Devika B Kumar (92217009) <br/><br/>
Daily human activities recognition using heterogeneous sensors from smart devices 

 - MSC DA 2017-2019 
<br/><br/> Physical activities play a very important role in our physical and mental<br/>well-being. The lack of physical activities can negatively aect our wellbeing.<br/>Though people know the importance of physical activities, still they<br/>need regular motivational feedback to remain active in their daily life. In<br/>order to give them proper motivational feedback, we need to recognize their<br/>physical activates rst (in our case, the main target group is knowledge workers).<br/>Therefore, this research is about recognizing human context (condition,<br/>activity and situation etc.) using heterogeneous sensors. If recognized reliably,<br/>this context can enable novel well-being applications in dierent elds,<br/>for example, healthcare. As a rst step to achieve this goal, we recognize<br/>some physical activities using smartphone sensors like accelerometer, gyroscope,<br/>and magnetometer. Moreover, we are simulating a smartphone on a<br/>wrist position as a smart watch and want to see the possibilities of activity<br/>recognition with upcoming smart watches. We want to reliably recognize<br/>physical activities using heterogeneous sensor information, that may be incomplete<br/>or unreliable. We are currently working on improving the existing<br/>work by investigating and solving the open challenges in activity recognition<br/>using smartphone sensors. 
<br/><br/>
<br/><br/><label>Subjects--Topical Terms: </label><br/>RANDOM FOREST<br/>DECISION TREE<br/>NAIVE BAYES<br/>MACHINE LEARNING