WO2023020229A1 - 洗护用户画像建立的方法、装置、存储介质和程序产品 - Google Patents
洗护用户画像建立的方法、装置、存储介质和程序产品 Download PDFInfo
- Publication number
- WO2023020229A1 WO2023020229A1 PCT/CN2022/108347 CN2022108347W WO2023020229A1 WO 2023020229 A1 WO2023020229 A1 WO 2023020229A1 CN 2022108347 W CN2022108347 W CN 2022108347W WO 2023020229 A1 WO2023020229 A1 WO 2023020229A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- user
- information
- washing
- portrait
- washing machine
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000005406 washing Methods 0.000 claims abstract description 165
- 238000012549 training Methods 0.000 claims description 47
- 230000008569 process Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000011664 signaling Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 210000002268 wool Anatomy 0.000 description 1
Images
Classifications
-
- D—TEXTILES; PAPER
- D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
- D06F—LAUNDERING, DRYING, IRONING, PRESSING OR FOLDING TEXTILE ARTICLES
- D06F2103/00—Parameters monitored or detected for the control of domestic laundry washing machines, washer-dryers or laundry dryers
- D06F2103/68—Operation mode; Program phase
-
- D—TEXTILES; PAPER
- D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
- D06F—LAUNDERING, DRYING, IRONING, PRESSING OR FOLDING TEXTILE ARTICLES
- D06F33/00—Control of operations performed in washing machines or washer-dryers
- D06F33/30—Control of washing machines characterised by the purpose or target of the control
-
- D—TEXTILES; PAPER
- D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
- D06F—LAUNDERING, DRYING, IRONING, PRESSING OR FOLDING TEXTILE ARTICLES
- D06F33/00—Control of operations performed in washing machines or washer-dryers
- D06F33/30—Control of washing machines characterised by the purpose or target of the control
- D06F33/32—Control of operational steps, e.g. optimisation or improvement of operational steps depending on the condition of the laundry
- D06F33/36—Control of operational steps, e.g. optimisation or improvement of operational steps depending on the condition of the laundry of washing
Definitions
- This application relates to the technology of smart home appliances, and in particular to a method, device, storage medium and program product for creating care user portraits, which belong to the technical field of smart home appliances.
- smart home appliances are also popularized in daily life, and the smart home appliances are usually provided with a fixed operating program to control the operation of the smart home appliances.
- Smart washing machines are commonly used home appliances. Smart washing machines build user portraits and recommend washing programs to users based on the user portraits during the process of using the washing machine, bringing users a better laundry experience.
- user portraits are generally established based on user voiceprint information, user information (such as age, gender, etc.) and their washing habits. This requires a large amount of user information data and user washing information to create accurate user portraits. When washing information If there is little or incomplete user information, it is difficult to accurately construct user portraits for users.
- the present application provides a method, device, storage medium and program product for creating a cleaning user portrait to solve the problem that it is difficult to accurately build a user portrait due to little cleaning information or incomplete user information.
- the first aspect of this application is to provide a method for establishing a care user portrait, including:
- the washing information including washing time and washing program
- the user portrait model is trained according to a large amount of training data, and the training The data includes user information, user voiceprint information, user washing information and portrait tags, the user information includes user account information, user gender and user age, and the portrait tags are used to describe user portraits;
- the user portrait model is obtained by training a preset model with a large amount of training data.
- the user information included in the training data and the user voiceprint information have different weights during the training process, wherein the weight of the user information is greater than the weight of the user voiceprint information.
- the weight of the user information is 80%, and the weight of the user voiceprint information is 20%.
- the portrait tags include: a lady tag, a man tag and an old man tag.
- the second aspect of the present application is to provide a device for creating care user portraits, including:
- An acquisition module configured to acquire the washing information of the first user using the washing machine and the account information used by the first user, the washing information including washing time and washing program;
- a processing module configured to input the first user's washing information and used account information into a pre-trained user portrait model to obtain a user portrait of the first user, and the user portrait model is obtained through training based on a large amount of training data Yes, the training data includes user information, user voiceprint information, user cleaning information and portrait tags, the user information includes user account information, user gender and user age, and the portrait tags are used to describe user portraits;
- a saving module configured to save the user portrait of the first user.
- An acquisition module configured to acquire a large amount of the training data
- a training module configured to train a preset model with a large amount of training data to obtain the user portrait model.
- the third aspect of the present application is to provide a device for creating a care user portrait, including: at least one processor, memory, and transceiver;
- the processor controls receive actions and transmit actions of the transceiver.
- the memory stores computer-executable instructions
- the at least one processor executes the computer-implemented instructions stored in the memory, so that the at least one processor performs the method described in the first aspect.
- a fourth aspect of the present application is to provide a computer-readable storage medium, wherein computer-executable instructions are stored in the computer-readable storage medium, and the computer-executable instructions implement the method described in the first aspect when executed by a processor.
- a fifth aspect of the present application is to provide a computer program product, including a computer program.
- the computer program is executed by a processor, the method described in the first aspect is implemented.
- the method, device, storage medium, and program product provided in this application for establishing laundry user portraits When a user who has not added a family account controls the washing machine, the washing machine or the washing machine control device can be based on the acquired user's washing information and the user's Account information, input the user portrait model predicted by the pre-trained user portrait model of the server, when the user uses the washing machine again, the server queries the user portrait of the user according to the user account information, and based on the user portrait through the washing machine control device for the user Recommend the necessary washing programs to bring users a better washing experience.
