Method for evaluating intelligence degree of smart home equipment

Abstract

The invention provides a method capable of comprehensively evaluating the intelligent degree of household intelligent home equipment, and the method comprises the steps: reasonably selecting architecture parameters capable of reflecting the intelligent degree; data required by architecture parameters can be effectively obtained by collecting and analyzing multi-dimensional data such as categories, quantity and scores of household intelligent household equipment and App application use conditions, and meanwhile, intelligent degree scores required by model training are set in a reasonable mode, so that the intelligent degree of the intelligent household equipment is improved. And finally, a linear model is adopted to construct a scoring model, so that the development difficulty can be effectively reduced, the evaluation accuracy can be improved, and a convenient and effective evaluation means is provided for subsequent users, operators and the like. According to the method, whether the user uses the intelligent equipment or not can be judged through the App accessed by the user, a relatively reasonable score is made for a family of which the equipment model cannot be captured according to the App, and two sets of rating standards are further adopted, so that the matching between the score and the rating can be effectively ensured, and the reliability of an output result is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home, and in particular to a method for evaluating the intelligence level of smart home equipment. Background Art

[0002] With the development of the Internet of Things and artificial intelligence technologies, smart home devices are becoming increasingly popular in people's lives.

[0003] Scientific and comprehensive evaluation of the intelligence level of home smart home devices is the basis for users and smart home device operators to optimize and improve home smart home devices.

[0004] In the existing technology, an available method is to construct a scoring model and perform scoring based on the corresponding basic data. The type of model used and the architectural parameters of the model determine the training difficulty, development cost and subsequent scoring accuracy of the scoring model. However, the existing technology lacks an intelligent degree assessment model that can be used as a reference.

[0005] In addition, the device model data of home smart home devices is an indispensable basic data for building scoring models and scoring. Although the required data can be obtained by manually collecting the device model data of each household, this data collection method is difficult to obtain and has a limited data collection scope, which is prone to inaccurate and insufficient data collection. When network data is used to substitute the trained scoring model for scoring, there is a situation where the device model cannot be effectively obtained, which leads to inaccurate scoring and rating. Summary of the invention

[0006] The purpose of the present invention is to solve the above problems existing in the prior art and to provide a method for evaluating the intelligence level of smart home devices.

[0007] The purpose of the present invention is achieved through the following technical solutions:

[0008] A method for evaluating the intelligence level of a smart home device comprises the following steps:

[0009] S1, obtain network data of a household to be evaluated;

[0010] S2, determining a set of values ​​of intelligence rating parameters of the household to be evaluated according to the acquired network data;

[0011] S3, determining the intelligence score of the household to be evaluated according to the values ​​of a set of intelligence score parameters of the household to be evaluated and a scoring model, and determining the intelligence level of the smart home device according to the intelligence score, wherein the smart home device intelligence level assessment has two sets of rating standards;

[0012] The scoring model is determined by the following process:

[0013] S10, obtaining network data of multiple households;

[0014] S20, determining a set of intelligence rating parameters corresponding to each household according to the network data and device rating rules of each household; the set of intelligence rating parameters includes the number of device categories, the number of human-machine devices, the number of IoT devices, the number of common smart home operation apps used, the number of other apps used, the human-machine device rating, and the IoT device rating;

[0015] S30, sorting the families according to the values ​​of the intelligence level scoring parameters of the families and assigning an intelligence level score to each family according to the sorting;

[0016] S40, determining a relationship between the intelligence score parameter and the intelligence score as the scoring model according to the intelligence score assigned to each family and the value of the corresponding intelligence score parameter;

[0017] In S1 and S10, the topology information reported by the plug-in is obtained according to the MAC address of the home gateway and the data in the networking device, and the network topology relationship between the home gateway corresponding to the MAC address and its downstream devices is obtained to obtain the network data;

[0018] In S2, it is determined whether the acquired network data contains the device models of all the devices connected to the home gateway of the home to be evaluated. If not, the values ​​of the number of device categories, the number of human-machine devices, the number of IoT devices, the human-machine device scores and the IoT device scores are set to zero, and the values ​​of the number of common smart home operation apps used and the number of other apps used are calculated as the first data set; if yes, the values ​​of the number of device categories, the number of human-machine devices, the number of IoT devices, the number of common smart home operation apps used, the number of other apps used, the human-machine device scores and the IoT device scores are calculated according to the acquired network data and used as the second data set;

[0019] In S3, which set of rating standards to select is determined according to the data based on which the intelligence level score is calculated. When the intelligence level score is calculated based on the first data set, the first set of rating standards is used; otherwise, the second set of rating standards is used.

