The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
O uso de relatórios em ação cresceu significativamente nas últimas décadas, à medida que os dados foram digitalizados. No entanto, os métodos estatísticos tradicionais já não funcionam devido à expansão incontrolável e à complexidade dos dados brutos. Portanto, é crucial limpar e analisar dados financeiros usando métodos modernos de aprendizado de máquina. Neste estudo, os relatórios trimestrais (ou seja, registros do 10T) de empresas de capital aberto nos Estados Unidos foram analisados utilizando métodos de mineração de dados. O estudo utilizou 8905 relatórios trimestrais de empresas de 2019 a 2022. A abordagem proposta consiste em duas fases com uma combinação de três métodos diferentes de aprendizado de máquina. Os dois primeiros métodos foram utilizados para gerar um conjunto de dados a partir dos arquivamentos 10Q com extração de novos recursos, e o último método foi utilizado para o problema de classificação. O método Doc2Vec no framework Gensim foi usado para gerar vetores a partir de tags textuais em arquivamentos 10Q. Os vetores gerados foram agrupados utilizando o algoritmo K-means para combinar as tags de acordo com sua semântica. Desta forma, 94000 tags representando diferentes itens financeiros foram reduzidos para 20000 clusters compostos por essas tags, tornando a análise mais eficiente e gerenciável. O conjunto de dados foi criado com os valores correspondentes às tags nos clusters. Além disso, a métrica PriceRank foi adicionada ao conjunto de dados como um rótulo de classe que indica a força dos preços das empresas para o próximo trimestre financeiro. Assim, pretende-se determinar o efeito dos relatórios trimestrais de uma empresa no preço de mercado da empresa para o próximo período. Por fim, um modelo de Rede Neural Convolucional foi utilizado para o problema de classificação. Para avaliar os resultados, todas as etapas do método híbrido proposto foram comparadas com outras técnicas de aprendizado de máquina. Esta nova abordagem poderá ajudar os investidores a examinarem as empresas colectivamente e a inferirem conhecimentos novos e significativos. O método proposto foi comparado com diferentes abordagens para criação de conjuntos de dados por meio da extração de novos recursos e tarefas de classificação e, eventualmente, testado com diferentes métricas. A abordagem proposta teve um desempenho comparativamente melhor do que outros métodos de aprendizado de máquina para prever a força futura dos preços com base em relatórios anteriores com uma precisão de 84% no conjunto de dados de registros do 10T criado.
Mustafa Sami KACAR
Konya Technical Univ.
Semih YUMUSAK
KTO Karatay Univ.
Halife KODAZ
Konya Technical Univ.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Mustafa Sami KACAR, Semih YUMUSAK, Halife KODAZ, "Price Rank Prediction of a Company by Utilizing Data Mining Methods on Financial Disclosures" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 9, pp. 1461-1471, September 2023, doi: 10.1587/transinf.2022OFP0002.
Abstract: The use of reports in action has grown significantly in recent decades as data has become digitized. However, traditional statistical methods no longer work due to the uncontrollable expansion and complexity of raw data. Therefore, it is crucial to clean and analyze financial data using modern machine learning methods. In this study, the quarterly reports (i.e. 10Q filings) of publicly traded companies in the United States were analyzed by utilizing data mining methods. The study used 8905 quarterly reports of companies from 2019 to 2022. The proposed approach consists of two phases with a combination of three different machine learning methods. The first two methods were used to generate a dataset from the 10Q filings with extracting new features, and the last method was used for the classification problem. Doc2Vec method in Gensim framework was used to generate vectors from textual tags in 10Q filings. The generated vectors were clustered using the K-means algorithm to combine the tags according to their semantics. By this way, 94000 tags representing different financial items were reduced to 20000 clusters consisting of these tags, making the analysis more efficient and manageable. The dataset was created with the values corresponding to the tags in the clusters. In addition, PriceRank metric was added to the dataset as a class label indicating the price strength of the companies for the next financial quarter. Thus, it is aimed to determine the effect of a company's quarterly reports on the market price of the company for the next period. Finally, a Convolutional Neural Network model was utilized for the classification problem. To evaluate the results, all stages of the proposed hybrid method were compared with other machine learning techniques. This novel approach could assist investors in examining companies collectively and inferring new, significant insights. The proposed method was compared with different approaches for creating datasets by extracting new features and classification tasks, then eventually tested with different metrics. The proposed approach performed comparatively better than the other machine learning methods to predict future price strength based on past reports with an accuracy of 84% on the created 10Q filings dataset.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022OFP0002/_p