- the server receives a large amount of user information, user voiceprint information, and user washing information sent by the washing machine or the control device of the washing machine, makes a data set, and trains the user portrait model based on these information, and trains the user portrait model based on these information , these information acquisition costs are low, and the authenticity is high
- FIG. 1 is a schematic diagram of an application scenario applicable to the present application
- FIG. 2 is a schematic diagram of adding a page to a family account
- FIG. 3 is a schematic flowchart of a method for establishing a care user portrait provided in Embodiment 1 of the present application;
- FIG. 4 is a signaling flow chart of a method for establishing a care user portrait provided in Embodiment 2 of the present application;
- FIG. 5 is a schematic structural diagram of a device for establishing a care user portrait provided in Embodiment 3 of the present application;
- FIG. 6 is a schematic structural diagram of a device for establishing a care user portrait provided in Embodiment 4 of the present application.
- Smart washing machines are commonly used household appliances.
- the user portraits can be understood as abstracting each specific information of the user into labels, and using these labels to concretize the user image, thereby providing users with useful information. targeted services.
- the user can recommend the required washing program for the user according to the user portrait, so as to bring the user a better washing experience.
- user portraits are generally established based on user voiceprint information, user information (such as age, gender, etc.) and their washing habits. This requires a large amount of user information data and user washing information to create accurate user portraits. When washing information If there is little or incomplete user information, it is difficult to accurately construct user portraits for users.
- the present application provides a method, device, storage medium and program product for creating a portrait of a washing machine user, by obtaining the information (such as age, gender, etc.) Get the user's voiceprint information.
- the user information and user voiceprint information of a large number of family accounts obtained are used to make training data sets according to different weights. Among them, the weight of user information is greater than that of user voiceprint information, and different portrait tags are set.
- the portrait labels are used to describe user portraits .
- the training data set is used for the training of the neural network. Using the trained model, according to the user's washing information and the account information used, the portrait of the user who does not join the family but operates the washing machine is predicted.
- the user's portrait can also be obtained by obtaining the user's washing habits, and the server can identify the washing machine corresponding to the family account or The washing machine's control device recommends a more suitable washing program.
- FIG. 1 is a schematic diagram of an application scenario applicable to this application.
- the control device 101 of the washing machine, the washing machine 102 and the server 103 perform information interaction through the Internet.
- the processor in the server 103 is configured to execute corresponding computer programs to implement corresponding control operations on various functions of the washing machine 102 and the washing machine control device 101 .
- the server 103 may also be a server cluster, which is not limited in this embodiment. It can be understood that there may be multiple washing machine control devices 101 , washing machines 102 and servers 103 , which are not shown in the figure.
- the embodiment of the present application does not limit the type of the washing machine 102.
- the washing machine 102 can be various types of washing machines such as a drum washing machine and an agitator washing machine, and the washing machine may or may not have a voice recognition function.
- the embodiment of the present application does not limit the type of the control device 101 of the washing machine.
- the control device 101 of the washing machine can be an electronic device with a washing machine APP installed on a mobile phone, a tablet computer, a desktop computer, etc., and realize the washing program through the APP installed on the control device 101 of the washing machine display and selection.
- FIG. 2 is a schematic diagram of a family account adding page. As shown in FIG. 2 , account information 201 and an account adding area 202 are displayed in the adding page.
- the user can click the account information viewing control in the added page displayed by the control device 20 of the washing machine to view the account information of the family. account or a user-defined account.
- the "My Family” control is displayed in the account addition area 202, and the user can view the information of the family members added to the account by clicking on the "My Family" control.
- an adding control 203 is also displayed in the account adding area 202, and the adding control 203 is used to add a new family member. After the user clicks the add control 203, information such as gender and age of the newly added member can be further added.
- multiple family members can be added under one family account, the user information of the multiple family members is different, and the multiple family members share a family account.
- FIG. 3 is a schematic flowchart of a method for establishing a care user portrait provided in Embodiment 1 of the present application.
- the method can be executed by a server, a washing machine control device, or a washing machine.
- the execution subject can be
- the server is taken as an example for description, and the method includes the following steps.
- Step 301 acquiring washing information of a first user who uses a washing machine and account information used by the first user, where the washing information includes washing time and washing programs.
- the first user is a user who has not added a family account, and when the first user selects a washing program through the washing machine, the washing machine can obtain the washing information of the first user.
- the washing machine can send the washing information to the control device or server of the washing machine, and the control device or server of the washing machine can obtain the account information used by the first user according to the identity of the washing machine.
- the control device of the washing machine acquires the washing information of the first user.
- the control device of the washing machine may send the washing information of the first user and the account number used by the first user to the server.
- the washing information obtained by the washing machine or the control device of the washing machine is: the washing program is a large-scale washing program, and the washing time is 16:00 in the afternoon.
- Step 302 Input the first user's washing information and used account information into the pre-trained user portrait model to obtain the first user's user portrait.
- the above-mentioned user portrait model is obtained by the server based on a large amount of training data collected, and the training data includes user information, user voiceprint information, user washing information and portrait tags.
- User information includes user information including user account information, user gender and user age.
- User voiceprint information is used for users to set portrait tags based on voiceprint characteristics. For example, information with dialect accents can be set as the "elderly" label.
- the weight of the user information is greater than the weight of the user voiceprint information, for example, the weight of the user information is 80%, and the weight of the user voiceprint information is 20%, or the weight of the user information is 85%. %, the weight of the user's voiceprint information is 15%.
- the weight of user information and the weight of user voiceprint information may also be the same.
