{"393301":{"#nid":"393301","#data":{"type":"event","title":"Ph.D. Proposal Defense by Daniel Kohlsdorf","body":[{"value":"Title: \u003Cstrong\u003EData Mining in Large Audio Collections of Dolphin Signals\u003C\/strong\u003E\u003Cstrong\u003E\u003Cbr \/\u003E\u003C\/strong\u003E\u003Cstrong\u003E\u003Cbr \/\u003E\u003C\/strong\u003E\u003Cstrong\u003EDaniel Kohlsdorf\u003C\/strong\u003EPh.D. StudentSchool of Interactive ComputingCollege of ComputingGeorgia Institute of Technology\u003Cbr \/\u003E\u003Cbr \/\u003E\u003Cstrong\u003EDate: Thursday, April 9th, 2015\u003C\/strong\u003E\u003Cstrong\u003ETime: 10 AM to 12 Noon EST\u003C\/strong\u003E\u003Cstrong\u003ELocation: TSRB 223\u003C\/strong\u003E\u003Cbr \/\u003E\u003Cstrong\u003ECommittee:\u003C\/strong\u003EDr. Thad Starner, School of Interactive Computing, Georgia TechDr. Irfan Essa, School of Interactive Computing, Georgia TechDr. Charles Isbell, School of Interactive Computing, Georgia TechDr. Michael Beetz, University BremenDr. Denise Herzing, Wild Dolphin Project\u003Cstrong\u003E\u003Cbr \/\u003E\u003C\/strong\u003E\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003E\u0022The presented research addresses signal processing and data mining in large collections of audible dolphin communication.\u003C\/p\u003E\u003Cp\u003EThe goal is to develop a novel system that is capable of automatically finding patterns and their correspondences to dolphin behavior.\u003C\/p\u003E\u003Cp\u003EThe system will help marine biologists to perform communication analysis automatically.\u003C\/p\u003E\u003Cp\u003EBiologists can interactively find and test novel hypothesis using a user interface on top of the data mining system.\u003C\/p\u003E\u003Cp\u003ECurrent research in animal communication research suffers from the slow speed of manual data analysis.\u003C\/p\u003E\u003Cp\u003EOften researchers search and annotate audio and video material using manual measurements. Often these measurements are subjective and not formally defined.\u003C\/p\u003E\u003Cp\u003EFinding patterns of communication that relate to observable behavior without metrics for comparison is a tedious process.\u003C\/p\u003E\u003Cp\u003EThe process can take several years from data collection to publication.\u003C\/p\u003E\u003Cp\u003ETherefore, I propose a data mining system for audible animal communication.\u003C\/p\u003E\u003Cp\u003EThe system automatically learns a representation in which animal communication patterns can be easily found and compared.\u003C\/p\u003E\u003Cp\u003EFurthermore, the system will be able to find communication patterns and segmentations using\u003C\/p\u003E\u003Cp\u003Ealgorithms adopted from speech recognition and time series motif discovery.\u003C\/p\u003E\u003Cp\u003EIf integrated in a user interface,\u003C\/p\u003E\u003Cp\u003Ean algorithm that just segments and finds patterns in animal communication can already\u003C\/p\u003E\u003Cp\u003Ehelp researchers in their research effort. However, the proposed system is capable\u003C\/p\u003E\u003Cp\u003Eof testing hypothesis about animal communication in addition.\u003C\/p\u003E\u003Cp\u003EFor example, a researcher might ask if different groups of animals use different patterns.\u003C\/p\u003E\u003Cp\u003EIn this case the researchers could collect communication data from multiple groups,\u003C\/p\u003E\u003Cp\u003Ediscover all patterns in the communication jointly and then compare the statistical distributions for\u003C\/p\u003E\u003Cp\u003Eeach group against the others.\u003C\/p\u003E\u003Cp\u003EIn this way, researchers can not only visually inspect the resulting patterns but also gain a novel, quantitive analysis method.\u003C\/p\u003E\u003Cp\u003EI hypothesize that feature learning and automatic segmentation of audible dolphin communication along with statistical\u003C\/p\u003E\u003Cp\u003Ecommunication models can provide valuable insight into dolphin behavior useful to marine biologists\u003C\/p\u003E\u003Cp\u003Efor retrospective analysis as well as scientific hypothesis generation and testing.\u0022\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Data Mining in Large Audio Collections of Dolphin Signals"}],"uid":"27707","created_gmt":"2015-04-02 11:46:33","changed_gmt":"2016-10-08 02:11:31","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2015-04-09T11:00:00-04:00","event_time_end":"2015-04-09T13:00:00-04:00","event_time_end_last":"2015-04-09T13:00:00-04:00","gmt_time_start":"2015-04-09 15:00:00","gmt_time_end":"2015-04-09 17:00:00","gmt_time_end_last":"2015-04-09 17:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"11038","name":"CoC PhD Thesis Proposal Announcement"},{"id":"1808","name":"graduate students"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78771","name":"Public"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}