A data processing method and system, electronic equipment, and storage medium

Abstract

The disclosure provides a data processing method and system, electronic equipment and storage medium, and belongs to the field of data processing. The method comprises the following steps: in response to receiving an automatic control instruction, determining a plurality of target standard information from a plurality of standard information based on room information; determining a first control model based on the matching degree between the basic information of a target person and the plurality of target standard information; the standard information is the basic information of a historical person; inputting the behavior characteristics of the target person in the target room into the first control model to obtain a control strategy of the target room; the target room is a room where the target person lives; and controlling the intelligent devices in the target room based on the control strategy. The data processing method and system, electronic equipment and storage medium provided by the disclosure can improve the accuracy and reliability of hotel control.

Description

TECHNICAL FIELD

[0001] The present disclosure belongs to the field of data processing, and more particularly relates to a data processing method and system, an electronic device, and a storage medium. BACKGROUND

[0002] In the fierce competition environment of the hotel industry today, improving the quality and intelligent level of room service has become a key development trend. With the rapid progress of technology, more and more intelligent devices are equipped in hotel rooms, such as intelligent air conditioners, intelligent lighting systems, intelligent curtains, etc., aiming to provide guests with more comfortable, convenient and personalized check-in experience.

[0003] The traditional room equipment control method is often simple and fixed, and is mostly based on preset schedules or simple sensor threshold triggers, which is difficult to accurately adapt to the diversified needs of different guests under different environmental conditions.

[0004] Therefore, there is an urgent need for an accurate and reliable data processing method. SUMMARY

[0005] The present disclosure aims to provide a data processing method and system, an electronic device, and a storage medium to improve the accuracy and reliability of hotel control.

[0006] In a first aspect, the present disclosure provides a data processing method, comprising:

[0007] In response to receiving an automatic control instruction, determining a plurality of target standard information from a plurality of standard information based on room information;

[0008] Determining a first control model based on the matching degree of the basic information of the target personnel and the plurality of target standard information; the standard information is the basic information of historical personnel;

[0009] Inputting the behavior characteristics of the target personnel in the target room into the first control model to obtain a control strategy of the target room; the target room is the room where the target personnel resides;

[0010] Controlling the intelligent equipment in the target room based on the control strategy.

[0011] In a second aspect, the present disclosure provides a data processing system, comprising:

[0012] A standard information determination module configured to determine a plurality of target standard information from a plurality of standard information based on room information in response to receiving an automatic control instruction;

[0013] A model determination module configured to determine a first control model based on the matching degree of the basic information of the target personnel and the plurality of target standard information; the standard information is the basic information of historical personnel;

[0014] The control strategy output module is configured to input the behavior characteristics of the target person in the target room into a first control model to obtain a control strategy of the target room; the target room is a room in which the target person resides;

[0015] The control module is configured to control the smart device in the target room based on the control strategy.

[0016] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the data processing method when executing the computer program.

[0017] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of the data processing method.

[0018] The data processing method and system, the electronic device, and the storage medium provided by the embodiments of the present disclosure have the following beneficial effects:

[0019] The present disclosure can filter target standard information that meets the current room from a pre-set standard information library according to room information, which narrows the range of filtering and makes the filtered target standard information more in line with the actual situation of the current room, thereby improving the pertinence of data processing. The present disclosure takes the environmental characteristics of the target room as input and outputs a control strategy through a first control model, so that the present disclosure can adjust the control strategy according to the real-time changing environmental characteristics, thereby ensuring that the environment of the target room is always in the best state and improving the accuracy and reliability of data processing. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor.

[0021] Figure 1 A flowchart of a data processing method provided by an embodiment of the present disclosure is shown in the figure;

[0022] Figure 2 A structural block diagram of a data processing system provided by an embodiment of the present disclosure is shown in the figure;

[0023] Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure is shown in the figure. DETAILED DESCRIPTION

[0024] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present disclosure. However, it will be apparent to those skilled in the art that the present disclosure can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present disclosure with unnecessary detail.

[0025] In order to make the objectives, technical solutions and advantages of the present disclosure clearer, the following will be described with specific embodiments in conjunction with the accompanying drawings.

[0026] Referring to Figure 1 , Figure 1 A flowchart of a data processing method provided by an embodiment of the present disclosure is shown in FIG. 1. The method comprises the following steps.

[0027] S101: In response to receiving an automatic control instruction, determining a plurality of target standard information from a plurality of standard information based on room information.

[0028] In the embodiment, the automatic control instruction can be an instruction that the occupant has checked in. For example, the instruction can be that the room state is detected as a check-in state and the door opening information is received. The state of the room can be determined by the registration system of the hotel, and the door opening signal can be detected by the intelligent door lock. It is considered that when the room is in an idle state, the hotel cleaning personnel can also enter the room for cleaning. At this time, because the room state is in an idle state, the control will not be started at this time, thereby avoiding the waste of resources.