- the server is a process of continuous optimization, which can continuously adjust the weight ratio of user information weight and user voiceprint information according to the training results.
- the portrait tags may include: a "lady” tag, a "man” tag, and an “elderly” tag.
- the portrait tag can also be set as a "single nobleman” tag, a “full-time mother” tag, a “fitness person” tag, and the like.
- the information with the simplification of washing procedures and the majority of voiceprint recognition features in dialects is set as the "elderly" label.
- the description of the setting of the image tag and the setting manner of the image tag in this embodiment is only an example, and does not constitute a limitation.
- the variety of washing information may make the setting method of the portrait label different.
- the user portrait prediction can be performed by the server, the control device of the washing machine, or the washing machine.
- the user portrait model is trained by the server, and the server receives a large amount of user information and user information sent by the washing machine or the control device of the washing machine.
- the voiceprint information and the user's cleaning information are made into a data set, and the user portrait model is trained based on these information.
- the server obtains the user portrait model through training, it can predict the user portrait based on the user portrait model.
- the training and use of the user portrait model can be realized by two different servers, and the user portrait model can be trained by a dedicated server, and the user portrait model is sent to another dedicated server for user portrait prediction model after the training is completed.
- the server can send the trained user portrait model to the washing machine or the washing machine's control device.
- the washing machine or the control device of the washing machine can be input into the user portrait model based on the obtained washing information of the user and the account information used by the user to obtain the user portrait of the user.
- Step 303 saving the user portrait of the first user.
- the user portrait of the first user is associated with the account information used by the first user. Later, when the first user uses the washing machine again, the first user can be queried based on the account information used by the first user based on the user portrait, and recommend the desired washing program to the user through the control device of the washing machine based on the user portrait.
- the washing machine or the control device of the washing machine can input the user portrait model pre-trained by the server based on the obtained user's washing information and user account information to obtain the user profile.
- Portrait when the user uses the washing machine again, the server queries the user portrait of the user according to the user account information, and based on the user portrait, recommends the required washing program to the user through the washing machine control device, bringing the user a better laundry experience.
- the server receives a large amount of user information, user voiceprint information, and user washing information sent by the washing machine or the control device of the washing machine, makes a data set, and trains the user portrait model based on these information, and trains the user portrait model based on these information , these information acquisition costs are low, and the authenticity is high.
- Fig. 4 is a signaling flowchart of a method for establishing a care user portrait provided in Embodiment 2 of the present application.
- This embodiment describes the interaction process between devices in detail on the basis of Embodiment 1, and
- the server performs user portrait model training and user portrait prediction as an example for illustration.
- the method provided in this embodiment includes the following steps.
- Step 401 the control device of the washing machine records user information, user voiceprint information and user washing information.
- Step 402 the washing machine records user information, user voiceprint information and user washing information.
- step 401 and step 402 are executed in no sequence.
- Step 403 the control device of the washing machine sends the user information, user voiceprint information and user washing information to the server.
- Step 404 the washing machine sends user information, user voiceprint information and user washing information to the server.
- step 403 and step 404 are executed in no sequence.
- Step 405 the server creates a training data set according to different weights of user information and user voiceprint information.
- Step 406 the server uses the training data set to train the user portrait model.
- Step 407 the control device of the washing machine records the washing information used by the first user and the account information used by the user.
- Step 408 the control device of the washing machine sends the washing information used by the first user and the account information used by the user to the server.
- Step 409 the server inputs the washing information used by the first user and the account information used by the user into the user portrait model to obtain the user image of the first user.
- Step 410 the server saves the user portrait of the first user.
- FIG. 5 is a schematic structural diagram of a device for establishing a care user portrait provided in Embodiment 3 of the present application.
- the device 50 includes: an acquiring module 501 , a processing module 502 and a saving module 503 .
- the obtaining module 501 is configured to obtain the washing information of the first user using the washing machine and the account information used by the first user, where the washing information includes washing time and washing program.
- the processing module 502 is configured to input the washing information of the first user and the account information used into the pre-trained user portrait model to obtain the user portrait of the first user.
- the user portrait model is trained according to a large amount of training data, and the training data includes User information, user voiceprint information, user cleaning information and portrait tags, user information includes user account information, user gender and user age, and portrait tags are used to describe user portraits.
- obtain a large amount of training data and use the large amount of training data to train the preset model to obtain a user portrait model.
- the user information included in the training data and the user voiceprint information have different weights during the training process, wherein the weight of the user information is greater than the weight of the user voiceprint information.
- the weight of user information is 80%, and the weight of user voiceprint information is 20%.
- the portrait tags include: a lady tag, a man tag and an old man tag.
- the saving module 503 is configured to save the user portrait of the first user.
- the device can be applied to the washing machine or the control equipment of the washing machine, and can also be applied to the server.
- the device in this embodiment can be used to implement the method for creating a care user portrait in Embodiment 1.
- the specific implementation and technical effect are similar, and will not be repeated here.
- Fig. 6 is a schematic structural diagram of a device for establishing a care user portrait provided in Embodiment 4 of the present application.
- the device 60 may be a washing machine control device, or a server.
- the device includes: a processor 601, a memory 602, and a transceiver 603.
- the processor 601 executes the computer-executed instructions stored in the memory 602, and controls the transceiver 603 to receive
- the actions and sending actions enable at least one processor to execute the method steps performed by the control device or the server of the washing machine in Embodiment 1 or Embodiment 2.
- the specific implementation methods and technical effects are similar and will not be repeated here.
- Embodiment 5 of the present application provides a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium.
