WO2021000554A1

WO2021000554A1 - Method and system for optimizing recommended massage program of smart massage chair

Info

Publication number: WO2021000554A1
Application number: PCT/CN2019/130295
Authority: WO
Inventors: 王铭琪; 陈水金; 冯时; 张云龙
Original assignee: 奥佳华智能健康科技集团股份有限公司
Priority date: 2019-07-01
Filing date: 2019-12-31
Publication date: 2021-01-07
Also published as: CN110347449A; CN110347449B; KR102470171B1; KR20210005544A

Abstract

A method and a system for optimizing a recommended massage program of a smart massage chair. Said method comprises the steps of: 1) acquiring personal information of a user; 2) obtaining an initial massage recommendation combination according to the personal information, the massage recommendation combination comprising a manipulation combination and adjustable parameters; 3) calculating a correction amount according to historical evaluation data, and adjusting the adjustable parameters according to the correction amount, so as to obtain a final massage recommendation combination; 4) a massage device executing the final massage recommendation combination; and 5) collecting user evaluation data and storing same in the historical evaluation data. The method and system realize an overall closed loop of massage programs, user feedback, analysis and optimization, thereby continuously improving recommendation logics, and providing a better massage experience.

Description

Method and system for optimizing recommended massage procedures for intelligent massage chairs

Technical field

The invention relates to the field of massage equipment, in particular to a method and system for optimizing recommended massage procedures for an intelligent massage chair.

Background technique

The existing massage chair presets a variety of modes, and the user manually selects the massage mode for massage during the process of using the massage chair. With the continuous development of the massage chair industry, people have put forward a higher demand for intelligent interactive massage. This demand is not only reflected in the intelligent improvement of operation interaction, but also in the intelligent optimization of massage techniques.

Therefore, it appears that the massage mode for the target user is determined according to the physiological data of the user, and the massage mode suitable for the target user is recommended to the target user. However, whether it is manual selection or smart recommendation, the types of massage modes involved are limited, and the massage modes are preset and stored in the massage chair. They include fixed massage techniques and related parameters, which cannot be adjusted and changed according to personal needs. To achieve more personalized and accurate recommendation, the user experience is not good enough.

Summary of the invention

technical problem

The main purpose of the present invention is to overcome the above-mentioned shortcomings in the prior art, and propose a method and system for optimizing the massage program recommended by a smart massage chair that can combine personal information to make personalized and accurate recommendations, and combine feedback opinions to optimize the massage program.

The solution to the problem

The present invention adopts the following technical solutions:

A method for optimizing recommended massage procedures for an intelligent massage chair is characterized in that the steps include the following:

1) Collect user personal information;

2) Obtain an initial recommended massage combination according to personal information, the recommended massage combination includes a combination of techniques and adjustable parameters, and the adjustable parameters are used to adjust the working process of the combination of techniques;

3) Calculate the correction amount according to the historical evaluation data, and adjust the adjustable parameters according to the correction amount to obtain the final massage recommendation combination;

4) The massage equipment implements the recommended combination of massage;

5) Collect user evaluation data and store it in historical evaluation data.

The personal information includes physiological parameter information, and the physiological parameter information includes one or more of blood pressure, blood oxygen, heart rate, fatigue index, or soreness index.

The personal information includes basic information, and the basic information includes one or more of height, weight, gender, age, or tolerance.

There are several massage instruction units in advance, and each massage instruction unit is equipped with several combination packages; according to personal information, one of the combination packages of each massage instruction unit is taken to form the combination of techniques.

The several combination packages of the massage instruction unit are divided according to one or more of massage time, massage intensity, massage times or massage speed.

There are several combinations of techniques in advance, and one combination of techniques is selected based on personal information.

The adjustable parameters include one or more of massage time, massage intensity, massage times, massage speed, or accuracy of massaged parts.

Let the adjustable parameters of the initial recommended massage combination be (l _i , t _i ), l _i represents massage intensity, t _i represents massage duration; historical evaluation data is (ΔL _i , ΔT _i ), ΔL _i represents massage intensity evaluation, ΔT _i represents the evaluation of massage duration; the correction amount is (Δl _i , Δt _i ), the preset weight is a _i , i∈[1,N], N is the number of historical evaluation data taken; then calculated based on historical evaluation data Correction amount:

Δl _i =ΔL _iN *a _iN +ΔL _i-N+1 *a _i-N+1 ... +ΔL _i-1 *a _i-1

Δt _i =ΔT _iN *a _iN +ΔT _i-N+1 *a _i-N+1 ... +ΔT _i-1 *a _i-1 ;

Adjust the adjustable parameters according to the correction amount, and the final recommended combination of adjustable parameters is (l _i +Δl _i , t _i +Δt _i ).

