based on machine learning.

[0117] The above-mentioned machine learning-based home-based elderly care service matching processing device can be implemented in the form of a computer program. The computer program can be used in Figure 5 Runs on the computer device shown.

[0118] See also Figure 5 , Figure 5 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a computer device such as a desktop computer or a server, or may be a component or part in other devices.

[0119] See also Figure 5 The computer device 500 includes a processor 502, a memory and a network interface 505 connected via a system bus 501, wherein the memory may include a non-volatile storage medium 503 and an internal memory 504, and the memory may also be a volatile storage medium.

[0120] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute the above-mentioned home-based elderly care service matching processing method based on machine learning.

[0121] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0122] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the above-mentioned home-based elderly care service matching processing method based on machine learning.

[0123] The network interface 505 is used to communicate with other devices over the network. Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the present application scheme, and does not constitute a limitation on the computer device 500 to which the present application scheme is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. For example, in some embodiments, the computer device may only include a memory and a processor. In such an embodiment, the structure and function of the memory and the processor are the same as those of the present application scheme. Figure 5 The embodiments shown are consistent and will not be described again here.

[0124] Among them, the processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps: obtain physical and mental health information data of different dimensions of home-based elderly care customers, and the physical and mental health information data include health status information data, self-care ability information data, living status information data, and mental health information data; fuse the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data; determine the user portrait corresponding to the home-based elderly care customer based on the fused data; determine the customer category corresponding to the home-based elderly care customer based on the user portrait; predict the home-based elderly care service demand corresponding to the home-based elderly care customer based on a preset prediction model for machine learning, and based on the user portrait and the customer category; match the target home-based elderly care service project corresponding to the home-based elderly care service demand based on a preset home-based elderly care service project database, and match the target home-based elderly care service project to the home-based elderly care customer.

[0125] In one embodiment, the processor 502 specifically implements the following steps when acquiring different dimensions of physical and mental health information data of home-based elderly care clients:

[0126] Determine the medical information of home care clients;

[0127] Based on a preset wearable smart device, obtaining health information of the home-based elderly care customer;

[0128] The health status of the home-based elderly care client is classified according to the medical information and the health information to obtain health status information data of the home-based elderly care client.

[0129] In one embodiment, the processor 502 implements the following steps when determining the medical information of the home-based elderly care client:

[0130] Obtain medical image data for home-based elderly care clients;

[0131] Based on a preset OCR recognition method, the medical image data is recognized to obtain the medical text data of the home-based elderly care customer;

[0132] The medical text data is structured and preset target medical content is extracted to obtain the medical information of the home-based elderly care client.

[0133] In one embodiment, the processor 502 specifically implements the following steps when acquiring different dimensions of physical and mental health information data of home-based elderly care clients:

[0134] Based on the preset home cameras, collect behavioral images of home-based elderly care customers;

[0135] Based on the Internet of Things communication, the behavior image is obtained, and based on a preset cross-modal learning model, the behavior image is identified to obtain the behavior information text;

[0136] Based on a preset intelligent client terminal, acquiring a call voice corresponding to a call with the home-based elderly care client;

[0137] Based on a preset automatic speech recognition model, identifying the call text corresponding to the call voice;

[0138] The self-care ability information data of the home-based elderly care client is determined based on the medical text data, the behavior information text and the call text.

[0139] In one embodiment, the processor 502 specifically implements the following steps when acquiring different dimensions of physical and mental health information data of home-based elderly care clients:

[0140] Based on the preset home cameras, facial expressions of home-based elderly care customers are collected;

[0141] Based on Internet of Things communication, obtaining the expression image;

[0142] Based on a preset artificial intelligence recognition model, the facial expression image is identified to obtain the mental health information data of the home-based elderly care client.

[0143] In one embodiment, the user portrait includes several characteristics of the physical and mental health information of the customer; when the processor 502 determines the customer category corresponding to the home-based elderly care customer according to the user portrait, the processor 502 specifically implements the following steps:

[0144] Based on a preset Multiple Hot encoding method, all the customer's physical and mental health information features are encoded into the same feature to obtain a unified encoding feature;

[0145] Based on a preset feature embedding method, mapping the unified coding feature from high-dimensional data to a low-dimensional space to obtain an embedding vector;

[0146] Based on the preset Wide & Deep model and according to the embedding vector, the customer category corresponding to the home-based elderly care customer is determined.

[0147] In one embodiment, after matching the target home-based elderly care service item to the home-based elderly care client, the processor 502 further implements the following steps:

[0148] Based on the preset Internet of Things communication method, the target home-based elderly care service item is pushed to the preset Internet of Things terminal corresponding to the home-based elderly care customer.

[0149] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 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.

[0150] Those skilled in the art will appreciate that all or part of the processes in the method for implementing the above embodiment can be completed by a computer program, which can be stored in a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the process steps of the embodiment of the above method.

[0151] Therefore, the present application also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, and the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the following steps:

[0152] A computer program product, when running on a computer, enables the computer to execute the steps of the machine learning-based home-based elderly care service matching processing method described in the above embodiments.

[0153] The computer-readable storage medium may be an internal storage unit of the aforementioned device, such as a hard disk or memory of the device. The computer-readable storage medium may also be an external storage device of the device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the device and an external storage device.

[0154] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0155] The storage medium is a physical, non-transient storage medium, for example, it can be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, etc., which can store computer programs.