- the computer-executable instructions are executed by a processor, they are used to implement any one of the above-mentioned Embodiment 1 or Embodiment 2.
- the steps of the method for establishing a care user portrait of the item, the specific implementation method and the technical effect are similar, and will not be repeated here.
- Embodiment 6 of the present application provides a computer program product, including a computer program.
- the computer program When executed by a processor, it implements the steps of the method for creating a care user portrait as in any one of Embodiment 1 or Embodiment 2 above. Specifically The implementation method and the technical effect are similar, and will not be repeated here.
Landscapes
- Engineering & Computer Science (AREA)
- Textile Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
洗护用户画像建立的方法、装置、存储介质和程序产品，当有未加入家庭账号的用户操控洗衣机时，洗衣机或者洗衣机的控制设备可以基于获取到的用户的洗涤信息和用户的账号信息，输入服务器预先训练好的用户画像模型预测得到用户画像，当该用户再次使用洗衣机，服务器根据用户的账号信息查询到该用户的用户画像，并基于该用户画像通过洗衣机的控制设备为用户推荐需要的洗涤程序，带给用户更好的洗衣体验。另外，服务器接收到洗衣机或者洗衣机的控制设备发送的大量的用户信息、用户声纹信息、用户的洗涤信息，制作成数据集，并基于这些信息训练用户画像模型，这些信息获取成本低，且真实性高。
Description
本申请要求于2021年08月16日提交中国专利局、申请号为202110938521.1、申请名称为“洗护用户画像建立的方法、装置、存储介质和程序产品”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本申请涉及本申请涉及智能家电技术，尤其涉及一种洗护用户画像建立的方法、装置、存储介质和程序产品，属于智能家电技术领域。
随着科技的发展，智能家电也在日常生活中得到普及，智能家电中通常设置有固定的运行程序以控制智能家电的运行。
智能洗衣机是常用的家电设备，智能洗衣机通过建立用户画像，在用户使用洗衣机的过程中，可以根据该用户画像为用户推荐需要的洗涤程序，带给用户更好的洗衣体验。
目前，一般是基于用户声纹信息、用户的信息(例如年龄、性别等)及其洗涤习惯来建立用户画像，这需要海量的用户信息数据和用户洗涤信息才能建立精准的用户画像，当洗涤信息很少或者用户信息不全，则很难准确的对用户进行用户画像的构建。
发明内容
本申请提供一种洗护用户画像建立的方法、装置、存储介质和程序产品，用以解决由于洗涤信息很少或者用户信息不全难以准确的对用户进行用户画像构建的问题。
本申请的第一方面是提供一种洗护用户画像建立的方法，包括：
获取使用洗衣机的第一用户的洗涤信息和所述第一用户使用的账号信息，所述洗涤信息包括洗涤时间和洗涤程序；
将所述第一用户的洗涤信息和使用的账号信息输入预先训练得到的用户画像模型，得到所述第一用户的用户画像，所述用户画像模型是根据大量训练数据训练得到的，所述训练数据包括用户信息、用户声纹信息、用户的洗涤信息 和画像标签，所述用户信息包括用户账号信息、用户性别和用户年龄，所述画像标签用于描述用户画像；
保存所述第一用户的用户画像。
可选的，还包括：
获取大量所述训练数据；
通过大量所述训练数据对预设模型进行训练得到所述用户画像模型。
可选的，所述训练数据包括的用户信息和用户声纹信息在训练过程中的权重不同，其中，所述用户信息的权重大于所述用户声纹信息的权重。
可选的，所述用户信息的权重为80％，所述用户声纹信息的权重为20％。
可选的，所述画像标签包括：女士标签、男士标签和老人标签。
本申请的第二方面是提供一种洗护用户画像建立的装置，包括：
获取模块，用于获取使用洗衣机的第一用户的洗涤信息和所述第一用户使用的账号信息，所述洗涤信息包括洗涤时间和洗涤程序；
处理模块，用于将所述第一用户的洗涤信息和使用的账号信息输入预先训练得到的用户画像模型，得到所述第一用户的用户画像，所述用户画像模型是根据大量训练数据训练得到的，所述训练数据包括用户信息、用户声纹信息、用户的洗涤信息和画像标签，所述用户信息包括用户账号信息、用户性别和用户年龄，所述画像标签用于描述用户画像；
保存模块，用于保存所述第一用户的用户画像。
可选的，还包括：
获取模块，用于获取大量所述训练数据；
训练模块，用于通过大量所述训练数据对预设模型进行训练得到所述用户画像模型。
本申请的第三方面是提供一种洗护用户画像建立的装置，包括：至少一个处理器、存储器、收发器；
所述处理器控制所述收发器的接收动作和发送动作。
所述存储器存储计算机执行指令；
所述至少一个处理器执行所述存储器存储的计算机执行指令，使得所述至少一个处理器执行第一方面所述的方法。
本申请的第四方面是提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机执行指令，所述计算机执行指令被处理器执行时实现第一方面所述的方法。
本申请的第五方面是提供一种计算机程序产品，包括计算机程序，所述计算机程序被处理器执行时，实现第一方面所述的方法。
