{"@type": "dcat:Dataset", "accessLevel": "public", "accrualPeriodicity": "irregular", "bureauCode": ["026:00"], "contactPoint": {"@type": "vcard:Contact", "fn": "Elizabeth Foughty", "hasEmail": "mailto:elizabeth.a.foughty@nasa.gov"}, "description": "ANALYZING AVIATION SAFETY REPORTS: FROM TOPIC MODELING TO\r\nSCALABLE MULTI-LABEL CLASSIFICATION\r\n\r\nAMRUDIN AGOVIC*, HANHUAI SHAN*, AND ARINDAM BANERJEE*\r\n\r\nAbstract. The Aviation Safety Reporting System (ASRS) is used to collect voluntarily submitted\r\naviation safety reports from pilots, controllers and others. As such it is particularly useful\r\nin researching aviation safety deficiencies. In this paper we address two challenges related to the\r\nanalysis of ASRS data: (1) the unsupervised extraction of meaningful and interpretable topics\r\nfrom ASRS reports and (2) multi-label classification of ASRS data based on a set of predefined\r\ncategories. For topic modeling we investigate the practical usefulness of Latent Dirichlet Allocation\r\n(LDA) when it comes to modeling ASRS reports in terms of interpretable topics. We also\r\nutilize LDA to generate a more compact representation of ASRS reports to be used in multi-label\r\nclassification. For multi-label classification we propose a novel and highly scalable multi-label classification\r\nalgorithm based on multi-variate regression. Empirical results indicate that our approach\r\nis superior to several baseline and state-of-the-art approaches.", "distribution": [{"@type": "dcat:Distribution", "description": "ANALYZING AVIATION SAFETY REPORTS: FROM TOPIC MODELING TO SCALABLE MULTI-LABEL CLASSIFICATION", "downloadURL": "https://c3.nasa.gov/dashlink/static/media/publication/Paper_7_.pdf", "format": "PDF", "mediaType": "application/pdf", "title": "Paper 7 .pdf"}], "identifier": "DASHLINK_229", "issued": "2010-10-13", "keyword": ["ames", "dashlink", "nasa"], "landingPage": "https://c3.nasa.gov/dashlink/resources/229/", "modified": "2025-03-31", "programCode": ["026:029"], "publisher": {"@type": "org:Organization", "name": "Dashlink"}, "title": "ANALYZING AVIATION SAFETY REPORTS: FROM TOPIC MODELING TO SCALABLE MULTI-LABEL CLASSIFICATION"}