Abstract: Methods, systems, and apparatuses for content recommendations based on user activity data are described herein. An analytics subsystem may receive first activity data. The first activity may be indicative of a plurality of first engagements with a first plurality of media assets. At least one machine learning model may be configured to receive an input of activity data, such as the first activity data, and to determine at least one content recommendation on that basis. The at least one content recommendation may comprise a recommendation for at least one media asset. The at least one media asset may be associated with at least one media asset classification. The analytics subsystem may send the at least one content recommendation.

Abstract: A humidification system can include a heater base, a humidification chamber, and a breathing circuit. A cartridge can be removably coupled to the heater base. The cartridge can include various sensors, probes, sensor wire connectors, heater wire connectors, and/or other features. The cartridge can include features configured to mate with corresponding features on the humidification chamber and the heater base. The cartridge includes a memory, such as an EEPROM, or other suitable storage device. When the cartridge is installed on the heater base, the memory is electrically connected to a processor and/or memory of the heater base. Various models of cartridges can be produced for use with different humidification chambers, breathing circuits, and/or therapies. A connector can be configured to couple an inspiratory conduit to an outlet port of the humidification chamber. The connector can provide a pneumatic connection to the outlet port and an electrical connection to the cartridge.

Abstract: Methods, systems, and apparatuses for content recommendations based on user activity data are described herein. An analytics subsystem may receive first activity data. The first activity may be indicative of a plurality of first engagements with a first plurality of media assets. At least one machine learning model may be configured to receive an input of activity data, such as the first activity data, and to determine at least one content recommendation on that basis. The at least one content recommendation may comprise a recommendation for at least one media asset. The at least one media asset may be associated with at least one media asset classification. The analytics subsystem may send the at least one content recommendation.