[0020] Preferably, the human-machine device score and the IoT device score are determined according to the following formula:

[0021]

[0022] Among them, score is the score of human-machine equipment or IoT equipment, A is the basic score of human-machine equipment or IoT equipment, and x is the basic score of human-machine equipment or IoT equipment. iis the additional score of the i-th human-machine device or IoT device among all human-machine devices or IoT devices in a household; n is the number of human-machine devices or IoT devices.

[0023] Preferably, the additional points are determined according to the device attribute table and device scoring rules.

[0024] Preferably, in S2, if it is determined that the values ​​of the number of human-machine devices and the number of IoT devices are respectively greater than 100, the values ​​of the number of human-machine devices and the number of IoT devices are respectively set to 100.

[0025] Preferably, S30 includes the following steps:

[0026] S310, sorting the households in descending order according to the sum of the number of device categories and the number of common smart home operation apps used, the sum of the number of human-machine devices and the number of IoT devices, the sum of the human-machine device scores and the IoT device scores, and the number of other apps used;

[0027] S320, assigning an intelligence score to each household according to the ranking, and the intelligence score assigned to each household decreases in sequence according to the ranking.

[0028] Preferably, before S310, the acquired values ​​of the intelligence score parameters of all households are standardized.

[0029] Preferably, the scoring model is a linear model.

[0030] Preferably, the scoring model is as follows:

[0031] S=a*X1+b*X2+c*X3+d*X4+e*X5+f*X6+g*X7;

[0032] Among them, S is the intelligence level score, X1 is the number of device categories, X2 is the number of common smart home operation apps used, X3 is the number of human-machine devices, X4 is the number of IoT devices, X5 is the human-machine device score, X6 is the IoT device score, X7 is the number of other apps used, and a, b, c, d, e, f, and g are the weight coefficients corresponding to each intelligence level score parameter.

[0033] Preferably, the scoring model is as follows:

[0034] S=9.7801986*X1+3.9060424*X2+0.1953693*X3+0.2725942*X4+0.2478278*X5+0.1521893*X6+0.0425248*X7.

[0035] The advantages of the technical solution of the present invention are mainly reflected in:

[0036] The present invention provides a method for comprehensively evaluating the intelligence level of home smart home devices, by reasonably selecting architecture parameters that can reflect the intelligence level, and by collecting and analyzing multi-dimensional data such as the category, quantity, rating, and App application usage of home smart home devices, so as to ensure that the data required for the architecture parameters can be effectively obtained, and at the same time, by obtaining the intelligence level score required for model training in a reasonable way, and finally using a linear model to construct a scoring model, it can effectively reduce the development difficulty and improve the accuracy of the evaluation, and provide a convenient and effective means for subsequent users, operators, etc. to evaluate the intelligence level of home smart home devices, which is convenient for industry research, market analysis, and user self-diagnosis. At the same time, after obtaining the network data of the home to be evaluated, the present invention can determine whether the user uses the intelligent device through the App application accessed by the user when reliable device model data cannot be obtained, so that a more reasonable score can be given to the home that cannot capture the device model, and two sets of rating standards are adopted, which can effectively ensure the matching of the score and the rating, and improve the reliability of the output results. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of constructing the scoring model of the present invention;

[0038] Figure 2 It is a process schematic diagram of the scoring method of the present invention;

[0039] Figure 3 It is a schematic diagram of the process of calculating the intelligence degree score and performing the rating by using different scoring parameters in the scoring method of the present invention. DETAILED DESCRIPTION

[0040] The purpose, advantages and features of the present invention will be illustrated and explained by the non-limiting description of the following preferred embodiments. These embodiments are only typical examples of the application of the technical solution of the present invention, and any technical solution formed by equivalent replacement or equivalent transformation falls within the scope of protection claimed by the present invention.