本申请提供的洗护用户画像建立的方法、装置、存储介质和程序产品，当有未加入家庭账号的用户操控洗衣机时，洗衣机或者洗衣机的控制设备可以基于获取到的用户的洗涤信息和用户的账号信息，输入服务器预先训练好的用户画像模型预测得到用户画像，当该用户再次使用洗衣机，服务器根据用户的账号信息查询到该用户的用户画像，并基于该用户画像通过洗衣机的控制设备为用户推荐需要的洗涤程序，带给用户更好的洗衣体验。另外，服务器接收到洗衣机或者洗衣机的控制设备发送的大量的用户信息、用户声纹信息、用户的洗涤信息，制作成数据集，并基于这些信息训练用户画像模型，并基于这些信息训练用户画像模型，这些信息获取成本低，且真实性高
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并与说明书一起用于解释本公开的原理。
图1为本申请适用的一种应用场景的示意图；
图2为家庭账户添加页面的一种示意图；
图3为本申请实施例一提供的一种洗护用户画像建立的方法的流程示意图；
图4为本申请实施例二提供的一种洗护用户画像建立的方法的信令流程图；
图5为本申请实施例三提供的一种洗护用户画像建立的装置的结构示意图；
图6为本申请实施例四提供的一种洗护用户画像建立的装置的结构示意图。
通过上述附图，已示出本公开明确的实施例，后文中将有更详细的描述。这些附图和文字描述并不是为了通过任何方式限制本公开构思的范围，而是通过参考特定实施例为本领域技术人员说明本公开的概念。
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本申请相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本申请的一些方面相一致的装置和方法的例子。
智能洗衣机是常用的家电设备，通过为使用智能洗衣机的用户建立用户画像，该用户画像可以理解为将用户的每个具体信息抽象成标签，利用这些标签将用户形象具体化，从而为用户提供有针对性的服务。在用户使用洗衣机的过程中，可以根据该用户画像为用户推荐需要的洗涤程序，带给用户更好的洗衣体验。目前，一般是基于用户声纹信息、用户的信息(例如年龄、性别等)及其洗涤习惯来建立用户画像，这需要海量的用户信息数据和用户洗涤信息才能建立精准的用户画像，当洗涤信息很少或者用户信息不全，则很难准确的对用户进行用户画像的构建。
针对上述问题，本申请提供一种洗护用户画像建立的方法、装置、存储介质和程序产品，通过获取加入家庭账号的用户的信息(例如年龄、性别等)，并通过洗衣机或者洗衣机的控制设备获取用户的声纹信息。将获取到的大量的家庭账号的用户信息和用户声纹信息按照不同权重制作训练数据集，其中，用户信息的权重大于用户声纹信息，设置不同的画像标签，该画像标签用于描述用户画像。该训练数据集用于神经网络的训练，利用训练得到的模型，根据用户的洗涤信息和使用的账号信息预测得到未加入家庭但操控洗衣机的用户画像。在没有加入家庭账号的用户操控洗衣机时并且没有获取到该用户的声纹信息的情况下，也能通过获取用户的洗涤习惯，得到用户画像，服务器根据该用户画像为该家庭账号对应的洗衣机或洗衣机的控制设备推荐更适合的洗涤程序。
图1为本申请适用的一种应用场景的示意图。如图1所示，洗衣机的控制设备101、洗衣机102和服务器103之间通过互联网进行信息交互。服务器103中的处理器被配置可以执行相应的计算机程序以实现对洗衣机102和洗衣机的控制设备101的各种功能的相应的控制操作。服务器103也可以为服务器集群，本实施例对此不作限定。可以理解，洗衣机的控制设备101、洗衣机102和服务器103的数量均可以为多个，图中未示出。
另外，本申请实施例对于洗衣机102的类型不作限定，洗衣机102可以为滚筒洗衣机、搅拌式洗衣机等各种类型的洗衣机，且该洗衣机可以有语音识别功能，也可以没有语音识别功能。本申请实施例对于洗衣机的控制设备101的类型不作限定，洗衣机的控制设备101可以为手机、平板电脑、台式电脑等安装洗衣机APP的电子设备，通过洗衣机的控制设备101上安装的APP实现洗涤程序的展示和选择。
图2为家庭账户添加页面的一种示意图，如图2所示，该添加页面内显示有账户信息201和账户添加区域202。
用户可以在洗衣机的控制设备20显示的添加页面内点击账户信息查看控件，查看该家庭的账户信息，该账户信息可以包括账号、家庭名称等，账号可以是手机号、邮箱号、即时通讯软件的账号或者用户自定义的账号。
账户添加区域202内显示有“我的家人”控件，用户通过对“我的家人”控件进行点击操作查看该账户已经添加的家庭成员的信息。同时，在账户添加区域202内还显示有添加控件203，添加控件203用于添加新的家庭成员。用户点击添加控件203之后，可以进一步添加新增成员的性别、年龄等信息。
需要说明的是，一个家庭账号下可以添加多个家庭成员，该多个家庭成员的用户信息不同，该多个家庭成员共同使用一个家庭账号。
下面以具体地实施例对本申请的技术方案以及本申请的技术方案如何解决上述技术问题进行详细说明。下面这几个具体的实施例可以相互结合，对于相同或相似的概念或过程可能在某些实施例中不再赘述。下面将结合附图，对本申请的实施例进行描述。
参考图3，图3为本申请实施例一提供的一种洗护用户画像建立的方法的流程示意图，该方法可以由服务器、洗衣机的控制设备或者由洗衣机执行，本实施例可以以执行主体为服务器为例进行说明，该方法包括如下步骤。
步骤301、获取使用洗衣机的第一用户的洗涤信息和第一用户使用的账号信息，洗涤信息包括洗涤时间和洗涤程序。
该第一用户为未加入家庭账号的用户，当第一用户通过洗衣机选择洗涤程序后，洗衣机能够获取到第一用户的洗涤信息。可选的，洗衣机可以将该洗涤信息发送给洗衣机的控制设备或者服务器，洗衣机的控制设备或者服务器根据洗衣机的标识能够获取到第一用户使用的账号信息。
或者，用户通过洗衣机的控制设备选择洗涤程序后，洗衣机的控制设备获取到第一用户的洗涤信息。可选的，洗衣机的控制设备可以将第一用户的洗涤信息和第一用户使用的账号发送给服务器。
例如，当洗衣机被未加入家庭的第一用户操控时，洗衣机或者洗衣机的控制设备获取到的洗涤信息是：洗涤程序为大物洗程序，洗涤时间为下午16:00。