The evaluation data includes one or more of massage time evaluation, massage intensity evaluation, massage frequency evaluation, massage speed evaluation, or massage part evaluation.

An intelligent massage recommendation system, which is characterized in that it comprises

The massage chair stores a number of manipulation combinations and adjustable parameters; it is used to collect some personal information and send it to the mobile terminal, and to receive massage recommendations and perform massage actions;

The mobile terminal is used to collect some personal information and send it to the cloud platform, and send the massage recommendation combination from the cloud platform to the massage chair;

The cloud platform stores a number of technique combinations, adjustable parameters and historical evaluation data. According to personal information, the initial massage recommendation combination is obtained, the correction amount is calculated according to the historical evaluation data, and the adjustable parameters are adjusted according to the correction amount to obtain the final massage recommendation combination, and Send to the mobile terminal.

According to the above description of the present invention, compared with the prior art, the present invention has the following beneficial effects:

1. The optimization method of the present invention obtains the initial massage recommendation combination according to personal information, calculates the correction amount through historical evaluation data, uses the correction amount to adjust the adjustable parameters to obtain the final massage recommendation combination, and realizes the massage program, user feedback, and analysis optimization The overall closed loop, so as to continuously improve the recommendation logic, and bring a better massage experience.

2. The optimization method of the present invention divides the combination of techniques into several massage instruction units, and each massage instruction unit is set with multiple combination packages. By collecting personal information and combining the corresponding calculation model, the combination package of each massage instruction unit is determined By composing massage recommendations to achieve precise recommendations for users, massages are more personalized, rather than uniform.

3. The method of the present invention decomposes the combination of manipulations into several massage instruction units, which can be decomposed according to the massage location or massage sequence. Each massage instruction unit can be set with multiple combinations according to massage time, massage intensity, massage times or massage speed, etc. If you only store the basic massage package, you can combine a large number of massage programs to achieve a thousand people in the true sense.

4. In the method of the present invention, a corresponding calculation model is designed for each massage instruction unit, and the calculation model can also be optimized and adjusted as required, which is convenient for maintenance and upgrade.

5. The method of the present invention uses the evaluation data collected after the massage as the historical evaluation data, and combines the correction amount and the preset weight value to automatically optimize the adjustable parameters in the recommendation logic, which saves time, effort and high efficiency.

6. In the method of the present invention, it is recommended that the massage program is a massage program combination package code, that is, an instruction set, rather than a complete massage program, with small size and fast transmission speed.

7. The device, information collection device, storage device and recommendation device of the present invention can be integrated on the massage chair to implement the recommendation method of the present invention.

8. The system of the present invention detects the soreness value, blood oxygen, pulse rate and other parameters of the body part by integrating physiological parameter detection equipment on the massage chair; designs a questionnaire survey page on the mobile terminal APP to collect basic user information, including Height, weight, age, gender, massage tolerance, and submit them to the cloud platform. Based on the collected information and supplemented by user feedback, the cloud platform optimizes massage procedures and recommendation logic through big data analysis, different users, different physiology Parameters and different personal information will have different intelligent recommendation results, achieving personalized massage effects for thousands of people. Based on the collected user physical information, supplemented by user feedback.

9. The system of the present invention also integrates relevant detection equipment on the massage chair to detect body parts' soreness, blood oxygen, pulse rate and other parameters, and collects the user's height, weight, age, gender, massage tolerance, and submits To the cloud platform, the cloud platform optimizes the massage program and recommendation logic through big data analysis based on the collected information and user feedback. Different users, different physiological parameters, and different personal information will have different intelligent recommendations. As a result, a personalized massage effect for thousands of people is achieved.

Brief description of the drawings

[Figure 1 is a flow chart of the method in Embodiment 1 of the present invention];

[Figure 2 is a system composition diagram of Embodiment 1 of the present invention];

[Figure 3 is a composition diagram of the combination of techniques in the first embodiment of the present invention].