[0029] The room information can be the size, area, and orientation of the room. The standard information is the basic information of the historical personnel. It is considered that the size, orientation, and size of the room can all lead to different control methods in the room. Therefore, the standard information suitable for the room can be first determined from the plurality of standard information according to the room information, that is, the target standard information, that is, the basic information of the occupant who has checked in the room or the room of the same type. The standard information and the target standard information are both basic information of the occupant.

[0030] For example, when the room area is relatively small, the temperature rising and falling effect of the air conditioner is relatively good, and the occupant is inclined to frequently adjust the temperature of the air conditioner. When the room faces south, the sun is relatively glaring in the room, and the occupant is inclined to control the curtain to be partially closed during the day. When the size of the room is relatively luxurious, the occupant has a higher expectation for the quality of service and personalized experience, and the number of intelligent devices is different from that of the standard size room.

[0031] In this embodiment, detailed information of all rooms is obtained from the management system of the hotel, including room specifications such as deluxe suite, standard room, single room, etc.; area, square meters; house orientation, east, south, west, north, etc. At the same time, the basic information of the historical personnel is collected as standard information, which can include the age, gender, occupation, geographical origin, and frequency of the guest. If the current room is a deluxe suite, the basic information of the guest who has ever stayed in a deluxe suite should be selected as the target standard information. Because the guests of the deluxe suite may have higher expectations for the quality of service, equipment function, and personalized experience, which is different from the needs of ordinary room guests.

[0032] Different intervals are divided according to the size range of the room area. For example, the room area is divided into three intervals of less than 20 square meters, 20-30 square meters, and greater than 30 square meters. If the current room area is between 20-30 square meters, the basic information of the guest who has ever stayed in a room with similar area is selected as the target standard information.

[0033] Different matching rules are formulated for different house orientations. If the room is south-facing, the target standard information should focus on selecting the basic information of the guest who has ever stayed in a south-facing room and has a record of controlling sunlight into the room.

[0034] The basic information of the occupant who has ever stayed in the room can be used as the target standard information, or the basic information of the occupant who has ever stayed in a similar room can be used as the target standard information. Whether it is a similar room can be judged by calculating the similarity of the room features of the room and the room features of other rooms. The similarity of the room features of the room and the room features of other rooms can be calculated by the Euclidean distance method or the cosine similarity method, etc. The similarity of the room features of the two rooms, different calculation methods have different evaluation criteria, that is, the threshold value.

[0035] For example, in response to the value calculated by the Euclidean distance method being greater than the first distance threshold value, the room features of the room where the target personnel stays and the room features of the comparison room are not similar.

[0036] In response to the value calculated by the Euclidean distance method being less than or equal to the first distance threshold value, the room features of the room where the target personnel stays and the room features of the comparison room are similar, and the room where the target personnel stays is the room where the target personnel stays.

[0037] The first distance threshold value can be determined according to actual needs.

[0038] S102: Determine the first control model based on the matching degree of the basic information of the target personnel and the plurality of target standard information; the standard information is the basic information of the historical personnel.

[0039] In the embodiment, the target person is a hotel resident, and the basic information can include age, gender, occupation, and geographical origin, etc. The basic information of the resident can be obtained when the resident makes a reservation or registers at the hotel front desk. Since residents of different ages, different genders, different occupations, and different geographical origins have different feelings and adjustment levels for temperature, humidity, light intensity, etc., the control model cannot be determined based on only the above room information.

[0040] In the embodiment, the similarity between the basic information of the target person and each standard information is calculated as a matching degree, and the control model corresponding to the standard information with the highest matching degree is taken as the first control model. That is, each standard information corresponds to a control model, but one control model can correspond to multiple standard information. The control model is obtained by training a historical resident's basic information, behavior characteristics, and historical environmental characteristics of the resident's room.

[0041] The behavior characteristics refer to the behavior characteristics and habits of the historical resident in the room, which can reflect the needs and preferences of the guest for the room environment and device functions. For example, the behavior characteristics can include the guest's switching operation of the light at different time periods in a day, the frequency and amplitude of adjusting the light brightness, the temperature range set when using the air conditioner, and the number of times of adjusting the temperature, whether the guest often pulls or opens the curtain, and the time period of operation, etc. These behavior characteristics can be collected by the sensors (such as light sensors, temperature sensors, etc.) installed in the room and the operation data recorded by the intelligent device itself. The historical environmental characteristics are the historical temperature, humidity, light intensity, and external sound intensity of the room.

[0042] S103: input the environmental characteristics of the target room into the first control model to obtain the control strategy of the target room; the target room is the room where the target person resides.

[0043] In the embodiment, the target room is the room where the target person resides, and the first control model is the most suitable control model selected based on the target person and the target room. The first control model is selected from multiple control models according to the basic information of the target person and the room characteristics of the target room.

[0044] The environmental characteristics are the temperature, humidity, light intensity, and external sound intensity of the room, which can be obtained by the temperature sensor, humidity sensor, light intensity sensor, and sound intensity sensor installed in the room.

