Unsupervised Activity Recognition Using Automatically Mined Common Sense

By Danny Wyatt, Matthai Philipose, Tanzeem Choudhury. AAAI Conference on Artificial Intelligence, 2005

316 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/4867b6a9d556f7191da9b59f56f30734943c3304

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Navigate like a cabbie: probabilistic reasoning from observed context-aware behavior (2008, background): “This has been explored in a large number of domains including classifying a user’s posture [17], tracking and classifying user exercises [5], classifying a user’s interruptibility [6] and detecting a variety of daily activities within the home [26, 24].”
Collaborative location and activity recommendations with GPS history data (2010, background): “In [17], activities of daily living such as brushing teeth or making a snack in indoor environment are recognized by using RFID sensors.”
A Scalable Approach to Activity Recognition based on Object Use (2007, methodology): “5) or generated automatically by mining common-sense knowledge from web as described in [27].”
Cross-domain activity recognition (2009, methodology): “also proposed to mine recipes of the activities from the Web as the labeled data, but they only use this knowledge as a prior and trained a Hidden Markov Model from the unlabeled RFID sensor data [29].”
A generalized taxonomy of explanations styles for traditional and social recommender systems (2012, background): “Previous work focused on recognizing an activity from sensor data by ubiquitous computing (Liao et al. 2005; Wyatt et al. 2005; Zheng et al. 2005).”
Sensor-Based Activity Recognition (2012, methodology): “[106] followed Perkowitz’s approach by mining the web to create DBN activity models.”
A Survey on Ambient-Assisted Living Tools for Older Adults (2013, methodology): “complex real world ontology models, some researchers have used online resources such as eHow or WordNet to determine the similarity between known concepts and unknown concepts [109], [110].”
Collaborative Filtering Meets Mobile Recommendation: A User-Centered Approach (2010, methodology): “Beyond the tensor to model the userlocation-activity, we also extract another matrix E ∈ Rm×n from the GPS data to model the user-location visiting relations.”