步骤302、将第一用户的洗涤信息和使用的账号信息输入预先训练得到的用户画像模型，得到第一用户的用户画像。
上述用户画像模型是服务器根据采集到的大量的训练数据得到，该训练数据包括用户信息、用户声纹信息、用户的洗涤信息和画像标签。用户信息包括用户信息包括用户账号信息、用户性别和用户年龄。用户声纹信息用于用户根 据声纹特征设定画像标签，例如，有方言口音的信息可设定为“老人”标签。
可选的，在训练过程中，用户信息的权重大于用户声纹信息的权重，例如，该用户信息权重为80％，该用户声纹信息的权重为20％，或者，该用户信息权重为85％，该用户声纹信息的权重为15％。当然，用户信息的权重和用户声纹信息的权重也可以相同。
本实施例对训练数据集中用户信息权重和用户声纹信息权重的说明只是举例，并不构成限定。服务器在训练用户画像模型过程中，是一个不断优化的过程，可以根据训练结果不断调整用户信息权重和用户声纹信息的权重比值。
画像标签可以包括：“女士”标签、“男士”标签和“老人”标签。
作为一种可能的实现方式，画像标签还可以设定为“单身贵族”标签、“全职妈妈”标签、“健身人士”标签等。
作为一种可能的实现方式，对于画像标签的设定有如下方式：
具有洗涤程序多样化(例如，宝宝洗、羊毛洗大物洗等等)洗涤行为特征的信息，设定标签为“女士”。
具有洗涤程序单一化(例如，只使用一种洗涤程序)，并且只有在周末时间才洗衣的洗涤行为特征的信息，设定标签为“男士”。
具有早晨洗衣洗涤习惯特征且声纹识别是方言居多特征的信息，设定标签为“老人”。
画像标签的设定还可以有如下方式：
具有频繁的在晚上时间洗衣的洗涤行为特征的信息，可设定为“女士”标签。
具有洗涤程序的选择大多为针对特殊材质衣物的(例如为牛仔、衬衣等)洗涤行为特征的信息，设定为“男士”标签。
具有洗涤程序单一化且声纹识别是方言居多特征的信息，设定为“老人”标签。
本实施例对画像标签的设定以及画像标签的设定方式的说明只是举例，并不构成限定。洗涤信息的多样化，可能会使画像标签的设定方式不同。
本实施例中，用户画像预测可以由服务器、洗衣机的控制设备或者洗衣机执行，通常情况下，该用户画像模型由服务器训练得到，服务器接收到洗衣机或者洗衣机的控制设备发送的大量的用户信息、用户声纹信息、用户的洗涤信息，制作成数据集，并基于这些信息训练用户画像模型。服务器训练得到用户画像模型之后，可以基于该用户画像模型进行用户画像的预测。
可以理解，用户画像模型的训练和使用可以由两个不同的服务器实现，可以由一个专用服务器训练用户画像模型，在训练完成后将用户画像模型发送给另一个专用服务器进行用户画像预测的模型。
当用户画像预测可以由洗衣机的控制设备或者洗衣机执行时，服务器可以将该训练好的用户画像模型发送到洗衣机或者洗衣机的控制设备上。当有未加入家庭账号的用户操控洗衣机时，使得洗衣机或者洗衣机的控制设备可以基于获取到的用户的洗涤信息和用户使用的账号信息输入该用户画像模型中，得到该用户的用户画像。
步骤303、保存第一用户的用户画像。
保存第一用户的用户画像时将该第一用户的用户画像与第一用户使用的账号信息关联，后续当第一用户再次使用洗衣机，可以根据第一用户使用的账号信息，查询到第一用户的用户画像，并基于该用户画像通过该洗衣机的控制设备为用户推荐需要的洗涤程序。
本实施例中，当有未加入家庭账号的用户操控洗衣机时，洗衣机或者洗衣机的控制设备可以基于获取到的用户的洗涤信息和用户的账号信息，输入服务器预先训练好的用户画像模型预测得到用户画像，当该用户再次使用洗衣机，服务器根据用户的账号信息查询到该用户的用户画像，并基于该用户画像通过洗衣机的控制设备为用户推荐需要的洗涤程序，带给用户更好的洗衣体验。另外，服务器接收到洗衣机或者洗衣机的控制设备发送的大量的用户信息、用户声纹信息、用户的洗涤信息，制作成数据集，并基于这些信息训练用户画像模型，并基于这些信息训练用户画像模型，这些信息获取成本低，且真实性高。
参考图4，图4为本申请实施例二提供的一种洗护用户画像建立的方法的信令流程图，本实施例在实施例一的基础上，详细描述设备之间的交互过程，且本实施例以服务器进行用户画像模型训练以及用户画像预测为例进行说明，如图4所示，本实施例提供的方法包括如下步骤。
步骤401、洗衣机的控制设备记录用户信息、用户声纹信息和用户洗涤信息。
步骤402、洗衣机记录用户信息、用户声纹信息和用户洗涤信息。
可以理解，步骤401和步骤402在执行时并没有先后顺序。
步骤403、洗衣机的控制设备发送用户信息、用户声纹信息和用户洗涤信息到服务器。
步骤404、洗衣机发送用户信息、用户声纹信息和用户洗涤信息到服务器。
可以理解，步骤403和步骤404在执行时并没有先后顺序。
步骤405、服务器按照用户信息与用户声纹信息不同权重制作训练数据集。
步骤406、服务器利用训练数据集训练用户画像模型。
步骤407、洗衣机的控制设备记录第一用户使用的洗涤信息和用户使用的账号信息。
步骤408、洗衣机的控制设备发送第一用户使用的洗涤信息和用户使用的账号信息到服务器。
步骤409、服务器将第一用户使用的洗涤信息和用户使用的账号信息输入用户画像模型，得到第一用户的用户图像。
步骤410、服务器保存第一用户的用户画像。
本实施例的流程，可用于执行实施例一的洗护用户画像建立的方法，具体实现方式和技术效果类似，这里不再赘述。
参考图5，图5为本申请实施例三提供的一种洗护用户画像建立的装置的结构示意图。如图5所示，该装置50包括：获取模块501，处理模块502和保存模块503。
获取模块501，用于获取使用洗衣机的第一用户的洗涤信息和第一用户使用的账号信息，洗涤信息包括洗涤时间和洗涤程序。
处理模块502，用于将第一用户的洗涤信息和使用的账号信息输入预先训练得到的用户画像模型，得到第一用户的用户画像，用户画像模型是根据大量训练数据训练得到的，训练数据包括用户信息、用户声纹信息、用户的洗涤信息和画像标签，用户信息包括用户账号信息、用户性别和用户年龄，画像标签用于描述用户画像。
可选的，获取大量训练数据，通过大量训练数据对预设模型进行训练得到用户画像模型。
可选的，训练数据包括的用户信息和用户声纹信息在训练过程中的权重不同，其中，用户信息的权重大于用户声纹信息的权重。
可选的，用户信息的权重为80％，用户声纹信息的权重为20％。