Description of Examples

The present invention will be further described below through specific embodiments.

Example one

Refer to Figure 1 for a method for optimizing recommended massage procedures for a smart massage chair. The specific steps include the following:

1) Collect user personal information, including physiological parameter information and basic information. The physiological parameter information includes one or more of blood pressure, blood oxygen, heart rate, or soreness index. The physiological parameter information is the information detected by the related detection device. The basic information includes one or more of height, weight, gender, age, or tolerance, and the height, weight, gender, age, or tolerance can be obtained through user input.

2) Obtain the initial recommended massage combination according to personal information. The recommended massage combination includes a combination of techniques and adjustable parameters. The adjustable parameters are used to adjust the working process of the combination of techniques. The adjustable parameters include massage time, massage intensity, and massage. One or more of times or massage speed. details as follows:

There are several massage instruction units in advance, and each massage instruction unit has several combination packages for selection. The several massage instruction units are used to control multiple massage devices to work at the same time or work in an interval between each other. They can be divided according to massage parts or according to the order of massage, etc. The division method is not unique, for example: according to the massage parts: head, neck Part, back, waist, etc.; according to the massage sequence, it is divided into: cushion, middle, end, etc., or according to the massage part and combined with the massage sequence, such as cushion, middle (head, neck, back, waist, etc.), end.

Several combination packages of each massage instruction unit can be divided according to one or more of massage time, massage intensity, massage times or massage speed. For example, the massage time is divided into 3 combination packages from long to short, and 3 packages according to the number of massages from more to less, or according to the massage speed and number of massages, it is divided into 5 packages, etc. The division method and number are not unique.

Set the first group of evaluation scores for each combination package, and generate the second group of evaluation scores for personal information; for each massage instruction unit, select the first group of evaluation scores that are closest to the second group of evaluation scores The combination package corresponding to the value generates the massage recommendation combination. Wherein, a preset calculation function may be used to calculate the input personal information to obtain the second group of evaluation scores, and the calculation function includes commonly used mathematical calculation models such as score calculation functions and collaborative filtering algorithms.

for example

Assuming that M is the recommended combination of massage, that is, the instruction set, decomposed into n massage instruction units, including head segment + neck segment + shoulder segment + back segment + waist segment + hip segment, then M = M1∪M2∪ M3...∪Mn. Each massage instruction unit M _x , x∈i∈[1, N], including q _x combination packages

These combination packages can have different massage techniques and effects. Refer to Figure 3 and Table 1, including 6 massage instruction units (head, neck, shoulders, back, waist, buttocks), and each massage instruction unit has 5 combination packages.

Table 1

Massage instruction calculation function unit for each design _{_{value, J x (r 1, r}} 2, r 3 ...., R p), r 1, r 2, r 3 ..., r p personal information. Combination package corresponding to the massage command unit M _x

Set points

Table 2

程序片段Program fragment	计算函数Calculation function	计算结果Calculation results
头部head	J1(r1，r2，r3….，r11)J1(r1, r2, r3...., r11)	分值2，对应组合包P12Score 2, corresponding to package P12
颈部neck	J2(r1，r2，r3….，r11)J2(r1, r2, r3...., r11)	分值3，对应组合包P23Score 3, corresponding to package P23

肩部Shoulder	J3(r1，r2，r3….，r11)J3(r1, r2, r3...., r11)	分值4，对应组合包P34Score 4, corresponding to package P34
背部Back	J4(r1，r2，r3….，r11)J4(r1, r2, r3...., r11)	分值5，对应组合包P45Score 5, corresponding to package P45
腰部waist	J5(r1，r2，r3….，r11)J5(r1, r2, r3...., r11)	分值2，对应组合包P52Score 2, corresponding to package P52
臀部Buttocks	J6(r1，r2，r3….，r11)J6(r1, r2, r3...., r11)	分值1，对应组合包P61Score 1, corresponding to package P61

Collect input parameters r ₁ , r ₂ , r ₃ …., r _p , including soreness index (the soreness level is divided into 3 levels, light, moderate, and severe, replaced by 1, 2, and 3 respectively), blood oxygen, heart rate, height , Weight, gender, age, massage tolerance (divided into three levels: light, normal, and heavy, represented by 0, 1, 2 respectively), fatigue index (divided into 4 levels, replaced by 1, 2, 3, 4 respectively) )Wait.