[0045] The control strategy is a specific operation instruction and adjustment mode given to each intelligent device in the target room based on the analysis and calculation of the first control model, so as to realize reasonable control of the room environment and meet the expectations and needs of the target person.

[0046] For example, the control strategy can be to set the air conditioner temperature to a specific degree, adjust the light brightness to a certain percentage, control the curtain to open or close to a certain extent, and whether to close the window, etc. It is a comprehensive and specific operation scheme for each device.

[0047] S104: Control the intelligent device in the target room based on the control strategy.

[0048] In this embodiment, the intelligent device is a device installed in the target room, which has intelligent functions such as automatic control, remote operation, and flexible adjustment of its own running state according to preset conditions or external instructions. The intelligent device can be a smart air conditioner, a smart curtain, a smart lighting system, a smart humidifier, and a smart window, etc.

[0049] This embodiment can control the corresponding intelligent device to perform the corresponding action according to the generated control strategy.

[0050] From the above, it can be concluded that the present disclosure can filter the target standard information that meets the room from the preset standard information library according to the room information, which narrows the range of filtering and also makes the filtered target standard information more consistent with the actual situation of the current room, thereby improving the pertinence of data processing. The present disclosure takes the environmental characteristics of the target room as input and outputs the control strategy through the first control model, so that the present disclosure can adjust the control strategy according to the real-time changing environmental characteristics, thereby ensuring that the environment of the target room is always in the best state, and improving the accuracy and reliability of data processing.

[0051] In an embodiment of the present disclosure, the basic information includes a plurality of dimensions of first information; the target standard information includes a plurality of dimensions of second information;

[0052] The number of dimensions of the target standard information is greater than or equal to the number of dimensions of the basic information;

[0053] Determine the first control model based on the matching degree of the basic information of the target person and the plurality of target standard information, comprising:

[0054] Calculate the matching degree of the first information of each dimension of the target person and the second information of the corresponding dimension respectively, to obtain a plurality of matching degrees of each dimension;

[0055] Weighted calculation is performed on the plurality of matching degrees of each dimension respectively to obtain a plurality of target matching degrees;

[0056] The control model corresponding to the target standard information with the highest matching degree is taken as the first control model.

[0057] In the embodiment, the basic information is a set of data describing the characteristics of the target person (i.e. the guest currently staying in the hotel room), which is described from multiple different angles or aspects, and the different angles are referred to as dimensions, and the specific data content under each dimension is the first information.

[0058] For example, for a guest staying in the hotel, the dimensions of the basic information can include age, gender, occupation, geographical origin, membership, etc., and each dimension has corresponding specific content to reflect the guest's situation in that aspect.

[0059] The target standard information is the information related to the room of the target person selected from the basic information of a plurality of historical persons. It is also described from multiple dimensions, and the corresponding specific content under each dimension is the second information. These dimensions are partially repeated with the dimensions of the basic information, but cover more dimensions, so the number of dimensions is greater than or equal to the number of dimensions of the basic information of the target person.

[0060] Because the target standard information is the information we want to select from, its dimensions should be more comprehensive and comprehensive, so that the dimensions of the basic information are greater than the dimensions of the target standard information. Since each guest fills in the reservation information differently, the comprehensive degree of the information reserved is different, i.e. the number of dimensions is different, so in the process of selection and training, the information of the historical occupants with comprehensive basic information should be selected.

[0061] The matching degree of the first information of each dimension of the target person and the second information of the corresponding dimension is calculated to obtain a plurality of matching degrees of each dimension, which means that each dimension should be matched with the information of the same dimension to calculate the matching degree, for example, the basic information includes age, gender and geographical origin, and the standard information includes age, gender, geographical origin and occupation. When calculating the matching degree, the corresponding dimension of the age dimension should be age, not gender, geographical origin, etc.

[0062] The calculation of the age matching degree can be performed by setting the matching degree corresponding to the age difference, for example, the age difference is between 0-1 years, the matching degree is 1, the age difference is set to 1-5 years, the matching degree is assigned to 0.8, the age difference is between 5-10 years, the matching degree is assigned to 0.6, and the difference is greater than 10 years, the matching degree is assigned to 0.2.

[0063] The calculation of the gender matching degree defines the same gender as 1 and different gender as 0.

[0064] The professional matching degree and the regional source matching degree can be obtained through a preset mapping table. For example, the mapping table for the regional source matching degree includes the matching degrees between various regions, as shown in Table 1.

[0065] Table 1: Mapping table for regional source matching degree (part)

[0066]

[0067] The mapping table described above is only a part. A complete mapping table should have the matching degrees of all cities or be determined through certain rules. For example, the matching degree of the same city is 1, the regional matching degree of the same province but different cities is 0.8, and so on. Alternatively, the matching degree can be measured by cosine similarity or Euclidean distance method.

