By the same authors

Ontology Graph Embeddings and ILP for Financial Forecasting

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

Ontology Graph Embeddings and ILP for Financial Forecasting. / Erten, Can; Kazakov, Dimitar Lubomirov.

Inductive Logic Programming, Proceedings of the 30th International Conference. Springer, 2021. (LNAI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Erten, C & Kazakov, DL 2021, Ontology Graph Embeddings and ILP for Financial Forecasting. in Inductive Logic Programming, Proceedings of the 30th International Conference. LNAI, Springer, 30th International Conference on Inductive Logic Programming, Athens, Greece, 25/10/21.

APA

Erten, C., & Kazakov, D. L. (Accepted/In press). Ontology Graph Embeddings and ILP for Financial Forecasting. In Inductive Logic Programming, Proceedings of the 30th International Conference (LNAI). Springer.

Vancouver

Erten C, Kazakov DL. Ontology Graph Embeddings and ILP for Financial Forecasting. In Inductive Logic Programming, Proceedings of the 30th International Conference. Springer. 2021. (LNAI).

Author

Erten, Can ; Kazakov, Dimitar Lubomirov. / Ontology Graph Embeddings and ILP for Financial Forecasting. Inductive Logic Programming, Proceedings of the 30th International Conference. Springer, 2021. (LNAI).

Bibtex - Download

@inproceedings{8b579807395b44cbb6b0e389e56d3497,
title = "Ontology Graph Embeddings and ILP for Financial Forecasting",
abstract = " There is a history of hybrid machine learning approaches where the result of an unsupervised learning algorithm is used to provide data annotation from which ILP can learn in the usual supervised manner. Here we consider the task of predicting the property of cointegration between the time series of stock price of two companies, which can be used to implement a robust pair-trading strategy that can remain profitable regardless of the overall direction in which the market evolves. We start with an original FinTech ontology of relations between companies and their managers, which we have previously extracted from SEC reports, quarterly filings that are mandatory for all US companies. When combined with stock price time series, these relations have been shown to help find pairs of companies suitable to pair trading. Here we use node2vec embeddings to produce clusters of companies and managers, which are then used as background predicates in addition to the relations linking companies and staff present in the ontology, and the values of the target predicate for a given time period. Progol is used to learn from this mixture of predicates combining numerical with structural relations of the entities represented in the data set to reveal rules with and predictive power.",
keywords = "Ontology, financial forecasting, SEC financial reports, Inductive Logic Programming (ILP), unsupervised learning, graph embedding",
author = "Can Erten and Kazakov, {Dimitar Lubomirov}",
note = "{\textcopyright} 2021 Springer Nature Switzerland AG. ; 30th International Conference on Inductive Logic Programming : ILP2020-21@IJCLR ; Conference date: 25-10-2021 Through 27-10-2021",
year = "2021",
month = sep,
day = "20",
language = "English",
series = "LNAI",
publisher = "Springer",
booktitle = "Inductive Logic Programming, Proceedings of the 30th International Conference",
url = "http://lr2020.iit.demokritos.gr/ilp/index.html",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - Ontology Graph Embeddings and ILP for Financial Forecasting

AU - Erten, Can

AU - Kazakov, Dimitar Lubomirov

N1 - Conference code: 30

PY - 2021/9/20

Y1 - 2021/9/20

N2 - There is a history of hybrid machine learning approaches where the result of an unsupervised learning algorithm is used to provide data annotation from which ILP can learn in the usual supervised manner. Here we consider the task of predicting the property of cointegration between the time series of stock price of two companies, which can be used to implement a robust pair-trading strategy that can remain profitable regardless of the overall direction in which the market evolves. We start with an original FinTech ontology of relations between companies and their managers, which we have previously extracted from SEC reports, quarterly filings that are mandatory for all US companies. When combined with stock price time series, these relations have been shown to help find pairs of companies suitable to pair trading. Here we use node2vec embeddings to produce clusters of companies and managers, which are then used as background predicates in addition to the relations linking companies and staff present in the ontology, and the values of the target predicate for a given time period. Progol is used to learn from this mixture of predicates combining numerical with structural relations of the entities represented in the data set to reveal rules with and predictive power.

AB - There is a history of hybrid machine learning approaches where the result of an unsupervised learning algorithm is used to provide data annotation from which ILP can learn in the usual supervised manner. Here we consider the task of predicting the property of cointegration between the time series of stock price of two companies, which can be used to implement a robust pair-trading strategy that can remain profitable regardless of the overall direction in which the market evolves. We start with an original FinTech ontology of relations between companies and their managers, which we have previously extracted from SEC reports, quarterly filings that are mandatory for all US companies. When combined with stock price time series, these relations have been shown to help find pairs of companies suitable to pair trading. Here we use node2vec embeddings to produce clusters of companies and managers, which are then used as background predicates in addition to the relations linking companies and staff present in the ontology, and the values of the target predicate for a given time period. Progol is used to learn from this mixture of predicates combining numerical with structural relations of the entities represented in the data set to reveal rules with and predictive power.

KW - Ontology

KW - financial forecasting

KW - SEC financial reports

KW - Inductive Logic Programming (ILP)

KW - unsupervised learning

KW - graph embedding

M3 - Conference contribution

T3 - LNAI

BT - Inductive Logic Programming, Proceedings of the 30th International Conference

PB - Springer

T2 - 30th International Conference on Inductive Logic Programming

Y2 - 25 October 2021 through 27 October 2021

ER -