Papers - Takahashi, Hiroshi

Construction of news article evaluation system using language generation model

Nishi Y, Suge A, Takahashi H

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies)  186   313 - 320 2020

Joint Work,  ISSN  9789811557637

　View Summary

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2020. This study constructed a news article evaluation system that utilizes a language generation model to analyze financial markets. This system enables us to analyze the effect of news articles distributed in financial markets on the stock price of a company. We added the generated news articles as data for analysis through GPT-2 and verified the accuracy of the constructed system. As a result of empirical analyses, we confirmed that the accuracy of the model with the generated news articles improved. More detailed analyses are planned for the future.

• Access to Document (DOI)

Constructing a valuation system through patent document analysis

Fujiwara S, Matsumoto Y, Suge A, Takahashi H

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies)  186   321 - 330 2020

Joint Work,  ISSN  9789811557637

　View Summary

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2020. In the Initial Public Offering (IPO), many prior studies have discussed underpricing, where the public price is below the initial price. In this study, we propose a novel system for company valuation employing patent document analysis. Specifically, we first extracted patent document data published from 2001 to 2015 and vectorize them through Sparse Composite Document Vectors (SCDV). Similar companies in this study refer to the same industry, products handled, and business scale. However, there are various theories on what should be treated as similar companies. Next, the center of gravity is calculated from the vectorized patent document, and comparable companies are selected by expressing the distance between companies. A detailed analysis is planned for further study.

• Access to Document (DOI)

A formal, descriptive model for the business case of managerial decision-making

Kunigami M, Kikuchi T, Takahashi H, Terano T

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies)  186   355 - 365 2020

Joint Work,  ISSN  9789811557637

　View Summary

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2020. This paper proposes a formal descriptive model of organizational decision-making called the Managerial Decision-Making Description Model (MDDM). This model introduces visual representations to describe managerial decisions that redefine relationships between their objectives and resources. The MDDM describes various business cases and enables us to compare these decision-making processes. This paper presents the MDDM’s methodologies and describes, and compares, the decision diagrams extracted from actual business cases and organizational agent-based simulation (ABS) logs as virtual business cases.

• Access to Document (DOI)

飲食業界における労働市場-エージェントベースモデルによる分析

高橋 大志

第 82 回全国大会講演論文集  2020

Single Work

機械学習を用いた上場企業の労働分配率および平均給与の説明可能性に関する検討

高橋 大志

第 82 回全国大会講演論文集  2020

Single Work

大規模言語生成モデルによるニュース生成を用いたニュース評価モデルの構築

高橋 大志

第 82 回全国大会講演論文集  2020

Single Work

大規模ニューステキストを用いたナレッジネットワーク構築

高橋 大志

第 82 回全国大会講演論文集  2020

Single Work

Language generation by GPT-2 and text analysis of economic fluctuations forecast through LSTM

NISHI Yoshihiro, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 08 2019.10

　View Summary

<p>Economic fluctuations have a great impact on business management. In this study, we explorethe method to improve the accuracy of the prediction model of the economic fluctuation using the economicwatcher survey reported by the Cabinet Office, the Government of Japan. The economic watcher surveyincludes interviewees' answers in the text and the economic business trend tag for each answer. Theclassification was attempted through the LSTM model. As a result of the analysis, we found the predictionaccuracy could be improved by using texts generated by GPT-2. Further examination of the classificationmodel will be planned.</p>

• Access to Document (DOI)
• CiNii Research

Analyzing the factors that determine the labor share of listed companies in Japan

MATSUMOTO Akihiro, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 10 2019.10

　View Summary

<p>In recent years, the labor share has been declining in Japan, causing a hollowing out of the realeconomy. The purpose of this study is to clarify the relationship between labor shares and payout ratios. Weplan to examine whether the relationship between the labor share and payout ratio can be demonstrated byusing nonlinear analysis such as machine learning methods.</p>

• Access to Document (DOI)
• CiNii Research

Testing an agent-based model of the labor market: food-service industry and non-regular employment

SEO Hwangsaeng, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 11 2019.10

　View Summary

<p>Japan is suffering from a serious labor shortage due to the economic expansion anddemographic change in the non-regular employment of the labor market. In particular, non-regularemployees account for 80% of employees in the food-service industry is a critical issue. The purpose ofthe present study is to analyze a mechanism of job matching focusing on non-regular employmentthrough agent-based modeling.</p>

• Access to Document (DOI)
• CiNii Research

Personal Life Planning Simulation using Agent-based Model

KIKUCHI Takamasa, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 07 2019.10

　View Summary

<p>本邦では，社会の高齢化・長寿化が進展にするにつれ，資産の形成や取崩し，承継領域に係る論点について関心が高まっている．しかし、退職後の資産取崩しを前提とした資産形成・運用に関して，必ずしも理論や基本的な考え方が整備されているとは言い難い．そこで本稿では，著者らが提案している資産形成や取り崩しに係るエージェントモデルを拡張する．外生的に与える各種シナリオにおいて，個人の意思決定が資産枯渇に与える影響を分析する．</p>

• Access to Document (DOI)
• CiNii Research

Causal network construction using large-scale news text

ZHANG Ying, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 06 2019.10

　View Summary

<p>Investors refer to a variety of information when making investment decisions: The amount ofavailable information is increasing day by day due to the spread of the Internet, and it is difficult to graspall information such as causality. In this study, we attempt to construct an understanding support model thatcan visualize news through Attention-Based Bi-LSTM model using Reuters News as an analysis target.</p>

