Table 1. Machine learning techniques with FRM applications.
References [1] P. O. J. Kelliher, D. Wilmot, J. Vij, and P. J. M. Klumpes, "A common risk classification
system for the actuarial profession''
[3] W.-Y. Lin, Y.-H. Hu, and C.-F. Tsai, "Machine learning in financial crisis prediction: A
survey,'' IEEE Trans. Syst., Man, Cybern.
[4] J. Garrido, C. Genest, and J. Schulz, "Generalized linear models for dependent frequency
and severity of insurance claims''
RNN BASED SPELL CORRECTION MODEL FOR AZERBAIJANI LANGUAGE Sona Mehdizada Baku Higher Oil School Baku, Azerbaijan sona.mehtizada.std@bhos.edu.az Supervisor: Ramil Shukurov Keywords: Natural Language Processing, Recurrent Neural Networks, Spell Correction,
Attention Mechanism
Abstract The spell correction system carries utmost importance in the creation of documents in
any language. It is software that detects orthographic mistakes and suggests the correct
spelling of the word. Due to the issue of not having a proper spelling system for Azerbaijan
language, books, documents, or emails should be proofread and checked for typing errors by
other people which is also not as reliable by virtue of human factor and inefficiency in terms of
time. Compared to the other languages, Azerbaijan language has more complex morphological
Machine Learning
based learning methods
Supervised
Regression
Volatility Forecasting,
Claims Modelling,
Classification
Fraud Detection,
Credit Scoring & Bankruptcy
prediction
Unsupervised
Clustering
Credit Scoring & Bankruptcy
prediction
Anomaly Detection
Fraud Detection
[2] T. Bollerslev, "Generalized autoregressive conditional heteroskedasticity,'' J.
Econometrics, 1986.