Cluster-Based Sleep Analysis

Abstract

Various types of data can be collected regarding the physical or mental health of a user, as may relate to sleep of the user over a period of time. Health metrics can be determined from this data that can enable the user to be associated with a particular health type or category. For sleep, this can include associating the user with a sleep animal that has specific characteristics. This can help a user to better understand that user's sleep, and how that sleep compares to sleep of others. In addition to being able to provide health information in a way that is easy to understand, such an approach can also help to make more accurate recommendations or take specific actions to help improve the health of a user, such as to improve sleep. This can include making recommendations to a user or automatically adjusting operation of at least one device.

What is claimed is: 
     
         1 . A computer-implemented method, comprising:
 obtaining sleep-related data for a user collected over multiple sleep periods;   determining, from the sleep-related data, user values for a set of sleep metrics;   comparing the user values against sleep metric values for each of a set of sleeper types to identify a sleeper type that most closely represents the sleep of the user;   providing, for presentation to the user, information pertaining to the identified sleeper type for the user and a comparison of the user values to the sleep metric values for an average user of the identified sleeper type; and   causing an electronic device, associated with the user, to make at least one adjustment based at least in part upon the identified sleeper type for the user.   
     
     
         2 . The computer-implemented method of  claim 1 , further comprising:
 generating at least one recommendation to present to the user to help improve the sleep of the user, the at least one recommendation determined based at least in part upon the user values for the sleep metrics and the identified sleeper type for the user.   
     
     
         3 . The computer-implemented method of  claim 2 , further comprising:
 determining a current state of at least one of the user or an environment of the user, wherein the at least one recommendation or the at least one adjustment is determined based further upon the current state.   
     
     
         4 . The computer-implemented method of  claim 1 , wherein at least a subset of the sleep related data is obtained from a wearable monitoring device worn by the user, a computing device associated with the user, or a sensory device located in an environment surrounding the user. 
     
     
         5 . The computer-implemented method of  claim 1 , further comprising:
 collecting sleep-related data for a population of users;   determining an initial set of sleep features from the sleep-related data; and   performing clustering of the initial set of sleep features to generate a set of clusters corresponding to the set of sleeper types.   
     
     
         6 . The computer-implemented method of  claim 5 , further comprising:
 identifying the sleeper type for the user using one or more distance heuristics to compare the user values against the respective sleep metric values of the set of clusters.   
     
     
         7 . The computer-implemented method of  claim 1 , wherein the information pertaining to the identified sleeper type includes a graphical object representative of the identified sleeper type, the graphical object capable of being animated to convey a subset of the information to the user. 
     
     
         8 . The computer-implemented method of  claim 7 , wherein the graphical object is a sleep animal. 
     
     
         9 . The computer-implemented method of  claim 1 , wherein the at least one adjustment to be made by the electronic device includes at least one of a change in display, volume, brightness, mode, operational state, power level, communication, configuration, or operation, and wherein the adjustment is intended to assist a user in achieving a desired sleep goal. 
     
     
         10 . The computer-implemented method of  claim 1 , wherein the multiple sleep periods correspond to multiple days in which the user had an opportunity to sleep. 
     
     
         11 . A monitoring device, comprising:
 a display device;   a non-invasive measurement system;   at least one processor; and   memory including instructions that, when executed by the at least one processor, cause the monitoring device to:
 obtain, using the non-invasive measurement system, physiological data for a user collected over multiple data collection periods; 
 cause user values for a set of health metrics to be determined from at least the physiological data; 
 cause the user values to be compared against health metric values for each of a set of health types to identify a health type that most closely represents the health of the user; 
 display, on the display device, information pertaining to the identified health type for the user; and 
 trigger at least one automated change based at least in part upon the identified health type and the physiological data. 
   
     
     
         12 . The monitoring device of  claim 11 , wherein the automated change is to at least one of the monitoring device or an external electronic device, and wherein the automated change is intended to help improve the health of the user. 
     
     
         13 . The monitoring device of  claim 11 , wherein the instructions when executed further cause the monitoring device to:
 display, on the display device, at least one recommendation for improving the health of the user, the recommendation determined based at least in part upon the identified health type. cm  14 . The monitoring device of  claim 11 , wherein the user values for the set of health metrics are further determined based upon sleep-related data obtained from one or more external data sources.   
     
     
         15 . The monitoring device of  claim 11 , wherein the identified health type for the user is a sleeper type, and wherein the information pertaining to the identified health type includes a graphical object representative of the sleeper type, the graphical object capable of being animated to convey a subset of the information to the user. 
     
     
         16 . A sleep improvement system, comprising:
 one or more processors; and   memory including instructions that, when executed by the one or more processors, causes the sleep improvement system to:
 obtain, from one or more electronic devices, sleep-related data for a user collected over multiple sleep periods; 
 determine, from the sleep-related data, user values for a set of sleep metrics; 
 compare the user values against sleep metric values for each of a set of sleeper types to identify a sleeper type that most closely represents the sleep of the user; and 
 cause an electronic device, of the one or more electronic devices, to make at least one adjustment based at least in part upon the identified sleeper type for the user. 
   
     
     
         17 . The sleep improvement system of  claim 16 , wherein the at least one adjustment includes providing, for presentation to the user, information pertaining to the identified sleeper type for the user and a comparison of the user values to the sleep metric values for the average user of the identified sleeper type. 
     
     
         18 . The sleep improvement system of  claim 17 , wherein the information pertaining to the identified sleeper type includes a graphical object representative of the sleeper type, the graphical object capable of being animated to convey a subset of the information to the user. 
     
     
         19 . The sleep improvement system of  claim 16 , wherein the one or more electronic devices include at least one of a wearable monitoring device, a sensory device, a user computing device, or a network-connected smart device capable of providing at least a subset of the sleep-related data. 
     
     
         20 . The sleep improvement system of  claim 16 , wherein the instructions when executed further cause the sleep improvement system to:
 generate at least one recommendation to present to the user to help improve the sleep of the user, the at least one recommendation determined based at least in part upon the user values for the sleep metrics and the identified sleeper type for the user.