An algorithm for detecting leaks of insider information of financial markets in investment consulting
The paper focuses on revealing insider information leaks of financial markets during investment consulting. An original dataset was created, containing the records of the conversations between consultants and clients, presented in the form of dialogs in text format. The applicability of machine lear...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)
2021-06-01
|
| Series: | Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki |
| Subjects: | |
| Online Access: | https://ntv.ifmo.ru/file/article/20507.pdf |
| _version_ | 1818865477493981184 |
|---|---|
| author | Alisa A. Vorobeva Vladislav V. Gerasimov Yulia V. Li |
| author_facet | Alisa A. Vorobeva Vladislav V. Gerasimov Yulia V. Li |
| author_sort | Alisa A. Vorobeva |
| collection | DOAJ |
| description | The paper focuses on revealing insider information leaks of financial markets during investment consulting. An original dataset was created, containing the records of the conversations between consultants and clients, presented in the form of dialogs in text format. The applicability of machine learning methods for automating the detection of leaks arising in a conversation between a consultant and a client has been studied. The authors examined the applicability of the following supervised machine learning methods for constructing and training a classifier: probabilistic (Naïve Bayes
classifier), metric (k-nearest neighbors algorithm), logical (random forest), linear (support vector machine), and methods based on artificial neural networks. The paper considers various approaches to the construction of a natural language text model, such as tokenization (bag of words, word n-grams: bigrams and trigrams) and vectorization (one-hot encoding).
The proposed algorithm for detecting financial markets insider information leaks is based on the use of support vector machine (SVM) and tokenization by bigrams. The obtained results demonstrate that SVM and bigram tokenization
provide the highest leakage detection accuracy. The research results can be used in cybersecurity tools development, as well as for the further elaboration of natural language processing methods dealing with information security problems. |
| first_indexed | 2024-12-19T10:48:10Z |
| format | Article |
| id | doaj.art-d1645f278e1c408fa059a56948fe044e |
| institution | Directory Open Access Journal |
| issn | 2226-1494 2500-0373 |
| language | English |
| last_indexed | 2024-12-19T10:48:10Z |
| publishDate | 2021-06-01 |
| publisher | Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University) |
| record_format | Article |
| series | Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki |
| spelling | doaj.art-d1645f278e1c408fa059a56948fe044e2022-12-21T20:25:09ZengSaint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki2226-14942500-03732021-06-0121339440010.17586/2226-1494-2021-21-3-394-400An algorithm for detecting leaks of insider information of financial markets in investment consultingAlisa A. Vorobeva0https://orcid.org/0000-0001-6691-6167Vladislav V. Gerasimov1https://orcid.org/0000-0001-8099-2414Yulia V. Li2https://orcid.org/0000-0003-3280-8197PhD, Associate Professor, ITMO University, Saint Petersburg, 197101, Russian FederationEngineer, ITMO University, Saint Petersburg, 197101, Russian FederationEngineer, ITMO University, Saint Petersburg, 197101, Russian FederationThe paper focuses on revealing insider information leaks of financial markets during investment consulting. An original dataset was created, containing the records of the conversations between consultants and clients, presented in the form of dialogs in text format. The applicability of machine learning methods for automating the detection of leaks arising in a conversation between a consultant and a client has been studied. The authors examined the applicability of the following supervised machine learning methods for constructing and training a classifier: probabilistic (Naïve Bayes classifier), metric (k-nearest neighbors algorithm), logical (random forest), linear (support vector machine), and methods based on artificial neural networks. The paper considers various approaches to the construction of a natural language text model, such as tokenization (bag of words, word n-grams: bigrams and trigrams) and vectorization (one-hot encoding). The proposed algorithm for detecting financial markets insider information leaks is based on the use of support vector machine (SVM) and tokenization by bigrams. The obtained results demonstrate that SVM and bigram tokenization provide the highest leakage detection accuracy. The research results can be used in cybersecurity tools development, as well as for the further elaboration of natural language processing methods dealing with information security problems.https://ntv.ifmo.ru/file/article/20507.pdfnatural language processingmachine learningneural networkscompliance risksinsider information |
| spellingShingle | Alisa A. Vorobeva Vladislav V. Gerasimov Yulia V. Li An algorithm for detecting leaks of insider information of financial markets in investment consulting Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki natural language processing machine learning neural networks compliance risks insider information |
| title | An algorithm for detecting leaks of insider information of financial markets in investment consulting |
| title_full | An algorithm for detecting leaks of insider information of financial markets in investment consulting |
| title_fullStr | An algorithm for detecting leaks of insider information of financial markets in investment consulting |
| title_full_unstemmed | An algorithm for detecting leaks of insider information of financial markets in investment consulting |
| title_short | An algorithm for detecting leaks of insider information of financial markets in investment consulting |
| title_sort | algorithm for detecting leaks of insider information of financial markets in investment consulting |
| topic | natural language processing machine learning neural networks compliance risks insider information |
| url | https://ntv.ifmo.ru/file/article/20507.pdf |
| work_keys_str_mv | AT alisaavorobeva analgorithmfordetectingleaksofinsiderinformationoffinancialmarketsininvestmentconsulting AT vladislavvgerasimov analgorithmfordetectingleaksofinsiderinformationoffinancialmarketsininvestmentconsulting AT yuliavli analgorithmfordetectingleaksofinsiderinformationoffinancialmarketsininvestmentconsulting AT alisaavorobeva algorithmfordetectingleaksofinsiderinformationoffinancialmarketsininvestmentconsulting AT vladislavvgerasimov algorithmfordetectingleaksofinsiderinformationoffinancialmarketsininvestmentconsulting AT yuliavli algorithmfordetectingleaksofinsiderinformationoffinancialmarketsininvestmentconsulting |