[0041] In the description of the scheme, it should be noted that the terms "center", "upper", "lower", "left", "right", "front", "back", "vertical", "horizontal", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience and simplification of description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0042] A method for evaluating the intelligence level of smart home devices disclosed in the present invention is described below in conjunction with the accompanying drawings. The method is based on the network data of home users obtained to construct a scoring model and use the scoring model to score the intelligence level of smart home devices and to rate the intelligence level based on the intelligence level score.

[0043] As attached Figure 1 As shown, the scoring model is determined by the following process:

[0044] S10, obtaining network data of multiple households;

[0045] S20, determining a set of intelligence rating parameter values ​​corresponding to each household according to the network data and device rating rules of each household;

[0046] S30, sorting the families according to the values ​​of the intelligence level scoring parameters of the families and assigning an intelligence level score to each family according to the sorting;

[0047] S40, determining a relationship between the intelligence score parameter and the intelligence score as the scoring model according to the intelligence score assigned to each family and the value of the corresponding intelligence score parameter.

[0048] In S10, according to the MAC address of each home gateway and the topology information reported by the data acquisition plug-in in the networking device, the network topology relationship between the home gateway and its downstream devices corresponding to each MAC address is obtained to obtain the network data. The topology information reported by the data acquisition plug-in of the networking device will be stored locally and / or in the cloud, and then the network data corresponding to each home can be queried through the MAC address of each home gateway. The network data of each home can be data for one week, half a month or one month, which can be determined according to needs. The network data of each home obtained includes home gateway information, downstream device information and application information. Among them, the home gateway information includes the MAC address of the home gateway, the number of downstream devices, etc.; the downstream device information includes the model, brand, operating system and other data of the downstream device that can be identified by the data acquisition plug-in. The application information includes at least the App ID (application identifier) ​​of the App application used by a family reported by the data acquisition plug-in.

[0049] In S20, after acquiring the network data of multiple households, firstly, a comprehensive check is performed on the network data of each household, and missing values, abnormal values ​​and duplicate records are identified and processed to ensure the integrity, consistency and rationality of the data set. In addition, in order to ensure the reliability of the basic data, the acquired network data of the household can be screened, for example, according to the number of downstream devices of the home gateway and the number of downstream devices whose device models are acquired, it is determined whether there is a situation where the device model is not acquired in the downstream devices of a home gateway. If so, the network data of the household is deleted and is not used as data for scoring model training.

[0050] After data processing, the processed data is aggregated by household to obtain the values ​​of the set of intelligence score parameters for each household.

[0051] The set of intelligence rating parameters includes the number of device categories, the number of human-machine devices, the number of IoT devices, the number of common smart home operation apps used, the number of other apps used, the human-machine device rating and the IoT device rating. Among them, the device categories do not include routers, and the number of device categories refers to the number of types of devices under a home gateway. For example, if the devices under a home gateway include mobile phones, TVs, tablets, cameras, and speakers, the number of device categories is 5. The human-machine devices refer to mobile phones, computers, and tablets, and the IoT devices refer to other smart home devices other than routers and human-machine devices, such as cameras, speakers, smart lights, air conditioners, etc. The scope of the common smart home operation apps includes 52 App applications such as Mijia, Xiaotiancai, Migu Monitoring, Wukong Remote Control, Monitoring Eye, Ali Smart, Haier Youjia, Lecheng, Midea Home, and Huawei Smart Life. App applications beyond these 52 belong to other App applications.

[0052] For example, a router, a mobile phone, a laptop computer, and a speaker are connected to a home gateway, and the data acquisition plug-in can identify the model of each device connected to the home gateway. At the same time, the devices connected to the home gateway use two common smart home operation apps, Mijia and Wukong remote control. After the network data of the family is collected through the data acquisition plug-in, it can be determined that the number of device categories of the devices connected to the home network management is 3, the number of human-machine devices is 2, and the number of IoT devices is 1. At the same time, the human-machine device score can be determined based on the device information of the mobile phone and the laptop computer, and the IoT device score can be determined through the device information of the speaker. In addition, according to the App IDs of the two App applications used by the family, the App application corresponding to the App ID can be queried according to the App application and App ID correspondence table, so that it can be determined that the family uses the two App applications, Mijia and Wukong remote control, and the number of common smart home operation apps used is 2, and the number of other apps used is 0.