• Access to Document (DOI)
• CiNii Research

Identifying equity influencers on Twitter

KATAYAMA Shinya, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 03 2019.10

　View Summary

<p>The influence of SNS or the impact of influencers is expanding on our society. This studyanalyzes the relationships between tweets and news text of influencers in the Japanese stock market. Roleof social media in the financial market is also examined by extracting the data related to the tweetsinformation.</p>

• Access to Document (DOI)
• CiNii Research

Selection of comparable company using patent document vectors

FUJIWARA Shohei, MATSUMOTO Yusuke, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-013 ) 02 2019.10

　View Summary

<p>The number of M&A and IPO implementations has been increasing year by year, and the capitalmarket has become more active. Along with those situations, when a manager examines M&A and IPO, thecorporate value calculation is an important decision-making index. This paper attempts to improve theaccuracy of selecting similar companies in the process of comparable company analysis and DCF methods.Specifically, Sparse Composite Document Vector (SCDV) was created based on published patentdocuments. Next, similar companies were selected by calculating the distance between companies from thedocument vector made by SCDV.</p>

• Access to Document (DOI)
• CiNii Research

Estimating the Synergy Effect and the Corporate Activity

MATSUMOTO Yusuke, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-012 ) 02 2019.03

　View Summary

<p>Business behavior changes as a company grows. One of the behaviors which a company changeas it grows is diversification. While diversification is an advantage for a company, it can also be adisadvantage. This study analyzes how corporate diversification affects enterprise value, especially thesynergy effect between business units. We plan to analyze in detail in the future.</p>

• Access to Document (DOI)
• CiNii Research

Customer Behavior Research for Customer Relationship Management in High-end Sushi Shop

SHIN Kiyotaka, QU Xueyin, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-012 ) 04 2019.03

　View Summary

<p>This research aims to construct a system that leads the improvement of restaurantcompetitiveness by using POS data. The labor productivity of the food-service industry in Japan is very low, haslong suffered from lack of personnel and continues to struggle maintaining talent. In order to solve this problem,we analyzed the POS data and customer data of a high-end sushi shop with a high customer unit price andhypothesized that several factors may affect customers' loyalty through CRM (Customer RelationshipManagement). Besides, this study contributes to identify loyal customers through purchasing behavior.</p>

• Access to Document (DOI)
• CiNii Research

Proposal of An Agent Model for Personal Life Planning

KIKUCHI Takamasa, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-012 ) 09 2019.03

　View Summary

<p>本邦では，社会の高齢化・長寿化が進展にするにつれ，資産寿命を延伸することの重要性が高まってきている．しかし，資産の形成や取り崩し，承継領域に係る論点については，必ずしも議論が深まっているとは言えない状況にある．特に，退職後の資産取り崩しを前提とした資産形成・運用に関して，理論や基本的な考え方の整備，個人の特性に応じた助言・コンサルティングの枠組みの高度化など，様々な検討を進める必要がある．そこで本稿では，個人の特性やライフイベントを勘案したキャッシュフローマネジメントに係るシミュレーションの枠組みを提案する．資産形成や取り崩しに係る個人の意思決定を取り扱うエージェントモデルを構築し，外生的に与える各種シナリオにおいて，資産枯渇に与える影響を分析する．</p>

• Access to Document (DOI)
• CiNii Research

Analyzing the Relationship between Weekly Effect in Chinese Stock Markets and News Articles using a Financial Polarity Dictionary

QU Xueyin, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-012 ) 07 2019.03

　View Summary

<p>One of the anomalies in the financial markets is the week day effect. Although reports of theweekday effect have been conducted in China's stock market, the cause of the effect is still indecisive.The purpose of this study is to clarify the relevance of the weekday effect and news about the Chinesestock market. In this paper, we examined the polarity of the news and try to explain its relevance with theweekday effect.</p>

• Access to Document (DOI)
• CiNii Research

LSTM Model for Explaining the Association between News Data and Stock Price Fluctuations

TAKAYAMA Liliko, SUGE Aiko, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-011 ) 11 2019.01

　View Summary

<p>Research on the association between stock price fluctuations and news data is popular andnumerous studies have been conducted. This paper aims to improve the accuracy of text classification intothree categories (negative, neutral, or positive) by employing high frequency data and LSTM model. Oneof the novelties of our paper is to use several types of news articles in the analysis.</p>

• Access to Document (DOI)
• CiNii Research

Research of High-end Barbecue Beef Restaurants Based on Consumer Purchase Data by Machine Learning and FinTech Application

SHIN Kiyotaka, SUGE Aiko, YAMASHITA Yasuo, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence)  2019 ( BI-011 ) 06 2019.01

　View Summary

<p>The purpose of this research is to realize service that leads to improvement of management ability byusing POS data of restaurant service industry. The labor productivity of the food-service industry in Japan isvery low, has long suffered from lack of personnel and continues to struggle maintaining talent. Therefore, as ameasure to solve this problem, we analyzed the POS data of a High-end barbeque shop with a high customerunit price, making a hypothesis that the improvement per customer price per customer is effective.And Weclarify the characteristics of customer's purchasing behavior and examine the sales policy including FinTechapplication using machine learning.</p>

• Access to Document (DOI)
• CiNii Research