table 3

参数parameter	值value
颈部酸痛r1Neck pain r1	22
背部酸痛r2Back pain r2	11
腰部酸痛r3Waist pain r3	11
肩部酸痛r4Shoulder pain r4	33
心率r5Heart rate r5	6565
血氧r6Blood oxygen r6	9898
身高r7Height r7	165165
体重r8Weight r8	5757
性别r9Sex r9	11
年龄r10Age r10	3333
按摩耐受度r11Massage tolerance r11	11

For each massage instruction unit, input personal information r ₁ , r ₂ , r ₃ …., r _p into the corresponding score calculation function, and combine the calculation result _S'x with the score of the package

For comparison, take the combination package corresponding to the score with the smallest difference in each massage instruction unit

y _x ∈[1, q _x ], then the combination of techniques can be obtained

3) Calculate the correction amount according to the historical evaluation data, and adjust the adjustable parameters according to the correction amount to obtain the final massage recommendation combination. When the number of historical evaluation data is insufficient, the adjustable parameters can be adjusted according to the preset correction amount. When the number of historical evaluation data is sufficient, the calculation is as follows:

Let the adjustable parameters of the initial recommended massage combination be (l _i , t _i ), l _i represents massage intensity, t _i represents massage duration; historical evaluation data is (ΔL _i , ΔT _i ), ΔL _i represents massage intensity evaluation ( Too light, light, just right, heavy, too heavy), the scores are (2, 1, 0, -1, -2); ΔT _i represents the evaluation of massage time (too short, short, just right, long, too long) , The scores are (2, 1, 0, -1, -2); the correction amount is (Δl _i , Δt _i ), the default weight is a _i , the newer the date, the greater the weight, and a _iN + a _{i -N+1} ...+a _i-1 =1, i∈[1,N], N is the number of historical evaluation data taken; then the correction amount is calculated according to the historical evaluation data:

Δl _i =ΔL _iN *a _iN +ΔL _i-N+1 *a _i-N+1 ... +ΔL _i-1 *a _i-1

Δt _i =ΔT _iN *a _iN +ΔT _i-N+1 *a _i-N+1 ...+ΔT _i-1 *a _i-1 adjust the adjustable parameters according to the correction amount, then the final recommended combination can be The tuning parameter is (l _i +Δl _i , t _i +Δt _i )

For the i-th pushed Recommended combinations of massage can be represented as _{_{R (M i, L i,}} T i), where M _i denotes non-adjustable way _{_{combination, L i = l i + Δl}} i, T i = t i + Δt _i .

The types of adjustable parameters are not limited to massage strength and massage duration, but can also include the number of massages, massage speed, and accuracy of massage parts. For massage parts, they can be refined to the shoulders, neck, back, and waist. The accuracy includes partial Left, right, top, bottom. The calculation method of the correction amount is similar to the calculation method of massage intensity and massage duration.

4) Send the final massage recommendation combination to the massage equipment, and the massage equipment performs the massage action. The combination of manipulations and adjustable parameters included in the final recommended massage combination is a combination of massage command data, where each massage command unit of the manipulation combination can be used to control the corresponding massage device to perform massage actions, and the adjustable parameters can control the correlation of the massage device parameter.

5) Collect the user's evaluation data and store it in the historical evaluation data, that is, the historical evaluation data includes the evaluation data of the user after multiple massages. For the evaluation data, the composition is as follows: the evaluation data is (ΔL _i , ΔT _i ), where: ΔL _i represents the massage strength evaluation (too light, light, just, heavy, too heavy), and the scores are (2, 1) , 0, -1, -2), ΔT _i represents the massage duration evaluation (too short, short, just right, long, too long), and the scores are (2, 1, 0, -1, -2).

The specific types of evaluation data are not limited to massage intensity evaluation and massage duration evaluation, but can also include massage frequency evaluation, massage speed evaluation, massage part evaluation, etc., which can correspond to adjustable parameters.

The present invention also provides an intelligent massage recommendation device, including:

The information collection unit is used to collect the user's personal information and collect evaluation data. It can use detection handles, smart bracelets, mobile phones, tablets, etc., electrodes or sensor chips are set in the detection handles and smart bracelets to detect physiological parameter information , Enter personal information and collect evaluation data through smart terminals such as bracelets, mobile phones, and tablets.

