CN-122029018-A - Sensitivity classification of processing head users

Abstract

The presently disclosed subject matter relates to a computer-implemented method of classifying sensitivity of a user to a treatment head of an appliance (10). The computer-implemented method includes receiving (S500) data from a sensor of the appliance representative of a physical parameter associated with operating the appliance, assigning (S502) a user to a classification of sensitivity of a processing head of the appliance using a machine learning model based on the received data, and transmitting (S504) a signal indicative of the assigned classification.

Inventors

• S. Adiya
• K. B. Kush
• A. C. Stefan
• R. DE GROOT
• I. Catizia
• A. Korozen
• R.F.J. Van der Hill

Assignees

• 皇家飞利浦有限公司

Dates

Publication Date

20260512

Application Date

20241017

Priority Date

20231020

Claims (15)

1. 1. A computer-implemented method of classifying sensitivity of a user to a processing head (12) of an appliance (10), the computer-implemented method comprising: -receiving (S500) data representing a physical parameter associated with operating the appliance (10) from a sensor (18) of the appliance (10); Assigning (S502) a user to a classification of sensitivity of a treatment head (12) of the appliance (10) using a machine learning model based on the received data, and A signal indicating the assigned classification is transmitted (S504).
2. 2. The computer-implemented method of claim 1, wherein the physical parameter comprises current and/or power for driving a motor (16) of the processing head (12).
3. 3. The computer-implemented method of claim 1 or 2, wherein the machine learning model is a decision tree (30).
4. 4. A computer-implemented method according to claim 3, wherein assigning (S502) the user to the classification of sensitivity of the treatment head (12) of the appliance (10) using a machine learning model based on the received data comprises: Calculating a plurality of predictors (38) using the data representative of the physical parameter; inputting the plurality of predictors (38) to an input of the decision tree (30), and -Outputting from the decision tree (30) a classification of the sensitivity of the user to the treatment head (12) of the appliance (10).
5. 5. The computer-implemented method of claim 4, wherein the plurality of predictors (38) are selected from a list of predictors comprising: A binning distribution of motor power values from the first three uses, a binning distribution of motor power from each of the first through third uses, and a binning distribution of power values from the last use, wherein optionally the first class (34) is for non-sensitive users, the second class (35) is for normally sensitive users, and the third class (36) is for sensitive users.
6. 6. A computer-implemented method of calculating a degree of wear of a treatment head (12) of an appliance (10), comprising: Assigning a user to a classification of sensitivities using the method according to any of the preceding claims, wherein the machine learning model is a second machine learning model; determining (S704) a maximum head wear value for the user based on the classification of the user; the process head wear is estimated by: receiving (S200) from the sensor (18) a physical parameter associated with operating the appliance (10), Estimating (S202) process head wear based on the sensed physical parameters using a first machine learning model; calculating (S804) a process head wear level based on the determined maximum process head wear value and the estimated process head wear, and A signal is sent (S806) indicating the calculated degree of wear of the processing head.
7. 7. The computer-implemented method of claim 6, wherein the calculating the degree of wear of the processing head (12) includes using: 7 Where D is the process head wear or extent, cRPS represents the estimated process head wear, and EoL RPS represents the user's maximum process head wear.
8. 8. A computer-implemented method of training a machine learning model to assign a user to a classification of sensitivity of a processing head (12) of an appliance (10), the computer-implemented method comprising: -receiving (S600) a dataset comprising a plurality of classifications, and a plurality of predictors (38) derived from data sensed by a sensor (18) of the appliance (10), the data representing a plurality of physical parameters associated with operating the appliance (10); Inputting (S602) the plurality of predictors (38) into the machine learning model to assign the user to a category according to one of the plurality of categories, and The machine learning model is optimized (S604) to reduce errors between the assigned classification and the classification in the dataset.
9. 9. The computer-implemented method of claim 8, wherein the machine learning model is a decision tree, wherein the classification includes a first class (34) for non-sensitive users, a second class (35) for normal sensitive users, and a third class (36) for sensitive users.
10. 10. The computer-implemented method of claim 9, wherein optimizing the machine learning model is performed using classification and regression tree algorithms.
11. 11. The computer-implemented method of any of claims 8 to 10, wherein the physical parameter comprises current and/or power for driving a motor (16) of the processing head (12).
12. 12. The computer-implemented method of claim 11, wherein the plurality of predictors (38) are selected from a list of predictors including a binning distribution of motor power values from a previous three uses, a binning distribution of motor power from a first use, a binning distribution of motor power values from each of a first through third use, and a binning distribution of power values from a last use.
13. 13. The computer-implemented method of any of claims 8 to 12, comprising generating the plurality of classifications by receiving a score from a user for each aspect of using the appliance, and calculating a threshold for each classification based on differences between sub-populations of the user, wherein optionally the aspects include overall performance, comfort during use, dehairing during use, burning during use, and redness during use.
14. 14. The computer-implemented method of any of the preceding claims, wherein the appliance (10) is a personal care appliance (10), and optionally wherein the treatment head (12) is a cutting element (12).
15. 15. An appliance (10), comprising: an attachment for attaching a processing head (12) thereto; -a sensor (18) for sensing a physical parameter associated with operating the appliance (10); A controller (20) comprising a processor (22) and a memory (24) having instructions stored thereon which, when executed by the processor, cause the processor to perform the computer-implemented method of any of claims 1 to 7.