[0053] When determining the human-machine device score and the IoT device score, it is determined based on the score of each human-machine device and the score of the IoT device. The human-machine device score and the IoT device score are calculated based on a determined device scoring rule. The device scoring rule configures a basic score for each smart home device, which is set to 60 points in this embodiment. At the same time, additional points are configured for different smart home devices. The basic score and the additional score of each smart home device are added together to obtain the score of each smart home device.

[0054] The device scoring rules are scored according to the performance of different models of smart home devices. The better the performance and the more powerful the function of the device, the higher the score. Due to the large number of smart home devices, in order to distinguish the primary and secondary and reduce unnecessary data collection and calculation, only the main smart home devices with a greater impact on the degree of intelligence are formulated with detailed additional score scoring standards. The main smart home devices include mobile phones, tablets, TVs, speakers and cameras. For other smart home devices, the additional score is set to 0.

[0055] Among them, the specific device scoring rules for mobile phones are as follows:

[0056]

[0057] The specific device scoring rules for tablets and TVs are as follows:

[0058]

[0059] The specific equipment scoring rules for speakers are as follows:

[0060] The specific device scoring rules for cameras are as follows:

[0061]

[0062] For example, if a Huawei Mate60 phone is identified among the human-machine devices, it can be determined from the device attribute table that the phone model of Mate60 is 5G, the Wi-Fi protocol supports Wi-Fi6, and it will be launched in 2023 with a launch price of 5,999 yuan. Therefore, its additional score is 40 points, and the score of the phone is 100. In order to facilitate data statistical analysis, the attribute data of common smart home devices can be searched online in advance and stored as the device attribute table, and the device attribute table can be pre-stored in a memory or cloud, etc. Subsequently, the corresponding attribute data can be directly queried from the device attribute table according to the device model of the acquired smart home device and the additional score can be determined.

[0063] Therefore, the human-machine device score and the IoT device score are determined according to the following formula:

[0064]

[0065] Among them, score is the score of human-machine equipment or IoT equipment, A is the basic score of human-machine equipment or IoT equipment, and x is the basic score of human-machine equipment or IoT equipment. i is the additional score of the i-th human-machine device or IoT device among all human-machine devices or IoT devices in a household; n is the number of human-machine devices or IoT devices.

[0066] That is, when a family has multiple human-machine devices and multiple IoT devices, the average of the scores of the multiple human-machine devices is taken as the human-machine device score, and the average of the scores of the multiple IoT devices is taken as the IoT device score.

[0067] For some merchant households, considering that there will be a large number of non-resident human-machine devices and IoT devices, which will lead to a higher intelligence score, therefore, when it is determined that the number of human-machine devices and the number of IoT devices of a household are greater than 100, the values ​​of the number of human-machine devices and the number of IoT devices are set to 100 respectively. In addition, when calculating the human-machine device score and the IoT device score, 100 can be randomly selected from all human-machine devices and IoT devices for calculation. More preferably, in order to avoid the high score of non-resident human-machine devices leading to an inflated human-machine device score, at least 100 human-machine devices can be selected to calculate the human-machine device score based on the frequency of human-machine devices using common smart home operation apps.

[0068] The S30 includes the following steps:

[0069] S310, sorting the households in descending order according to the sum of the number of device categories and the number of common smart home operation apps used, the sum of the number of human-machine devices and the number of IoT devices, the sum of the human-machine device scores and the IoT device scores, and the number of other apps used;

[0070] S320, assigning an intelligence score to each family according to the ranking, and the intelligence score assigned to each family decreases in sequence according to the ranking. When assigning points specifically, for example, the points can be assigned in a proportional manner from 100 points to 1 point according to the number of families.

[0071] For example, there are three families a, b, and c whose intelligence rating parameter values ​​are as follows:

[0072] Family A:

[0073]

[0074] Family B:

[0075]

[0076] Family c:

[0077]

[0078] First, sort by the sum of the number of device categories and the number of common smart home operation apps used. The sum of the number of device categories and the number of common smart home operation apps used by family a (9) is greater than the sum of the number of device categories and the number of common smart home operation apps used by family b (7) and the sum of the number of device categories and the number of common smart home operation apps used by family c (7), so family a is ranked before family b and family c.