[0068] After obtaining the matching degree corresponding to each dimension, the matching degree of the basic information and each standard information can be obtained through weighted calculation. The control model corresponding to the target standard information with the highest matching degree can be taken as the first control model. In the weighted calculation, the weight can be allocated by experience or adjusted by calculating the correlation between each dimension and the behavior characteristics. For example:

[0069] determining an initial weight corresponding to each dimension;

[0070] increasing the weight corresponding to the dimension with a correlation greater than a first correlation threshold to a second weight according to a first correlation step, and decreasing the weights corresponding to other dimensions to a plurality of third weights according to a plurality of second correlation steps.

[0071] The initial weight corresponding to each dimension can be set in advance. When the correlation between a certain dimension and the behavior characteristics is greater than the first correlation threshold, it means that the degree of association between this dimension and the behavior characteristics is relatively large. Therefore, this dimension should be considered more in the weighted calculation. The correlation between the dimension and the behavior characteristics can be obtained by calculating the Pearson correlation coefficient. It should be noted that the absolute value obtained by the Pearson correlation coefficient calculation should be taken as the correlation value.

[0072] The first correlation threshold, the first correlation step and the second correlation step can be obtained by experiment. The first correlation step is equal to the sum of all second correlation steps, which aims to ensure that the weight after adjustment is still 1.

[0073] From the above, the present disclosure can be concluded that by dividing both the basic information and the target standard information into multiple dimensions and calculating the matching degree of each dimension respectively, the embodiment can more comprehensively capture the similarity between the target personnel and the target standard information, help to more accurately reflect the individualized needs of the target personnel, and thus provide a control model that better meets the needs of the target personnel. The present disclosure can more flexibly adjust the influence of each dimension in the decision-making process by assigning different weights to different dimensions, thereby enhancing the accuracy and rationality of the decision-making, and improving the accuracy and reliability of data processing.

[0074] In an embodiment of the present disclosure, the first control model is a support vector machine model trained according to historical environmental characteristics, historical personnel basic information, and historical personnel historical behavior characteristics;

[0075] Before determining the plurality of target standard information from the plurality of standard information based on the room characteristics in response to receiving the automatic control instruction, further comprising:

[0076] Determining a kernel function of the support vector machine model based on the historical environmental characteristics and the historical personnel historical behavior characteristics;

[0077] Determining a penalty parameter of the support vector machine model based on the historical environmental characteristics and the historical personnel historical behavior characteristics;

[0078] Training the support vector machine model based on the historical environmental characteristics, the historical personnel basic information, and the historical personnel historical behavior characteristics to obtain a plurality of control models; the first control model is one of the plurality of control models.

[0079] In an embodiment of the present disclosure, determining a kernel function of the support vector machine model based on the historical environmental characteristics and the historical personnel historical behavior characteristics comprises:

[0080] In response to the absolute value of the correlation coefficient of the historical environmental characteristics and the behavior characteristics of the historical personnel being greater than a first correlation threshold, the first type of kernel function is used as the kernel function of the support vector machine model;

[0081] In response to the absolute value of the correlation coefficient of the historical environmental characteristics and the behavior characteristics of the historical personnel being less than or equal to the first correlation threshold, the second type of kernel function is used as the kernel function of the support vector machine model;

[0082] The first type of kernel function and the second type of kernel function have different degrees of linear relationship.

[0083] In an embodiment of the present disclosure, determining a penalty parameter of the support vector machine model based on the historical environmental characteristics and the historical personnel historical behavior characteristics comprises:

[0084] Determining an initial penalty function;

[0085] In response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being greater than a second correlation threshold, the initial penalty function is increased by a first penalty step to obtain the penalty function of the support vector machine model;

[0086] In response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being less than or equal to the second correlation threshold, the initial penalty function is decreased by a second penalty step to obtain the penalty function of the support vector machine model.

[0087] In this embodiment, the first control model is a support vector machine model trained according to historical environment features, historical personnel basic information, and historical personnel behavior features. SVM has a penalty parameter and multiple kernel functions, which can include a linear kernel function, a polynomial kernel function, and a radial basis function kernel function. The linear kernel function mainly processes linearly related data, and the polynomial kernel function and the radial basis function kernel function mainly process non-linearly related data.

[0088] The penalty parameter is used to balance the fitting ability and generalization ability of the model. If the penalty parameter is too large, the model will try to correctly classify all samples in the training data, which may lead to overfitting. For example, for occasional abnormal device operations of guests (which may be misoperations), if the penalty parameter is too large, the model will learn these abnormal situations as normal behavior patterns. Conversely, when the penalty parameter is too small, the model may be too simple and unable to accurately capture the complex relationship between environmental factors and guest behavior, resulting in underfitting. For example, in a temperature control model, if the penalty parameter is too small, the model may not be able to learn the different temperature needs of guests in different seasons and weather conditions, and may simply adopt a fixed control strategy.

[0089] Therefore, the selection of the kernel function and the penalty parameter is related to the correlation between the environmental features and the behavior features. When the correlation between the environmental features and the behavior features is high, a linear kernel function can be used as the kernel function of the SVM to better express the relationship between the environmental features and the behavior features, and a higher penalty parameter can be selected to make the model fit the data more strictly and minimize misclassification. Conversely, when the correlation between the environmental features and the behavior features is low, a non-linear kernel function can be used as the kernel function of the SVM to make the model focus more on generalization ability and make the model relatively simple.

