#The forest mods 0.40 series#

Klassificeringsresultaten visade att random forest-metoden hade god fГ¶rmГҐga att prioritera bland features. MГ¶jligheten att anvГ¤nda den slutliga modellen pГҐ en processor med begrГ¤nsad hГҐrdvarukapacitet togs i beaktning dГҐ studiens metoder valdes. En decision tree-klassificerare anvГ¤ndes fГ¶r att utfГ¶ra klassificeringen av de mГ¤nskliga aktiviteterna fГ¶r bГҐda metoderna. BГҐda metoderna anvГ¤nde samma initiala samling av automatiskt genererade features. Hypotestestmetoden kombinerades med ett korrelationsfilter i denna studie. Den fГ¶rsta metoden vГ¤ljer features genom att anvГ¤nda individuella hypotestester och den andra metoden anvГ¤nder en random forest-klassificerare som en inbГ¤ddad feature-vГ¤ljare. Abstract Дenna studie utvГ¤rderar tvГҐ metoder som automatiskt extraherar features fГ¶r att klassificera accelerometerdata frГҐn periodiska och sporadiska mГ¤nskliga aktiviteter. In addition to the classification performance this thesis studies the potential business benefits that automation of feature extractioncan result in.

The first method, however, only had a macro average F1 score of 0.40 and a weighted average F1 score of 0.63 when using the same number of features. With 23 features selected it had a macro average F1 score of 0.84 and a weighted average F1 score of 0.93. The classification results showed that the random forest method was good at prioritizing among features. A decision tree classifier was used to perform the human activity recognition task for both methods.The possibility of running the developed model on a processor with limited computing power was taken into consideration when selecting methods for evaluation.

#The forest mods 0.40 series#

Both methods used the same initial pool of automatically generated time series features.

The hypothesis test was combined with a correlation filter in this study. The first method selects features using individual hypothesis tests and the second one is using a random forest classifier as an embedded feature selector. This thesis evaluates two methods for automatic feature extraction to classify the accelerometer data of periodic and sporadic human activities. 2019 (English) Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE credits Student thesis Abstract