[0156] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0157] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0158] The steps in the method of the embodiment of the present application can be adjusted in order, combined and deleted according to actual needs. The units in the device of the embodiment of the present application can be combined, divided and deleted according to actual needs. In addition, the functional units in the various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0159] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for an electronic device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0160] The software tools or components not provided by our company that appear in the embodiments of this application are merely examples and do not represent actual use.

[0161] The relevant data collection appearing in the embodiments of the present application complies with the requirements of relevant laws and regulations, such as GDPR (General Data Protection Regulation of the European Union) or information security standards of other countries and regions.

[0162] The above is only a specific implementation 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 replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for matching home-based elderly care services based on machine learning, characterized in that: include: Obtaining physical and mental health information data of different dimensions of home-based elderly care customers, wherein the physical and mental health information data includes health status information data, self-care ability information data, living status information data, and mental health information data; Fusing the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data; Determine the user portrait corresponding to the home-based elderly care customer based on the fused data; Determine the customer category corresponding to the home-based elderly care customer based on the user portrait; Based on a preset prediction model of machine learning, and according to the user portrait and the customer category, predict the home-based elderly care service needs corresponding to the home-based elderly care customer; Based on a preset home-based elderly care service project database, the target home-based elderly care service project corresponding to the home-based elderly care service demand is matched, and the target home-based elderly care service project is matched to the home-based elderly care customer.

2. The method for matching home-based elderly care services based on machine learning according to claim 1 is characterized in that: Obtain physical and mental health information data of home-based elderly care customers in different dimensions, including: Determine the medical information of home care clients; Based on a preset wearable smart device, obtaining health information of the home-based elderly care customer; The health status of the home-based elderly care client is classified according to the medical information and the health information to obtain health status information data of the home-based elderly care client.

3. The method for matching home-based elderly care services based on machine learning according to claim 2 is characterized in that: Identify medical information of home care clients, including: Obtain medical image data for home-based elderly care clients; Based on a preset OCR recognition method, the medical image data is recognized to obtain the medical text data of the home-based elderly care customer; The medical text data is structured and preset target medical content is extracted to obtain the medical information of the home-based elderly care client.

4. The method for matching home-based elderly care services based on machine learning according to claim 3 is characterized in that: Obtain physical and mental health information data of home-based elderly care customers in different dimensions, including: Based on the preset home cameras, collect behavioral images of home-based elderly care customers; Based on the Internet of Things communication, the behavior image is obtained, and based on a preset cross-modal learning model, the behavior image is identified to obtain the behavior information text; Based on a preset intelligent client terminal, acquiring a call voice corresponding to a call with the home-based elderly care client; Based on a preset automatic speech recognition model, identifying the call text corresponding to the call voice; The self-care ability information data of the home-based elderly care client is determined based on the medical text data, the behavior information text and the call text.

5. The method for matching home-based elderly care services based on machine learning according to claim 1 is characterized in that: Obtain physical and mental health information data of home-based elderly care customers in different dimensions, including: Based on the preset home cameras, facial expressions of home-based elderly care customers are collected; Based on Internet of Things communication, obtaining the expression image; Based on a preset artificial intelligence recognition model, the facial expression image is identified to obtain the mental health information data of the home-based elderly care client.

6. The method for matching home-based elderly care services based on machine learning according to any one of claims 1 to 5, characterized in that: The user portrait includes several characteristics of the physical and mental health information of the customer; according to the user portrait, the customer category corresponding to the home-based elderly care customer is determined, including: Based on a preset Multiple Hot encoding method, all the customer's physical and mental health information features are encoded into the same feature to obtain a unified encoding feature; Based on a preset feature embedding method, mapping the unified coding feature from high-dimensional data to a low-dimensional space to obtain an embedding vector; Based on the preset Wide & Deep model and according to the embedding vector, the customer category corresponding to the home-based elderly care customer is determined.

7. The method for matching home-based elderly care services based on machine learning according to claim 1 is characterized in that: After matching the target home-based elderly care service item to the home-based elderly care client, the method further includes: Based on the preset Internet of Things communication method, the target home-based elderly care service item is pushed to the preset Internet of Things terminal corresponding to the home-based elderly care customer.

8. A matching processing device for home-based elderly care services based on machine learning, characterized in that: include: The first acquisition unit is used to acquire physical and mental health information data of different dimensions of home-based elderly care customers, wherein the physical and mental health information data includes health status information data, self-care ability information data, living status information data, and mental health information data; A data fusion unit, used for fusing the health status information data, the self-care ability information data, the living status information data, and the mental health information data to obtain fused data; A first determining unit, configured to determine a user portrait corresponding to the home-based elderly care customer according to the fused data; A second determining unit is used to determine the customer category corresponding to the home-based elderly care customer according to the user portrait; A first prediction unit is used to predict the home-based elderly care service demand corresponding to the home-based elderly care customer based on a preset prediction model of machine learning and according to the user portrait and the customer category; The first matching unit is used to match the target home-based elderly care service items corresponding to the home-based elderly care service demand based on a preset home-based elderly care service item database, and match the target home-based elderly care service items to the home-based elderly care client.

9. A computer device, characterized in that: The computer device comprises a memory and a processor connected to the memory; the memory is used to store a computer program; the processor is used to run the computer program to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 can be implemented.

Images