[0090] Therefore, this embodiment sets a detailed adjustment method, that is, by comparing the correlation coefficient of the historical environment features and the historical personnel behavior features with a predetermined first correlation threshold, a suitable kernel function is selected, wherein the first type of kernel function is a linear kernel function, and the second type of kernel function is a non-linear kernel function, and the linear relationship degrees of the two types of kernel functions are different.

[0091] The correlation coefficient of the historical environment feature and the historical personnel behavior feature can be obtained by calculating the Pearson correlation coefficient. It should be noted that the result obtained by the Pearson correlation coefficient calculation has a value range of The closer to 1 indicates that the data is positively correlated to a greater extent; the closer to -1, the greater the negative correlation of the data; the closer to 0, the lower the correlation of the data.

[0092] The first correlation threshold and the second correlation threshold can be obtained by multiple tests of historical data. For example, when the first correlation threshold is greater than 0.8, the effect of using a linear kernel function is better, and when it is less than 0.8, the effect of a nonlinear kernel function is better. Therefore, the first correlation threshold can be 0.8, and the judgment of the second correlation threshold is the same.

[0093] The initial penalty parameter of the SVM can be a value with high applicability, or an initial value preliminarily determined based on the complexity of the data. For example, when the number or dimension of the data is greater than a preset value, the initial penalty parameter can be assigned a higher value to make the output more accurate.

[0094] The first penalty step and the second penalty step can be determined according to the actual situation, or according to the difference between the absolute value of the correlation coefficient and the second correlation threshold. For example, the first penalty step is determined based on the difference between the correlation coefficient of the historical environment feature and the historical personnel behavior feature and the second correlation threshold, and the first linear formula.

[0095] The second penalty step is determined based on the difference between the correlation coefficient of the historical environment feature and the historical personnel behavior feature and the second correlation threshold, and the second linear formula.

[0096] The first linear formula can be: The second linear formula can be: wherein represents the first penalty step, represents the second penalty step, both represent the slope, which can be determined according to the actual situation, represents the difference between the correlation coefficient of the historical environment feature and the historical personnel behavior feature and the second correlation threshold. Through the first linear formula and the second linear formula, the first penalty step and the second penalty step can be dynamically adjusted according to the size of the correlation coefficient.

[0097] From the above, the present disclosure can intelligently select a linear kernel function or a nonlinear kernel function as the kernel function of the SVM by comparing the correlation coefficient of the historical environmental features and the historical personnel behavior features with the first correlation threshold, and further enable the model to better adapt to data features in different situations, thereby improving the generalization ability and adaptability of the model. By comparing the correlation coefficient with the second correlation threshold and adjusting the penalty function accordingly, the present embodiment can balance the fitting ability and generalization ability of the model, avoid overfitting or underfitting, and enable the model to more flexibly cope with data of different complexity, thereby improving the robustness and stability of the model. By calculating the correlation coefficient and selecting and adjusting the model accordingly, the present embodiment can more accurately capture the internal relationship between the data, thereby improving the prediction performance and adaptability of the model.

[0098] In an embodiment of the present disclosure, in response to the absolute value of the correlation coefficient of the historical environmental features and the historical personnel behavior features being less than or equal to the first correlation threshold, a second type of kernel function is selected as the kernel function of the support vector machine model, including:

[0099] In response to the historical environmental features and the historical personnel behavior features satisfying the polynomial condition, a first sub-kernel function is selected as the kernel function of the support vector machine model.

[0100] In response to the historical environmental features and the historical personnel behavior features not satisfying the polynomial condition, a second sub-kernel function is selected as the kernel function of the support vector machine model.

[0101] The linear relationship of the first sub-kernel function and the second sub-kernel function is different.

[0102] In an embodiment of the present disclosure, an initial width parameter of the second sub-kernel function is determined.

[0103] In response to the historical environmental features and the historical personnel behavior features not satisfying the polynomial condition and the historical environmental features and the historical personnel behavior features satisfying the first distribution condition, the initial width parameter is reduced by a first width step.

[0104] In response to the historical environmental features and the historical personnel behavior features not satisfying the polynomial condition and the historical environmental features and the historical personnel behavior features not satisfying the first distribution condition, the initial width parameter is increased by a second width step.

[0105] In the embodiment, the second type of kernel function includes a polynomial kernel function and a radial basis function kernel function, the first sub-kernel function is a polynomial kernel function, and the second sub-kernel function is a radial basis function kernel function. Both of the two kernel functions are nonlinear kernel functions, but the polynomial kernel function is biased towards data with a polynomial distribution. The polynomial condition can be determined by drawing a two-dimensional or three-dimensional scatter plot of the environmental characteristics and the behavior characteristics and a trained convolutional neural network model to determine whether the scatter plot meets the polynomial feature. The convolutional neural network model is trained by a large number of two-dimensional and three-dimensional scatter plots meeting the polynomial distribution.

