Adaptive class association rule mining for human activity recognition
Citations
252 citations
19 citations
Cites background or methods from "Adaptive class association rule min..."
...• Rule combination strategy: Based on our experiments in [7] we investigated different strategies, cf....
[...]
...Confirming our initial results presented in [7] for the CARMA framework, we observe that the proposed approach outperforms both baselines in accuracy as well as in complex-...
[...]
...For a detailed discussion, we refer to [7]....
[...]
...As described in [7], confidence is estimated by the respective relative frequency of the class contained in the data records covered by the respective rule, while the Laplace value lval(r) of a rule r is determined by lvar(r) = pri+1 ∑ cj∈C p j +|C| , where p r j (and pi ) are the numbers of covered examples by rule r that belong to the respective classes cj considering all possible classes C, and class ci of the rule, respectively....
[...]
...In this paper, an adapted and substantially extended revision of [7], we propose a novel explicative machine learning approach utilizing class association rules (CARs): These are similar to standard association rules [8], however the righthand-side of the rule is assigned a specific class, while the lefthand-side of the rule is made up of a conjunction of attributevalue pairs making up a pattern....
[...]
10 citations
10 citations
8 citations
Cites methods from "Adaptive class association rule min..."
...Then, subgroup discovery can be adapted as a rule generator for class association rule mining, using our CARMA framework presented in [6]....
[...]
References
21,674 citations
8,046 citations
7,062 citations
Additional excerpts
...[1]:...
[...]
4,081 citations
3,223 citations
"Adaptive class association rule min..." refers background in this paper
...Also, Bao and Intille [16] asked 20 subjects to perform some everyday activities while wearing ve biaxial accelerometers on di erent parts of the body....
[...]