%0 Journal Article %1 keyhere %A Molnár, Csaba %A Kaplan, Frédéric %A Roy, Pierre %A Pachet, François %A Pongrácz, Péter %A Dóka, Antal %A Miklósi, Ádám %D 2008 %J Animal Cognition %K ML animal_cognition imported toread %P -- %T Classification of dog barks: a machine learning approach %U http://dx.doi.org/10.1007/s10071-007-0129-9 %X Abstract  In this study we analyzed the possible context-specific and individual-specific features of dog barks using a new machine-learning algorithm. A pool containing more than 6,000 barks, which were recorded in six different communicative situations was usedas the sound sample. The algorithm’s task was to learn which acoustic features of the barks, which were recorded in differentcontexts and from different individuals, could be distinguished from another. The program conducted this task by analyzingbarks emitted in previously identified contexts by identified dogs. After the best feature set had been obtained (with whichthe highest identification rate was achieved), the efficiency of the algorithm was tested in a classification task in whichunknown barks were analyzed. The recognition rates we found were highly above chance level: the algorithm could categorizethe barks according to their recorded situation with an efficiency of 43% and with an efficiency of 52% of the barking individuals.These findings suggest that dog barks have context-specific and individual-specific acoustic features. In our opinion, thismachine learning method may provide an efficient tool for analyzing acoustic data in various behavioral studies.

@article{keyhere, abstract = {Abstract  In this study we analyzed the possible context-specific and individual-specific features of dog barks using a new machine-learning algorithm. A pool containing more than 6,000 barks, which were recorded in six different communicative situations was usedas the sound sample. The algorithm’s task was to learn which acoustic features of the barks, which were recorded in differentcontexts and from different individuals, could be distinguished from another. The program conducted this task by analyzingbarks emitted in previously identified contexts by identified dogs. After the best feature set had been obtained (with whichthe highest identification rate was achieved), the efficiency of the algorithm was tested in a classification task in whichunknown barks were analyzed. The recognition rates we found were highly above chance level: the algorithm could categorizethe barks according to their recorded situation with an efficiency of 43% and with an efficiency of 52% of the barking individuals.These findings suggest that dog barks have context-specific and individual-specific acoustic features. In our opinion, thismachine learning method may provide an efficient tool for analyzing acoustic data in various behavioral studies.}, added-at = {2008-01-19T10:00:18.000+0100}, author = {Molnár, Csaba and Kaplan, Frédéric and Roy, Pierre and Pachet, François and Pongrácz, Péter and Dóka, Antal and Miklósi, Ádám}, biburl = {https://www.bibsonomy.org/bibtex/21221ba4b00c26300a9c37aec682c8fc6/unhammer}, description = {SpringerLink - Journal Article}, interhash = {f9ac88ecc3157b47d2c8e570e888349d}, intrahash = {1221ba4b00c26300a9c37aec682c8fc6}, journal = {Animal Cognition}, keywords = {ML animal_cognition imported toread}, pages = {--}, timestamp = {2008-01-19T10:00:19.000+0100}, title = {Classification of dog barks: a machine learning approach}, url = {http://dx.doi.org/10.1007/s10071-007-0129-9}, year = 2008 }