[0106] In the embodiment, the polynomial condition can also be determined by calculating a goodness-of-fit index. For example, a determination coefficient is calculated. The closer the determination coefficient is to 1, the better the polynomial model fits the data. If the determination coefficient of the first fitting degree is less than the first fitting threshold, the polynomial condition is not met. The first fitting degree and the first fitting threshold can be determined according to actual conditions.

[0107] If the SVM model is the second sub-kernel function, i.e., the radial basis function kernel function, an initial width parameter can be determined first, and the initial width parameter can be adjusted according to the distribution of the historical environmental characteristics and the historical personnel behavior characteristics.

[0108] In the embodiment, the expression of the radial basis function kernel function is: , and the width parameter is , represents the historical environmental characteristics, represents the historical personnel behavior characteristics, and the width parameter determines the influence range of each data point. If the hotel room data is relatively dispersed, for example, different guests have different requirements for temperature, and the data points are widely distributed in the feature space, the width parameter can be appropriately reduced. Because a smaller value will make the influence range of the data points larger, the model will pay more attention to the overall data distribution, which is helpful for processing such dispersed data. The first width step can be determined according to experience.

[0109] On the contrary, if the data distribution is relatively concentrated, for example, the operation behavior of a guest on a certain device is very similar under similar environmental conditions, and the data points are relatively concentrated, the width parameter can be appropriately increased. In this way, the model can pay more attention to the local data features and better fit such concentrated data.

[0110] The density of data points in the entire feature space can be calculated by statistical methods. For example, the space composed of environmental features (temperature, humidity, light intensity) and behavior features (device usage frequency, device combination usage) is divided into multiple small regions, and the number of data points in each small region is calculated. If the number of data points in each small region is relatively uniform, and there is no obvious high-density peak region, it can be preliminarily judged that the data is dispersed. On the contrary, if the density of data points in some small regions is obviously higher than that in other regions, forming a high-density region, it may mean that the data is concentrated in these regions.

[0111] For example, the feature space is divided into a first number of spaces to obtain a plurality of second feature spaces;

[0112] The density of data points in each second feature space is calculated respectively;

[0113] In response to the difference between the maximum density and the minimum density of data points in the plurality of second feature spaces being greater than or equal to a first density threshold, the historical environmental features and the historical personnel behavior features satisfy a first distribution condition;

[0114] In response to the difference between the maximum density and the minimum density of data points in the plurality of second feature spaces being less than the first density threshold, the historical environmental features and the historical personnel behavior features do not satisfy the first distribution condition.

[0115] The first number of divisions and the first density threshold can be determined according to experiments. The feature space is a space composed of historical environmental features and historical personnel behavior features.

[0116] In this embodiment, the data points in the feature space can also be clustered by a clustering algorithm, for example:

[0117] The data points in the feature space are clustered based on the clustering algorithm to obtain a clustering result;

[0118] In response to the number of clusters in the clustering result being greater than a first clustering threshold, the historical environmental features and the historical personnel behavior features satisfy a first distribution condition;

[0119] In response to the number of clusters in the clustering result being less than or equal to the first clustering threshold, the historical environmental features and the historical personnel behavior features do not satisfy the first distribution condition.

[0120] When the number of clusters in the clustering result is large, i.e. greater than the first clustering threshold, it indicates that the data distribution is relatively dispersed, and vice versa. The first clustering threshold can be determined according to actual conditions.

[0121] From the above, the present disclosure can dynamically adjust the width parameter by calculating the density and clustering results of the data points, so that the model can better fit the local or overall characteristics of the data, and the model can maintain optimal performance under different data distribution conditions. The present disclosure determines whether the polynomial condition and the first distribution condition are met in multiple ways, thereby improving the accuracy and reliability of data processing.

[0122] A data processing method corresponding to the above embodiment, Figure 2 A structural block diagram of a data processing system is provided for an embodiment of the present disclosure. For ease of illustration, only parts related to the embodiments of the present disclosure are shown. For reference Figure 2 The data processing system 20 includes a standard information determination module 21, a model determination module 22, a control strategy output module 23, and a control module 24.

[0123] The standard information determination module 21 is configured to determine a plurality of target standard information from a plurality of standard information based on room information in response to receiving an automatic control instruction.

[0124] The model determination module 22 is configured to determine a first control model based on a matching degree of the basic information of the target person and the plurality of target standard information; the standard information is the basic information of the historical person.

[0125] The control strategy output module 23 is configured to input the behavior characteristics of the target person in the target room into the first control model to obtain a control strategy of the target room; the target room is the room where the target person lives.

[0126] The control module 24 is configured to control the intelligent device in the target room based on the control strategy.

[0127] In an embodiment of the present disclosure, the target standard information includes a plurality of dimensions of second information.

[0128] The number of dimensions of the target standard information is greater than or equal to the number of dimensions of the basic information.