The storage unit stores several combination of techniques, adjustable parameters and historical evaluation data. It also stores a number of massage instruction units, and each massage instruction unit is provided with a number of combination packages; the correspondence relationship between the calculation model and the massage instruction unit and the correspondence relationship between parameters and the combination package are stored.

The recommendation unit obtains the initial massage recommendation combination according to personal information, calculates the correction amount according to the historical evaluation data, and adjusts the adjustable parameters according to the correction amount to obtain the final massage recommendation combination.

The intelligent massage recommendation device can be installed in a massage chair or on other massage equipment, recommend a combination of massage recommendations based on the user's personal information combined with evaluation data, and perform corresponding massage actions.

Referring to Figure 2, the present invention also proposes an intelligent massage recommendation system, including

The massage chair is equipped with an information collection unit, a storage unit, a communication unit, a control unit and several massage devices. The storage unit stores a number of manipulation combinations, adjustable parameters, etc. The manipulation combination includes a plurality of massage instruction units, and each massage instruction unit is provided with a number of combination packages. The information collection unit is used to collect physiological parameter information and can be designed as a detection handle or a built-in detection module. The communication unit is used to communicate with the mobile terminal, and can adopt Bluetooth or WIFI.

The several massage devices are used to perform massage actions, which can correspond to several massage instruction units, including head massage unit, neck massage unit, back massage unit, waist massage unit, hip massage unit, foot massage unit, etc., The massage device may not correspond to the massage instruction unit. The control unit is used to control the coordinated work of each unit.

The mobile terminal is used to collect the user’s basic information and evaluation data and send it to the cloud platform, and send the massage recommendation combination from the cloud platform to the massage chair. It can be a smart terminal such as a mobile phone or a tablet, with an operation interface, a communication module, etc. Collect basic information and evaluation data such as age, weight, gender and tolerance through questionnaire surveys.

The cloud platform is equipped with a database, communication unit, recommendation unit, etc. The database stores several technique combinations, adjustable parameters and historical evaluation data, etc. The technique combination includes several massage instruction units, and each massage instruction unit has several combination packages . The communication unit communicates with the mobile terminal and can receive personal information and send massage recommendations. Obtain the initial massage recommended combination according to personal information, calculate the correction amount according to the historical evaluation data, and adjust the adjustable parameters according to the correction amount to obtain the final massage recommendation combination.

In the intelligent massage recommendation system of the present invention, the user can establish a personal account on the cloud platform in advance through the mobile terminal. When in use, the massage chair collects the user’s physiological parameter information and sends the information to the mobile terminal. The mobile terminal collects the user’s basic information and sends it to the cloud platform together with the physiological parameter information. The cloud platform calculates the personal information through the relevant calculation model. The initial massage recommended combination is calculated based on the historical evaluation data, and the adjustable parameters are adjusted according to the correction value to obtain the final massage recommended combination and sent to the mobile terminal. The mobile terminal then sends the massage recommended combination to the massage chair. The recommended combination performs the corresponding massage action; after the massage is completed, the mobile terminal collects the evaluation data and transmits it to the cloud platform.

Example two

An intelligent massage chair recommended massage program optimization method. Its main features are the same as the first embodiment. The difference is that the combination of techniques is a preset massage mode, that is, a combination of several massage command units is a preset combination. Personal information is calculated to obtain relevant indicators, and one of the method combinations is determined according to the indicators. For example, the relevant index is the degree of fatigue, and each fatigue degree corresponds to a massage mode to relieve fatigue.

Example three

A smart massage chair recommended massage program optimization method, device and system. The optimization method and device are the same as those in the first embodiment. The difference lies in that the recommendation system includes:

The massage chair is equipped with an information collection unit, a storage unit, a communication unit, a control unit and several massage devices. The storage unit stores a number of manipulation combinations, adjustable parameters, historical evaluation data, etc. The manipulation combination includes a number of massage instruction units, and each massage instruction unit is provided with a number of combination packages. The information collection unit is used to collect physiological parameter information and can be designed as a detection handle or a built-in detection module; and an intelligent interactive interface is set to collect basic information and evaluation data such as age, weight, gender, and tolerance through questionnaire surveys. The communication unit is used to communicate with the cloud platform, which can be implemented through WIFI or internet or other internet communication protocols.