Description

Sensitivity classification of processing head users Technical Field The subject matter of the present disclosure relates to a computer-implemented method of classifying a user's sensitivity to a treatment head of an appliance, a computer-implemented method of calculating a degree of treatment head wear of an appliance, a computer-implemented method of training a machine learning model to assign a user to a classification of a sensitivity to a treatment head of an appliance, a transitory or non-transitory computer-readable medium, and an appliance. Background Currently, to understand the wear of the cutting elements, personal care appliances rely on pad printing for display or on extended shaving times, such as 170 minutes. Both of these approaches have their drawbacks, resulting in inability to adapt to the individual characteristics of a particular user's beard, or to cover multiple ways of using a personal care appliance. As a result, they may provide inaccurate information about the wear of the cutting elements, resulting in a suboptimal shaving experience. US 2021/0216891A1 discloses a computer-implemented method of analyzing shaves. The method involves receiving context data associated with a user from a data source. The received context data is used to train a machine learning model. User data is received from a user. A durability cluster of the user is determined based on the received user data and the trained machine learning model. A shaving improvement action is performed based on the determined durability cluster. The context data relates to shaving behavior, shaving performance scores, demographics, shaving habits, hair characteristics or skin characteristics. It is an object of the presently disclosed subject matter to improve upon the prior art. Disclosure of Invention According to a first aspect of the present invention there is provided a computer-implemented method of classifying sensitivity of a user to a treatment head of an appliance, the computer-implemented method comprising receiving data from a sensor of the appliance representative of a physical parameter associated with operation of the appliance, assigning the user to a classification of sensitivity of the treatment head of the appliance using a machine learning model based on the received data, and transmitting a signal indicative of the assigned classification. In this way, the user has a more consistent shaving experience and knows when to replace their cutting elements, thereby reducing the risk of having to go through painful shaving in order to understand when to replace the cutting elements. In one embodiment, the physical parameters include current and/or power for a motor driving the processing head. In one embodiment, the machine learning model is a decision tree. In one embodiment, assigning a user to a classification of sensitivity to a processing head of the appliance using a machine learning model based on the received data includes calculating a plurality of predictors using data representative of the physical parameter, inputting the plurality of predictors to an input of the decision tree, and outputting from the decision tree a classification of sensitivity of the user to a processing head of the appliance. In one embodiment, the plurality of predictors is selected from a list of predictors including a binning distribution of motor power values from a previous three uses, a binning distribution of motor power from a first use, a binning distribution of motor power values from each of the first through third uses, and a binning distribution of power values from a last use. The binned distribution of motor power values from the first to third trees may be the average or maximum motor power values from the first to third trees. In one embodiment, where users are classified, a first class is for non-sensitive users, a second class is for normally sensitive users, and a third class is for sensitive users. According to one aspect of the present invention there is provided a computer implemented method of calculating the degree of treatment head wear of an appliance, the method comprising assigning a user to a classification of sensitivity using the method of any preceding claim, wherein the machine learning model is a second machine learning model, determining a maximum treatment head wear value for the user based on the classification of the user, estimating treatment head wear by receiving a physical parameter associated with operating the appliance from the sensor, estimating treatment head wear based on the sensed physical parameter using the first machine learning model, calculating the degree of treatment head wear based on the determined maximum treatment head wear value and the estimated treatment head wear, and transmitting a signal indicative of the calculated treatment head wear degree. In one embodiment, calculating the degree of treatment head wear includes using: where D is the degree of wear of the proce