[0129] The model determination module 22 is specifically configured to calculate the matching degree of the first information of each dimension of the target person and the second information of the corresponding dimension respectively to obtain a plurality of matching degrees of each dimension.

[0130] The plurality of matching degrees of each dimension are respectively weighted to obtain a plurality of target matching degrees.

[0131] The control model corresponding to the target standard information with the highest target matching degree is taken as the first control model.

[0132] In an embodiment of the present disclosure, the first control model is a support vector machine model trained according to the historical environment features, the historical personnel basic information and the historical personnel historical behavior features.

[0133] The data processing system 20 further comprises a model construction module.

[0134] The model construction module is configured to determine a kernel function of the support vector machine model based on the historical environment features and the historical personnel historical behavior features.

[0135] The model construction module is configured to determine a penalty parameter of the support vector machine model based on the historical environment features and the historical personnel historical behavior features.

[0136] The model construction module is configured to train the support vector machine model based on the historical environment features, the historical personnel basic information and the historical personnel historical behavior features to obtain a plurality of control models; and the first control model is one of the plurality of control models.

[0137] In an embodiment of the present disclosure, the model construction module is specifically configured to, in response to an absolute value of a correlation coefficient of the historical environment features and the historical personnel behavior features being greater than a first correlation threshold, take a first type of kernel function as the kernel function of the support vector machine model.

[0138] In response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being less than or equal to the first correlation threshold, take a second type of kernel function as the kernel function of the support vector machine model.

[0139] The first type of kernel function and the second type of kernel function are different in linear relationship degree.

[0140] In an embodiment of the present disclosure, the model construction module is specifically further configured to determine an initial penalty function.

[0141] In response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being greater than a second correlation threshold, increase the initial penalty function by a first penalty step to obtain the penalty function of the support vector machine model.

[0142] In response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being less than or equal to the second correlation threshold, decrease the initial penalty function by a second penalty step to obtain the penalty function of the support vector machine model.

[0143] In an embodiment of the present disclosure, the model construction module is specifically further configured to, in response to the historical environment features and the historical personnel behavior features satisfying a polynomial condition, take a first sub-kernel function as the kernel function of the support vector machine model.

[0144] In response to the fact that the characteristics of the historical environment and the behavioral characteristics of historical personnel do not satisfy the polynomial condition, the second sub-kernel function is used as the kernel function of the support vector machine model;

[0145] The linear relationships between the first and second sub-kernel functions are different.

[0146] In one embodiment of this disclosure, the model building module is further configured to determine the initial width parameter of the second sub-kernel function;

[0147] In response to the fact that the characteristics of the historical environment and the behavior of historical personnel do not satisfy the polynomial condition, and the characteristics of the historical environment and the behavior of historical personnel satisfy the first distribution condition, the initial width parameter is reduced according to the first width step size.

[0148] In response to the fact that the characteristics of the historical environment and the behavior of historical personnel do not satisfy the polynomial condition, and that the characteristics of the historical environment and the behavior of historical personnel do not satisfy the first distribution condition, the initial width parameter is increased according to the second width step size.

[0149] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of the present disclosure. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of each module / unit in the above system embodiments, for example... Figure 2 The functions of modules 21 to 24 are shown.

[0150] It should be understood that, in the embodiments of this disclosure, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0151] The input device 302 can include a touchpad, a fingerprint collection sensor (for collecting fingerprint information and direction information of a fingerprint of a user), a microphone, and the like, and the output device 303 can include a display (LCD, etc.), a speaker, and the like.

[0152] The memory 304 can include a read-only memory and a random access memory, and provide the processor 301 with instructions and data. A portion of the memory 304 can also include a non-volatile random access memory. For example, the memory 304 can also store device type information.

[0153] In specific implementations, the processor 301, the input device 302, and the output device 303 described in the embodiments of the present disclosure can perform the implementation manners described in the first and second embodiments of the data processing method provided by the embodiments of the present disclosure, and can also perform the implementation manners of the electronic device described in the embodiments of the present disclosure, which will not be described here.

[0154] In another embodiment of the present disclosure, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program includes program instructions, and the program instructions are executed by a processor to implement all or part of the processes of the above-mentioned embodiments. The computer program can also be used to instruct related hardware to complete, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0155] The computer readable storage medium can be an internal storage unit of the electronic device of any of the preceding embodiments, such as a hard disk or a memory of the electronic device. The computer readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, or the like. Further, the computer readable storage medium can include both the internal storage unit and the external storage device of the electronic device. The computer readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer readable storage medium can also be used to temporarily store data that has been output or will be output.

[0156] Those skilled in the art can understand that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the foregoing description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present disclosure.

[0157] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic device and the units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0158] In several embodiments provided in the present application, it should be understood that the disclosed electronic device and method can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of units is only a logical function division, and actual implementation can have another division manner. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces or units, and can also be electrical, mechanical or other forms of connection.

