2018/05/13 by Andrei M. Butnaru, Radu Tudor Ionescu, Butnaru, Andrei M. +1 · 10 citations
Computer Science · Mathematics · #Arabic #Artificial intelligence #Authorship Attribution and Profiling #Character (mathematics) #Computation and Language (cs.CL) #Computer science #FOS: Computer and information sciences #Identification (biology) #Kernel (algebra) #Kernel Fisher discriminant analysis #Kernel method #Linear discriminant analysis #Linguistics #Machine learning #Macro #Mathematics #Modern Standard Arabic #Natural Language Processing Techniques #Natural language processing #Speech Recognition and Synthesis #Speech recognition #Support vector machine #Task (project management) #cs.CL
paper · pdf · doi:10.48550/arxiv.1805.04876
published in arXiv (Cornell University) (Cornell University) · This paper presents the UnibucKernel team's participation at the 2018 Arabic Dialect Identification Shared Task. Accepted at the VarDial Workshop of COLING 2018
openalex publication_date 2018/05/13 · arxiv created 2018/07/28 · arxiv updated 2018/07/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We present a machine learning approach that ranked on the first place in the Arabic Dialect Identification (ADI) Closed Shared Tasks of the 2018 VarDial Evaluation Campaign. The proposed approach combines several kernels using multiple kernel learning. While most of our kernels are based on character p-grams (also known as n-grams) extracted from speech or phonetic transcripts, we also use a kernel based on dialectal embeddings generated from audio recordings by the organizers. In the learning stage, we independently employ Kernel Discriminant Analysis (KDA) and Kernel Ridge Regression (KRR). Preliminary experiments indicate that KRR provides better classification results. Our approach is shallow and simple, but the empirical results obtained in the 2018 ADI Closed Shared Task prove that it achieves the best performance. Furthermore, our top macro-F1 score (58.92%) is significantly better than the second best score (57.59%) in the 2018 ADI Shared Task, according to the statistical significance test performed by the organizers. Nevertheless, we obtain even better post-competition results (a macro-F1 score of 62.28%) using the audio embeddings released by the organizers after the competition. With a very similar approach (that did not include phonetic features), we also ranked first in the ADI Closed Shared Tasks of the 2017 VarDial Evaluation Campaign, surpassing the second best method by 4.62%. We therefore conclude that our multiple kernel learning method is the best approach to date for Arabic dialect identification.