In the intelligent massage recommendation system of the present invention, the user can establish a personal account on the cloud platform in advance through a mobile terminal. When in use, the massage chair collects the user’s physiological parameter information and basic information and sends the information to the cloud platform. The cloud platform calculates the personal information through the relevant calculation model to generate the initial massage recommendation combination, and then calculate the correction amount based on the historical evaluation data. Adjust the adjustable parameters according to the correction amount to obtain the final recommended massage combination and send it to the massage chair. The massage chair executes the corresponding massage action according to the recommended massage combination. After the massage is completed, the massage chair collects the evaluation data and transmits it to the cloud platform.

Non-patent literature

Claims

A method for optimizing recommended massage procedures for an intelligent massage chair is characterized in that the steps include the following:

1) Collect user personal information;

2) Obtain an initial recommended massage combination according to personal information, the recommended massage combination includes a combination of techniques and adjustable parameters, and the adjustable parameters are used to adjust the working process of the combination of techniques;

3) Calculate the correction amount according to the historical evaluation data, and adjust the adjustable parameters according to the correction amount to obtain the final massage recommendation combination;

4) The massage equipment implements the final massage recommendation combination;

5) Collect user evaluation data and store it in historical evaluation data.
The method for optimizing a recommended massage program for an intelligent massage chair according to claim 1, wherein the personal information includes physiological parameter information, and the physiological parameter information includes one of blood pressure, blood oxygen, heart rate, fatigue index, or soreness index. Kind or more.
The method for optimizing recommended massage procedures for an intelligent massage chair according to claim 1, wherein the personal information includes basic information, and the basic information includes one of height, weight, gender, age, or tolerance. Many kinds.
An intelligent massage chair recommended massage program optimization method according to claim 1, wherein a number of massage instruction units are pre-installed, and each massage instruction unit is provided with a number of combination packages; each massage instruction unit is taken according to personal information One of the combination packages of, constitute the combination of techniques.
An intelligent massage chair recommended massage program optimization method according to claim 4, wherein the several combination packages of the massage instruction unit are based on one or more of massage time, massage intensity, massage frequency, or massage speed. Kind of division.
A method for optimizing a recommended massage program for an intelligent massage chair according to claim 1, characterized in that a number of manipulation combinations are preset, and one of the manipulation combinations is selected according to personal information.
The intelligent massage chair recommended massage program optimization method according to claim 1, wherein the adjustable parameters include one or more of massage time, massage intensity, massage times, massage speed, or accuracy of massage parts Kind.
The method for optimizing a recommended massage program for an intelligent massage chair according to claim 1, wherein the adjustable parameters of the initial recommended massage combination are (l i , t i ), where l i represents massage strength, and t i represents Massage duration; historical evaluation data is (ΔL i , ΔT i ), ΔL i represents massage intensity evaluation, ΔT i represents massage duration evaluation; correction amount is (Δl i , Δt i ), preset weight is a i , i∈[ 1, N], N is the number of historical evaluation data taken; then the correction amount is calculated according to the historical evaluation data:

Δl i =ΔL iN *a iN +ΔL i-N+1 *a i-N+1 ... +ΔL i-1 *a i-1

Δt i =ΔT iN *a iN +ΔT i-N+1 *a i-N+1 ... +ΔT i-1 *a i-1 ;

Adjust the adjustable parameters according to the correction amount, then the final recommended combination of adjustable parameters is

(l i +Δl i , t i +Δt i ).
The method for optimizing recommended massage procedures for an intelligent massage chair according to claim 1, wherein the evaluation data includes one of massage time evaluation, massage intensity evaluation, massage frequency evaluation, massage speed evaluation, or massage part evaluation. Many kinds.
An intelligent massage recommendation system, which is characterized in that it comprises

The massage chair stores a number of manipulation combinations and adjustable parameters; it is used to collect some personal information and send it to the mobile terminal, and to receive massage recommendations and perform massage actions;

The mobile terminal is used to collect some personal information and send it to the cloud platform, and send the massage recommendation combination from the cloud platform to the massage chair;

The cloud platform stores a number of technique combinations, adjustable parameters and historical evaluation data. According to personal information, the initial massage recommendation combination is obtained, the correction amount is calculated according to the historical evaluation data, and the adjustable parameters are adjusted according to the correction amount to obtain the final massage recommendation combination, and Send to the mobile terminal.