[0159] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. can be located in one place or can be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0160] In addition, each of the functional units in the various embodiments of the present disclosure can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0161] The above is only a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present disclosure, and these modifications or replacements should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A data processing method, characterized by, The method comprises the following steps: in response to an absolute value of a correlation coefficient of historical environment features and historical personnel behavior features being greater than a first correlation threshold, a first type of kernel function is used as a kernel function of a support vector machine model; in response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being less than or equal to the first correlation threshold, a second type of kernel function is used as the kernel function of the support vector machine model; the first type of kernel function and the second type of kernel function have different degrees of linearity; an initial penalty function is determined; in response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being greater than a second correlation threshold, the initial penalty function is increased by a first penalty step to obtain a penalty function of the support vector machine model; in response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being less than or equal to the second correlation threshold, the initial penalty function is decreased by a second penalty step to obtain the penalty function of the support vector machine model; a plurality of control models are obtained by training a support vector machine model based on the historical environment features, basic information of historical personnel and historical behavior features of the historical personnel; in response to receiving an automatic control instruction, a plurality of target standard information is determined from a plurality of standard information based on room information; a first control model is determined based on a matching degree of basic information of a target personnel and the plurality of target standard information; the standard information is basic information of historical personnel; the first control model is a support vector machine model obtained by training based on historical environment features, basic information of historical personnel and historical behavior features of historical personnel; the first control model is one of the plurality of control models; an environment feature of a target room is input into the first control model to obtain a control strategy of the target room; the target room is a room in which the target personnel resides; the target room is controlled based on the control strategy.

2. A data processing method as claimed in claim 1, characterized in that, The basic information comprises a plurality of dimensions of first information; the target standard information comprises a plurality of dimensions of second information; a number of dimensions of the target standard information is greater than or equal to a number of dimensions of the basic information; The first control model is determined based on a matching degree of basic information of a target personnel and the plurality of target standard information, comprising: a matching degree of first information of each dimension of the target personnel and second information of the corresponding dimension is calculated respectively to obtain a plurality of matching degrees of each dimension; a plurality of target matching degrees are obtained by weighted calculation of the plurality of matching degrees of each dimension; a control model corresponding to target standard information with the highest target matching degree is used as the first control model.

3. A data processing method as claimed in claim 1, characterized in that, The second type of kernel function is used as the kernel function of the support vector machine model in response to the absolute value of the correlation coefficient of the historical environment features and the historical personnel behavior features being less than or equal to the first correlation threshold, comprising: in response to the historical environment features and the historical personnel behavior features satisfying a polynomial condition, a first sub-kernel function is used as the kernel function of the support vector machine model. in response to the historical environment feature and the historical personnel behavior feature not satisfying the polynomial condition, taking the second sub-kernel function as a kernel function of the support vector machine model; the linear relationship of the first sub-kernel function and the second sub-kernel function is different.

4. A data processing method as claimed in claim 3, characterized in that, Further comprising: determining an initial width parameter of the second sub-kernel function; in response to the historical environment feature and the historical personnel behavior feature not satisfying the polynomial condition, and the historical environment feature and the historical personnel behavior feature satisfying a first distribution condition, reducing the initial width parameter by a first width step; in response to the historical environment feature and the historical personnel behavior feature not satisfying the polynomial condition, and the historical environment feature and the historical personnel behavior feature not satisfying the first distribution condition, increasing the initial width parameter by a second width step.

5. A data processing system, characterized by comprising: a model construction module, configured to, in response to an absolute value of a correlation coefficient of a historical environment feature and a historical personnel behavior feature being greater than a first correlation threshold, take a first type of kernel function as a kernel function of a support vector machine model; in response to the absolute value of the correlation coefficient of the historical environment feature and the historical personnel behavior feature being less than or equal to the first correlation threshold, take a second type of kernel function as the kernel function of the support vector machine model; the degree of linear relationship of the first type of kernel function and the second type of kernel function is different; determining an initial penalty function; in response to the absolute value of the correlation coefficient of the historical environment feature and the historical personnel behavior feature being greater than a second correlation threshold, increasing the initial penalty function by a first penalty step to obtain a penalty function of the support vector machine model; in response to the absolute value of the correlation coefficient of the historical environment feature and the historical personnel behavior feature being less than or equal to the second correlation threshold, reducing the initial penalty function by a second penalty step to obtain the penalty function of the support vector machine model; training the support vector machine model based on the historical environment feature, basic information of a historical personnel and historical behavior features of the historical personnel to obtain a plurality of control models; a standard information determination module, configured to, in response to receiving an automatic control instruction, determine a plurality of target standard information from a plurality of standard information based on room information; a model determination module, configured to determine a first control model based on a matching degree of basic information of a target personnel and the plurality of target standard information; the standard information is basic information of a historical personnel; the first control model is a support vector machine model trained based on a historical environment feature, basic information of a historical personnel and historical behavior features of the historical personnel; the first control model is one of the plurality of control models; a control strategy output module, configured to input a behavior feature of the target personnel in a target room into the first control model to obtain a control strategy of the target room; the target room is a room in which the target personnel resides; a control module, configured to control intelligent devices in the target room based on the control strategy.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 4.

Images