可选的，画像标签包括：女士标签、男士标签和老人标签。
保存模块503，用于保存第一用户的用户画像。
该装置可应用于洗衣机或者洗衣机的控制设备，还可以应用于服务器。
本实施例的装置，可用于执行实施例一中的洗护用户画像建立的方法，具体实现方式和技术效果类似，这里不再赘述。
参考图6，图6为本申请实施例四提供的一种洗护用户画像建立的装置的结 构示意图。该装置60可以是洗衣机的控制设备，也可以是服务器，该装置包括：处理器601、存储器602、收发器603，该处理器601执行存储器602存储的计算机执行指令，并控制收发器603的接收动作和发送动作，使得至少一个处理器执行实施例一或者实施例二中洗衣机的控制设备或者服务器执行的方法步骤，具体实现方式和技术效果类似，这里不再赘述。
本申请实施例五提供一种计算机可读存储介质，该计算机可读存储介质中存储有计算机执行指令，该计算机执行指令被处理器执行时用于实现如上述实施例一或者实施例二任一项的洗护用户画像建立的方法的步骤，具体实现方式和技术效果类似，这里不再赘述。
本申请实施例六提供一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时，实现如上述实施例一或者实施例二任一项的洗护用户画像建立的方法的步骤，具体实现方式和技术效果类似，这里不再赘述。
本领域技术人员在考虑说明书及实践这里公开的发明后，将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本申请的真正范围和精神由下面的权利要求书指出。
应当理解的是，本申请并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本申请的范围仅由所附的权利要求书来限制。
Claims (10)
- 一种洗护用户画像建立的方法，其特征在于，包括：获取使用洗衣机的第一用户的洗涤信息和所述第一用户使用的账号信息，所述洗涤信息包括洗涤时间和洗涤程序；将所述第一用户的洗涤信息和使用的账号信息输入预先训练得到的用户画像模型，得到所述第一用户的用户画像，所述用户画像模型是根据大量训练数据训练得到的，所述训练数据包括用户信息、用户声纹信息、用户的洗涤信息和画像标签，所述用户信息包括用户账号信息、用户性别和用户年龄，所述画像标签用于描述用户画像；保存所述第一用户的用户画像。
- 根据权利要求1所述的方法，其特征在于，还包括：获取大量所述训练数据；通过大量所述训练数据对预设模型进行训练得到所述用户画像模型。
- 根据权利要求2所述的方法，其特征在于，所述训练数据包括的用户信息和用户声纹信息在训练过程中的权重不同，其中，所述用户信息的权重大于所述用户声纹信息的权重。
- 根据权利要求3所述的方法，其特征在于，所述用户信息的权重为80％，所述用户声纹信息的权重为20％。
- 根据权利要求1-4任一项所述的方法，其特征在于，所述画像标签包括：女士标签、男士标签和老人标签。
- 一种洗护用户画像建立的装置，其特征在于，包括：获取模块，用于获取使用洗衣机的第一用户的洗涤信息和所述第一用户使用的账号信息，所述洗涤信息包括洗涤时间和洗涤程序；处理模块，用于将所述第一用户的洗涤信息和使用的账号信息输入预先训练得到的用户画像模型，得到所述第一用户的用户画像，所述用户画像模型是根据大量训练数据训练得到的，所述训练数据包括用户信息、用户声纹信息、用户的洗涤信息和画像标签，所述用户信息包括用户账号信息、用户性别和用户年龄，所述画像标签用于描述用户画像；保存模块，用于保存所述第一用户的用户画像。
- 根据权利要求6所述的装置，其特征在于，还包括：获取模块，用于获取大量所述训练数据；训练模块，用于通过大量所述训练数据对预设模型进行训练得到所述用户画像模型。
- 一种洗护用户画像建立的装置，其特征在于，包括：至少一个处理器、存储器、收发器；所述处理器控制所述收发器的接收动作和发送动作；所述存储器存储计算机执行指令；所述至少一个处理器执行所述存储器存储的计算机执行指令，使得所述至少一个处理器执行如权利要求1至5任一项所述的方法。
- 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有计算机执行指令，所述计算机执行指令被处理器执行时实现如权利要求1至5任一项所述的方法。
- 一种计算机程序产品，包括计算机程序，其特征在于，所述计算机程序被处理器执行时，实现权利要求1至5任一项所述的方法。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110938521.1 | 2021-08-16 | ||
| CN202110938521.1A CN115704153A (zh) | 2021-08-16 | 2021-08-16 | 洗护用户画像建立的方法、装置、存储介质和程序产品 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023020229A1 true WO2023020229A1 (zh) | 2023-02-23 |
Family
ID=85181259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/108347 WO2023020229A1 (zh) | 2021-08-16 | 2022-07-27 | 洗护用户画像建立的方法、装置、存储介质和程序产品 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115704153A (zh) |
| WO (1) | WO2023020229A1 (zh) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106227841A (zh) * | 2016-07-26 | 2016-12-14 | 北京小米移动软件有限公司 | 信息推送方法和装置 |
| CN106319834A (zh) * | 2015-06-17 | 2017-01-11 | 青岛海尔洗衣机有限公司 | 衣物洗涤管理方法 |