[0079] Since the sum of the number of device categories and the number of common smart home operation apps used by families B and C is equal, they are then sorted according to the sum of the number of human-machine devices and the number of IoT devices. Since the sum of the number of human-machine devices and the number of IoT devices of family B (11) is greater than the sum of the number of human-machine devices and the number of IoT devices of family C (9), family B is ranked after family A and before family C. Correspondingly, when assigning a smartness score to each family, the smartness score of family A is higher than that of family B, and the smartness score of family B is higher than that of family C.

[0080] Here, we consider that the number of device categories and the number of common smart home operation apps have the greatest impact on the intelligence rating and should have the highest weight. This process not only takes into account the richness of smart home devices, but also the activity and diversity of app applications, ensuring the comprehensiveness and fairness of the rating system.

[0081] Of course, in the S30, after obtaining the values ​​of the intelligence scoring parameters of each family, in order to eliminate the influence of different dimensions and magnitudes on the model, the values ​​of the intelligence scoring parameters of all families are standardized and sorted and scored according to the data after standard processing. Standardization is conducive to mapping the values ​​of all intelligence scoring parameters to a unified scale, ensuring the comparability between the intelligence scoring parameters, and is conducive to the stability and convergence speed of the model. The standardization can, for example, unify all values ​​to within 10, within 5, etc. The specific method is a known technology and is not limited here. Of course, standardization is not necessary and can be omitted.

[0082] In S40, the scoring model is a linear model. Specifically, the linear model is constructed by taking the intelligence score as an independent variable and taking the set of intelligence score parameters as independent variables, as follows:

[0083] S=a*X1+b*X2+c*X3+d*X4+e*X5+f*X6+g*X7;

[0084] Among them, S is the intelligence level score, X1 is the number of device categories, X2 is the number of common smart home operation apps used, X3 is the number of human-machine devices, X4 is the number of IoT devices, X5 is the human-machine device score, X6 is the IoT device score, X7 is the number of other apps used, a, b, c, d, e, f, g are the weight coefficients corresponding to each intelligence level score parameter, which represent the direct impact of each intelligence level score parameter on the intelligence level score.

[0085] By constructing a matrix to solve the RSS minimum value, the weight coefficients corresponding to each feature parameter are as follows:

[0086]

[0087]

[0088] The final fit is 0.9522151782527883.

[0089] The final linear model is as follows:

[0090] S=9.7801986*X1+3.9060424*X2+0.1953693*X3+0.2725942*X4+0.2478278*X5+0.1521893*X6+0.0425248*X7.

[0091] Correspondingly, as attached Figure 2 As shown, the method for evaluating the intelligence level of smart home devices comprises the following steps:

[0092] S1, obtain network data of a household to be evaluated;

[0093] S2, determining a set of values ​​of intelligence rating parameters of the household to be evaluated according to the acquired network data and device rating rules;

[0094] S3, determining the intelligence score of the household to be evaluated according to the values ​​of a set of intelligence score parameters of the household to be evaluated and a scoring model; determining the intelligence level of the smart home device according to the intelligence score of the household to be evaluated.

[0095] The specific processes of S1 and S2 refer to S10 and S20 mentioned above, which will not be described in detail here.

[0096] The difference is that in S2, before determining the values ​​of a set of intelligence scoring parameters for the household to be evaluated, it is necessary to first determine whether the acquired network data contains the device models of all downstream devices of the home gateway of the household to be evaluated. The reason for this operation is that when acquiring the network data of the household to be evaluated, there is a situation where the data acquisition plug-in cannot acquire the device models of all downstream devices under the home gateway of the household to be evaluated. If at least one downstream device has not acquired the device model, it is impossible to accurately determine the actual number of device categories, the number of human-machine devices, the number of IoT devices, the human-machine device score and the IoT device score of the household to be evaluated. Therefore, if the intelligence score is still calculated according to the value calculated at this time, it cannot accurately reflect the intelligence level of the household to be evaluated.

