Leveraging the power of Natural Language Processing for Financial Intelligence System

In this study, we aim to exploit natural language processing (NLP) techniques to develop a financial intelligence system that understands and analyzes online news channels on the basis of companies filtered by specific keywords. The system enables us to immediately notify potential opportunities and threats that may arise for the relevant company portfolio and to take the necessary actions. The architecture can enrich portfolio management, increase a company's credit profitability, offer finance-specific functions and use time and resources effectively in collecting and evaluating information through various metrics. In this direction, we designated an infrastructure and addressed a wide variety of NLP issues to execute the system modules. Various NLP tasks such as text classification, text regression, impact measurement, and Named-Entity Recognition have been successfully solved with the latest techniques. Not only traditional machine learning techniques but also modern deep learning architectures such as RNN and Transformers have been utilized to solve financial tasks.

[1]  Banu Diri,et al.  Comparison of Topic Modeling Methods for Type Detection of Turkish News , 2019, 2019 4th International Conference on Computer Science and Engineering (UBMK).

[2]  Suat Atan,et al.  Borsa İstanbul’da Finansal Haberler ile Piyasa Değeri İlişkisinin Metin Madenciliği ve Duygu (Sentiment) Analizi ile İncelenmesi , 2019, Ankara Üniversitesi SBF Dergisi.

[3]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[4]  Banu Diri,et al.  Türkçe Dokümanlar İçin N-gram Tabanlı Yeni Bir Sınıflandırma(Ng-ind): Yazar, Tür ve Cinsiyet , 2016 .

[5]  Dilek Küçük,et al.  A named entity recognition dataset for Turkish , 2016, 2016 24th Signal Processing and Communication Application Conference (SIU).

[6]  Quoc V. Le,et al.  Distributed Representations of Sentences and Documents , 2014, ICML.

[7]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[8]  Peter Tiño,et al.  Managing Diversity in Regression Ensembles , 2005, J. Mach. Learn. Res..

[9]  George Forman,et al.  An Extensive Empirical Study of Feature Selection Metrics for Text Classification , 2003, J. Mach. Learn. Res..

[10]  Thorsten Joachims,et al.  Text Categorization with Support Vector Machines: Learning with Many Relevant Features , 1998, ECML.

[11]  S. Hochreiter,et al.  Long Short-Term Memory , 1997, Neural Computation.

[12]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[13]  Fatih Sağlam.  Otomatik Duygu Sözlüğü Geliştirilmesi ve Haberlerin Duygu Analizi , 2019 .

[14]  Z. H. Kilimci,et al.  Financial sentiment analysis with Deep Ensemble Models (DEMs) for stock market prediction , 2019 .

[15]  Gyözö Gidófalvi.  Using News Articles to Predict Stock Price Movements , 2001 .