| CN109902461A (zh) * | 2019-03-01 | 2019-06-18 | 珠海格力电器股份有限公司 | 一种电器控制方法、装置、存储介质及电器 |
| CN110373859A (zh) * | 2019-07-09 | 2019-10-25 | 合肥美的洗衣机有限公司 | 衣物处理装置的洗涤控制方法、装置和衣物处理装置 |
-
2021
- 2021-08-16 CN CN202110938521.1A patent/CN115704153A/zh active Pending
-
2022
- 2022-07-27 WO PCT/CN2022/108347 patent/WO2023020229A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106319834A (zh) * | 2015-06-17 | 2017-01-11 | 青岛海尔洗衣机有限公司 | 衣物洗涤管理方法 |
| CN106227841A (zh) * | 2016-07-26 | 2016-12-14 | 北京小米移动软件有限公司 | 信息推送方法和装置 |
| CN109902461A (zh) * | 2019-03-01 | 2019-06-18 | 珠海格力电器股份有限公司 | 一种电器控制方法、装置、存储介质及电器 |
| CN110373859A (zh) * | 2019-07-09 | 2019-10-25 | 合肥美的洗衣机有限公司 | 衣物处理装置的洗涤控制方法、装置和衣物处理装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115704153A (zh) | 2023-02-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6653774B2 (ja) | 通知に対するユーザ可用性又は受容性を測定する方法、装置及び機械可読媒体 | |
| US9930102B1 (en) | Method and system for using emotional state data to tailor the user experience of an interactive software system | |
| CN106779088B (zh) | 执行机器学习流程的方法及系统 | |
| US20190303504A1 (en) | Distributed in-browser deep learning for predictive pre-fetching | |
| WO2017080170A1 (zh) | 群体用户画像方法及系统 | |
| US9785534B1 (en) | Method and system for using abandonment indicator data to facilitate progress and prevent abandonment of an interactive software system | |
| US20070300225A1 (en) | Providing user information to introspection | |
| US11100126B2 (en) | Creating reports using dimension trees | |
| US8392229B2 (en) | Activity-centric granular application functionality | |
| CN103827778B (zh) | 企业工具增强 | |
| US20190236466A1 (en) | Dynamic access of artificial intelligence engine in a cloud computing architecture | |
| CN108055320B (zh) | 资源下载的处理方法、终端设备及计算机可读存储介质 | |
| WO2022048515A1 (zh) | 一种测评实现方法、装置和存储介质 | |
| CN111898768A (zh) | 数据处理方法、装置、设备及介质 | |
| CN108549567A (zh) | 动画显示方法、装置、终端、服务器及存储介质 | |
| US11782943B2 (en) | Method and system for case management | |
| CN109284386A (zh) | 可定制的意图识别方法及装置 | |
| JP5973094B1 (ja) | 顧客対応支援システム、顧客対応支援方法、及び顧客対応支援用コンピュータプログラム | |
| CN105404514A (zh) | JavaScript渲染方法和装置 | |
| WO2023020229A1 (zh) | 洗护用户画像建立的方法、装置、存储介质和程序产品 | |
| US20200410049A1 (en) | Personalizing online feed presentation using machine learning | |
| JPWO2018030136A1 (ja) | 情報処理装置、情報処理方法、プログラム、および情報処理システム | |
| CN107368291A (zh) | 一种网页绘制装置、方法及电子设备 | |
| CN111552549A (zh) | 一种智能设备的多任务请求动态配置方法及系统 | |
| US11658926B2 (en) | Generating smart replies involving image files |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22857542 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 22857542 Country of ref document: EP Kind code of ref document: A1 |