[0097] When making a specific judgment, the number of downstream devices can be compared with the number of downstream devices whose device models have been obtained. If the two are consistent, it means that the device models of all downstream devices of the home gateway of the household to be evaluated have been obtained. Otherwise, it means that there are downstream devices of the home gateway of the household to be evaluated whose device models have not been obtained.

[0098] If not, that is, there is a situation where the device model is not obtained in the downstream device of the home gateway of the family to be evaluated, at this time, the above parameters can no longer be calculated, and the above device category number, human-machine device number, IoT device number, human-machine device score and IoT device score can be directly set to zero, and at the same time, the values ​​of the number of common smart home operation apps and the number of other apps used are determined based on network data, and the number of device categories, human-machine devices, IoT devices, human-machine device scores, IoT device scores and the calculated number of common smart home operation apps and other apps used at this time are used as the first data set. Since the number of common smart home operation apps has the largest proportion in the scoring model, the intelligence score calculated by bringing the above common smart home operation apps, other apps and device category numbers and the parameters set to zero into the calculation model can still reflect the degree of intelligence. Therefore, the intelligence score can still be calculated based on the value of the first data set and the above scoring model.

[0099] If yes, that is, the device models of all devices connected to the home gateway of the home to be evaluated are obtained, then the number of device categories, the number of human-machine devices, the number of IoT devices, the number of common smart home operation apps used, the number of other apps used, the human-machine device score and the IoT device score are calculated based on the obtained network data and used as the second data set;

[0100] In S3, two sets of rating standards are set for the smart home device intelligence rating. The reason for setting two sets of rating standards is that when the intelligence rating is performed based on the intelligence rating obtained from the first-level data set, due to the lack of scores corresponding to other intelligence rating parameters, the intelligence rating will be significantly reduced. Therefore, the intelligence rating range corresponding to the intelligence rating of each smart home device needs to be set smaller to truly reflect the actual intelligence level. The intelligence rating calculated based on the second data set has data of all parameters, so the score will be higher. Therefore, the intelligence rating range corresponding to the intelligence rating of each smart home device needs to be set larger.

[0101] Therefore, when the intelligence rating is calculated based on the first data set, the first set of rating standards is used to perform the intelligence rating; otherwise, the second set of rating standards is used to perform the intelligence rating.

[0102] The first set of rating criteria is exemplified by the following examples:

[0103] When the intelligence level score of the household to be evaluated is greater than or equal to 20 points, the intelligence level of the smart home device is excellent;

[0104] When the intelligence level score of the household to be evaluated is greater than or equal to 10 and less than 20, the intelligence level of the smart home device is good;

[0105] When the intelligence level score of the household to be evaluated is less than 10, the intelligence level of the smart home device is poor.

[0106] The second set of rating criteria is exemplified by the following examples:

[0107] When the intelligence level score of the household to be evaluated is greater than or equal to 70, the intelligence level of the smart home device is excellent;

[0108] When the intelligence level score of the household to be evaluated is greater than or equal to 40 and less than 70, the intelligence level of the smart home device is good;

[0109] When the intelligence level score of the household to be evaluated is less than 40, the intelligence level of the smart home device is poor.

[0110] After the intelligence score and intelligence level of the household to be evaluated are obtained, they can be pushed to the user via App application messages or stored in a cloud server for use by users, operators, etc.

[0111] Of course, in other embodiments, the intelligence scores corresponding to different levels in the first set of rating standards and the second set of rating standards can also be designed to other values ​​as needed. For example, in the first set of rating standards, the intelligence score for the excellent level is greater than or equal to 30 points.

[0112] There are many implementation methods of the present invention, and all technical solutions formed by equivalent transformation or equivalent transformation fall within the protection scope of the present invention.

