FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2276722833> ?p ?o ?g. }

• W2276722833 abstract "Recognizing human activities with sensors next to the body has become more important over the years, aiming to create or improve systems in elder care support, health/fitness monitoring, and assisting those with cognitive disorders. This task of recognizing activities taking place at a certain moment when considering only one individual user is called Activity Recognition. With the development of new technologies, like smartphones, it was possible to overcome the barrier of the person having to use multiple body worn sensors and passing to use only one, in his trousers’ front pocket. Being able to give a new solution for this problem is a huge motivation, besides having the pleasure of working with Android technology that is leading the market and helping the user experience with is phone. Having the smartphones a triaxial accelerometer built in it is possible to create applications that are capable of recognizing the activities of the user with great accuracy. This thesis aims to meet this new solution, creating an Android application, for the problem of recognizing the activities performed by the user and treating it as a classification problem. To embark into a path that leads to the solution it was necessary to study previous works in order to trace this path. The main objective was to understand what were considered the common human activities and what approaches could be taken when dealing with this classification problem, like supervised or semi-supervised learning. What were the most usual classifiers, what their differences were and how could we compare them. With this study we decided to built an Android application that explore supervised and semi-supervised learning with both one-step classification (classifying the data in Standing Idle, Sitting, Running and Walking) and hierarchical classification (having this approach two classifications, first in Dynamic and Static activities and then inside Dynamic into Running and Walking and inside Static into Standing Idle and Sitting). On supervised learning a model is created and it stays static along the time. On semi-supervised learning some instances labeled by the model are added to the training file in order to create a new model and have a training file up to date and with activities from the current user. The main components of the application, how they interact its basic architecture are presented. In order to compare the classifiers’ performance their accuracy was chosen. Curves of precision/recall were also created to understand and evaluate the models’ information retrieval system. Since we were working on a mobile application the memory and battery usage and time spent on the classification were also an issue. To check the feasibility of the application these issues had to be monitored. The classifiers used in the experiments were Naive Bayes and Hoeffding trees. The main conclusions from this study are: 1) hierarchical classification has better performance than one-step classification; 2) the best mix for the hierarchical classification is using Naive Bayes in the first classification and Hoeffding trees in the second one; 3) Semisupervised learning is globally better than supervised classification; 4) Naive Bayes consumes less battery than Hoeffding trees." @default.
• W2276722833 created "2016-06-24" @default.
• W2276722833 creator A5029075161 @default.
• W2276722833 date "2012-01-01" @default.
• W2276722833 modified "2023-09-23" @default.
• W2276722833 title "Activity recognition from smartphone sensing data" @default.
• W2276722833 cites W13530706 @default.
• W2276722833 cites W1481162702 @default.
• W2276722833 cites W1897824230 @default.
• W2276722833 cites W1968194699 @default.
• W2276722833 cites W1968458973 @default.
• W2276722833 cites W2008056655 @default.
• W2276722833 cites W2017634428 @default.
• W2276722833 cites W2068714596 @default.
• W2276722833 cites W2082553115 @default.
• W2276722833 cites W2105046342 @default.
• W2276722833 cites W2108307068 @default.
• W2276722833 cites W2135335717 @default.
• W2276722833 cites W2136504847 @default.
• W2276722833 cites W2139038299 @default.
• W2276722833 cites W2140190241 @default.
• W2276722833 cites W2170265372 @default.
• W2276722833 cites W2210036442 @default.
• W2276722833 cites W2497373516 @default.
• W2276722833 cites W3112596628 @default.
• W2276722833 cites W98188630 @default.
• W2276722833 hasPublicationYear "2012" @default.
• W2276722833 type Work @default.
• W2276722833 sameAs 2276722833 @default.
• W2276722833 citedByCount "0" @default.
• W2276722833 crossrefType "dissertation" @default.
• W2276722833 hasAuthorship W2276722833A5029075161 @default.
• W2276722833 hasConcept C107457646 @default.
• W2276722833 hasConcept C111919701 @default.
• W2276722833 hasConcept C121687571 @default.
• W2276722833 hasConcept C127413603 @default.
• W2276722833 hasConcept C138885662 @default.
• W2276722833 hasConcept C154945302 @default.
• W2276722833 hasConcept C2778707766 @default.
• W2276722833 hasConcept C41008148 @default.
• W2276722833 hasConcept C41895202 @default.
• W2276722833 hasConcept C557433098 @default.
• W2276722833 hasConcept C89805583 @default.
• W2276722833 hasConceptScore W2276722833C107457646 @default.
• W2276722833 hasConceptScore W2276722833C111919701 @default.
• W2276722833 hasConceptScore W2276722833C121687571 @default.
• W2276722833 hasConceptScore W2276722833C127413603 @default.
• W2276722833 hasConceptScore W2276722833C138885662 @default.
• W2276722833 hasConceptScore W2276722833C154945302 @default.
• W2276722833 hasConceptScore W2276722833C2778707766 @default.
• W2276722833 hasConceptScore W2276722833C41008148 @default.
• W2276722833 hasConceptScore W2276722833C41895202 @default.
• W2276722833 hasConceptScore W2276722833C557433098 @default.
• W2276722833 hasConceptScore W2276722833C89805583 @default.
• W2276722833 hasLocation W22767228331 @default.
• W2276722833 hasOpenAccess W2276722833 @default.
• W2276722833 hasPrimaryLocation W22767228331 @default.
• W2276722833 hasRelatedWork W2050230532 @default.
• W2276722833 hasRelatedWork W2051167698 @default.
• W2276722833 hasRelatedWork W2466310942 @default.
• W2276722833 hasRelatedWork W2512231435 @default.
• W2276722833 hasRelatedWork W2518241345 @default.
• W2276722833 hasRelatedWork W2739254393 @default.
• W2276722833 hasRelatedWork W2773553693 @default.
• W2276722833 hasRelatedWork W2782796106 @default.
• W2276722833 hasRelatedWork W2783576832 @default.
• W2276722833 hasRelatedWork W2891048219 @default.
• W2276722833 hasRelatedWork W2942392951 @default.
• W2276722833 hasRelatedWork W2945171767 @default.
• W2276722833 hasRelatedWork W2948777090 @default.
• W2276722833 hasRelatedWork W2957007523 @default.
• W2276722833 hasRelatedWork W3020876265 @default.
• W2276722833 hasRelatedWork W3045020010 @default.
• W2276722833 hasRelatedWork W3082714420 @default.
• W2276722833 hasRelatedWork W3134507349 @default.
• W2276722833 hasRelatedWork W3158460577 @default.
• W2276722833 hasRelatedWork W3166459988 @default.
• W2276722833 isParatext "false" @default.
• W2276722833 isRetracted "false" @default.
• W2276722833 magId "2276722833" @default.
• W2276722833 workType "dissertation" @default.