Title
Multi-label Categorization of French Death Certificates using NLP and Machine Learning
Abstract
The medical information represents an invaluable source of knowledge concerning the medical history of the patient, but the manner of their presentation make it badly exploited. The idea of this paper is based on the analysis of the death reports written in natural language, which are rich of information, and can be exploited in the calculation of mortality statistics, giving preventive solutions, as well as, help medical professional in their research. This paper proposes our approach to the task of Multi-label Categorization of French death certificates according to ICD-10 (International Classification of Diseases) codes. This approach is based on Machine learning techniques, which is evaluated over CépiDC corpus. The experiment showed that our approach gives interesting results, with an average F1-measue of 79.02%.
Year
DOI
Venue
2017
10.1145/3090354.3090384
BDCA
Field
DocType
ISBN
Data mining,Categorization,Text mining,Computer science,Natural language,Medical history,Artificial intelligence,Natural language processing,Machine learning
Conference
978-1-4503-4852-2
Citations 
PageRank 
References 
1
0.34
9
Authors
2
Name
Order
Citations
PageRank
Allaouzi Imane110.34
Mohamed Ben Ahmed219545.34