Abstract-Swallowing is an important part of the dietary non-invasive sensors. However, a rough estimation of food process. This paper presents an investigation to detect and type. e.g. ratio of fluid and solid nutrient combined with the classify normal swallowing during eating and drinking from timing information, e.g. event schedule and meal durations electromyography and microphone sensors. The non-invasive ' sensors are selected in order to integrate them into a collar-overthda,Already povides warsl bsisorsbehaviura like fabric for continuous monitoring of swallowing activity over monitoring. Although focusing on wearable sensors we expect a day. We compare methods for the detection of individual that additional information can be obtained in combination swallowing events from continuous sensor data. Furthermore with a supportive environment, e.g. food products with RFwe present a classifier comparison for the swallowing event identification tags, intelligent shopping lists or dietary moniproperties volume and viscosity. The methods are evaluated .a'es on experimental data and a performance analysis is shown. torWng tables.Moreover we present a class skew analysis based on the metricsWe target a non-invasive wearable system relying on inprecision and recall.formation from the following three sensing domains: 1) the Index Terms-Swallowing detection, event detection, bolus identification of characteristic arm and trunk movements asviscosity classification, bolus volume classification, sensor collar. sociated with food intake using inertial sensors [1], 2) the analysis of food chewing sounds from an ear microphone [2]