Copiar
@ARTICLE{e106-d_9_1461,
author={Mustafa Sami KACAR, Semih YUMUSAK, Halife KODAZ, },
journal={IEICE TRANSACTIONS on Information},
title={Price Rank Prediction of a Company by Utilizing Data Mining Methods on Financial Disclosures},
year={2023},
volume={E106-D},
number={9},
pages={1461-1471},
abstract={The use of reports in action has grown significantly in recent decades as data has become digitized. However, traditional statistical methods no longer work due to the uncontrollable expansion and complexity of raw data. Therefore, it is crucial to clean and analyze financial data using modern machine learning methods. In this study, the quarterly reports (i.e. 10Q filings) of publicly traded companies in the United States were analyzed by utilizing data mining methods. The study used 8905 quarterly reports of companies from 2019 to 2022. The proposed approach consists of two phases with a combination of three different machine learning methods. The first two methods were used to generate a dataset from the 10Q filings with extracting new features, and the last method was used for the classification problem. Doc2Vec method in Gensim framework was used to generate vectors from textual tags in 10Q filings. The generated vectors were clustered using the K-means algorithm to combine the tags according to their semantics. By this way, 94000 tags representing different financial items were reduced to 20000 clusters consisting of these tags, making the analysis more efficient and manageable. The dataset was created with the values corresponding to the tags in the clusters. In addition, PriceRank metric was added to the dataset as a class label indicating the price strength of the companies for the next financial quarter. Thus, it is aimed to determine the effect of a company's quarterly reports on the market price of the company for the next period. Finally, a Convolutional Neural Network model was utilized for the classification problem. To evaluate the results, all stages of the proposed hybrid method were compared with other machine learning techniques. This novel approach could assist investors in examining companies collectively and inferring new, significant insights. The proposed method was compared with different approaches for creating datasets by extracting new features and classification tasks, then eventually tested with different metrics. The proposed approach performed comparatively better than the other machine learning methods to predict future price strength based on past reports with an accuracy of 84% on the created 10Q filings dataset.},
keywords={},
doi={10.1587/transinf.2022OFP0002},
ISSN={1745-1361},
month={September},}
Copiar
TY - JOUR
TI - Price Rank Prediction of a Company by Utilizing Data Mining Methods on Financial Disclosures
T2 - IEICE TRANSACTIONS on Information
SP - 1461
EP - 1471
AU - Mustafa Sami KACAR
AU - Semih YUMUSAK
AU - Halife KODAZ
PY - 2023
DO - 10.1587/transinf.2022OFP0002
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2023
AB - The use of reports in action has grown significantly in recent decades as data has become digitized. However, traditional statistical methods no longer work due to the uncontrollable expansion and complexity of raw data. Therefore, it is crucial to clean and analyze financial data using modern machine learning methods. In this study, the quarterly reports (i.e. 10Q filings) of publicly traded companies in the United States were analyzed by utilizing data mining methods. The study used 8905 quarterly reports of companies from 2019 to 2022. The proposed approach consists of two phases with a combination of three different machine learning methods. The first two methods were used to generate a dataset from the 10Q filings with extracting new features, and the last method was used for the classification problem. Doc2Vec method in Gensim framework was used to generate vectors from textual tags in 10Q filings. The generated vectors were clustered using the K-means algorithm to combine the tags according to their semantics. By this way, 94000 tags representing different financial items were reduced to 20000 clusters consisting of these tags, making the analysis more efficient and manageable. The dataset was created with the values corresponding to the tags in the clusters. In addition, PriceRank metric was added to the dataset as a class label indicating the price strength of the companies for the next financial quarter. Thus, it is aimed to determine the effect of a company's quarterly reports on the market price of the company for the next period. Finally, a Convolutional Neural Network model was utilized for the classification problem. To evaluate the results, all stages of the proposed hybrid method were compared with other machine learning techniques. This novel approach could assist investors in examining companies collectively and inferring new, significant insights. The proposed method was compared with different approaches for creating datasets by extracting new features and classification tasks, then eventually tested with different metrics. The proposed approach performed comparatively better than the other machine learning methods to predict future price strength based on past reports with an accuracy of 84% on the created 10Q filings dataset.
ER -