Claims

1. A method for evaluating the intelligence level of smart home devices, characterized in that: The steps include: S1, obtain network data of a household to be evaluated; S2, determining a set of values ​​of intelligence rating parameters of the household to be evaluated according to the acquired network data; S3, determining the intelligence score of the household to be evaluated according to the values ​​of a set of intelligence score parameters of the household to be evaluated and a scoring model, and determining the intelligence level of the smart home device according to the intelligence score, wherein the smart home device intelligence level assessment has two sets of rating standards; The scoring model is determined by the following process: S10, obtaining network data of multiple households; S20, determining a set of intelligence rating parameters corresponding to each household according to the network data and device rating rules of each household; the set of intelligence rating parameters includes the number of device categories, the number of human-machine devices, the number of IoT devices, the number of common smart home operation apps used, the number of other apps used, the human-machine device rating, and the IoT device rating; S30, sorting the families according to the values ​​of the intelligence level scoring parameters of the families and assigning an intelligence level score to each family according to the sorting; S40, determining a relationship between the intelligence score parameter and the intelligence score as the scoring model according to the intelligence score assigned to each family and the value of the corresponding intelligence score parameter; In S1 and S10, the topology information reported by the plug-in is obtained according to the MAC address of the home gateway and the data in the networking device, and the network topology relationship between the home gateway corresponding to the MAC address and its downstream devices is obtained to obtain the network data; In S2, it is determined whether the acquired network data contains the device models of all the devices connected to the home gateway of the home to be evaluated. If not, the values ​​of the number of device categories, the number of human-machine devices, the number of IoT devices, the human-machine device scores and the IoT device scores are set to zero, and the values ​​of the number of common smart home operation apps used and the number of other apps used are calculated as the first data set; if yes, the values ​​of the number of device categories, the number of human-machine devices, the number of IoT devices, the number of common smart home operation apps used, the number of other apps used, the human-machine device scores and the IoT device scores are calculated according to the acquired network data and used as the second data set; In S3, which set of rating standards to select is determined according to the data based on which the intelligence level score is calculated. When the intelligence level score is calculated based on the first data set, the first set of rating standards is used; otherwise, the second set of rating standards is used.

2. The method for evaluating the intelligence level of smart home devices according to claim 1, characterized in that: The human-machine device score and IoT device score are determined according to the following formula: Among them, score is the score of human-machine equipment or IoT equipment, A is the basic score of human-machine equipment or IoT equipment, and x is the basic score of human-machine equipment or IoT equipment. i is the additional score of the i-th human-machine device or IoT device among all human-machine devices or IoT devices in a household, and n is the number of human-machine devices or IoT devices.

3. A method for evaluating the intelligence level of smart home devices according to claim 2, characterized in that: The additional points are determined according to the device attribute table and the device scoring rules.

4. The method for evaluating the intelligence level of smart home devices according to claim 2, characterized in that: In S2, if it is determined that the values ​​of the number of human-machine devices and the number of IoT devices are respectively greater than 100, the values ​​of the number of human-machine devices and the number of IoT devices are respectively set to 100.

5. A method for evaluating the intelligence level of smart home devices according to claim 1, characterized in that: S30 includes the following steps: S310, sorting the households in descending order according to the sum of the number of device categories and the number of common smart home operation apps used, the sum of the number of human-machine devices and the number of IoT devices, the sum of the human-machine device scores and the IoT device scores, and the number of other apps used; S320, assigning an intelligence score to each household according to the ranking, and the intelligence score assigned to each household decreases in sequence according to the ranking.

6. A method for evaluating the intelligence level of smart home devices according to claim 5, characterized in that: Before S310, the values ​​of the intelligence level scoring parameters of all the households obtained are standardized.

7. The method for evaluating the intelligence level of smart home devices according to claim 1, characterized in that: The scoring model is a linear model.

8. A method for evaluating the intelligence level of smart home devices according to any one of claims 1 to 7, characterized in that: The scoring model is as follows: S=a*X1+b*X2+c*X3+d*X4+e*X5+f*X6+g*X7; Among them, S is the intelligence level score, X1 is the number of device categories, X2 is the number of common smart home operation apps used, X3 is the number of human-machine devices, X4 is the number of IoT devices, X5 is the human-machine device score, X6 is the IoT device score, X7 is the number of other apps used, and a, b, c, d, e, f, and g are the weight coefficients corresponding to each intelligence level score parameter.

9. A method for evaluating the intelligence level of smart home devices according to claim 8, characterized in that: The scoring model is as follows: S=9.7801986*X1+3.9060424*X2+0.1953693*X3+0.2725942*X4+0.2478278*X5+0.1521893*X6+0.0425248*X7.

Images