Details of a Researcher - Takahashi, Hiroshi

Papers - Takahashi, Hiroshi

Division display 41 - 60 of about 283 ／ All the affair displays >>

Empirical Analysis of the Impact of Mergers and Acquisitions on Innovation Activities Through Clustering

Tamagawa N, Takahashi H

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)) 13856 LNAI 365 - 379 2023

ISSN 03029743

Using large-scale patent data, we analyze the impact of mergers and acquisitions (M&A) on innovation activities in companies. The analysis classifies M&A according to the trend of innovation output through their index clustering. In addition, we measure the technological distances among companies using patent document data, then apply the distances to the evaluation of M&A. As a result, we find two types of clusters: 1) one with the tendency to increase the postmerger innovation outputs; 2) the other with a trend to decrease innovation output around M&A deals. The former cluster tends to have a smaller innovation output scale around deals than the other clusters. Furthermore, compared to the other groups, the group whose postmerger innovation output decreases has the largest increase in stock price return after the deal announcement. This study’s major findings indicate that the combination of unstructured text data and machine learning methods applies to M&A and innovation.

Access to Document (DOI)

Bitcoin Fraudulent Transaction Detection Vulnerability

Ehara T, Takahashi H

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies) 354 SIST 183 - 193 2023

ISSN 21903018

Bitcoin is expected to play a vital role in the web3 and is attracting attention. Still, it is notorious as a hotbed for various fraudulent transactions, which are on the rise. One of the characteristics of Bitcoin is its anonymity. Bitcoin allows tracing transactions, but the owner cannot be identified. In addition, bitcoin is considered more difficult to trace criminals than bank transactions because of the unlimited number of addresses that can be created. This study shows that it is possible to detect criminal transactions with known transaction patterns using supervised learning. Still, it is difficult to detect criminal transactions with unknown transaction patterns using supervised learning. We also show that even with anomaly detection methods, illicit transactions are difficult to detect. From the above, we believe that Bitcoin is vulnerable to illicit transactions.

Access to Document (DOI)

Analysis of the relationship between technological diversification and enterprise value using patent data

Matsumoto Y, Suge A, Takahashi H

Information Technology and Management (Information Technology and Management) 2023

ISSN 1385951X

As natural language processing technology advances, its application to finance is growing. We focus on technological diversification resulting from R&D activities, one of the concerns for firms requiring efficient business management. We analyze the relationship between technological diversification and enterprise value using text information of patents and examine text information's usefulness in corporate finance. Specifically, we created a firm's technological diversification index from the text information in the patents and the firm's excess value from its financial data and analyzed the relationship between them. We analyzed Japanese firms that have been listed on the First Section of the Tokyo Stock Exchange and have applied for patents during the period 2002–2015. As a result, we found that (1) the spread of technology in multiple ways could impair enterprise value, and (2) the use of text information can be more valuable than the use of specific standards indexes such as the International Patent Classification (IPC). This study shows interesting results on the relationship between technological diversification and enterprise value and implies the applicability of unstructured data to finance research. A detailed analysis is for further study.

Access to Document (DOI)

Evaluation of Sponsor Matching in eSports Using Attribute Estimation Method of Twitter Sentences

SAITO Tatsuki, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-021 ) 09 2022.09

In this research, we will create a model that estimates the attributes of followers based on the information of utterances on SNS as clues about what kind of customers support e-sports teams and their sponsor companies, and compare the attributes. The purpose is to clarify the sponsor's integrity by The survey targets e-sports teams that operate Twitter accounts and their sponsor companies. Using the information posted on Twitter, we will compare the attributes of followers between target accounts and analyze the trend of followers. Based on the analysis results, we propose indicators for optimal matching of teams and sponsors.

Improving the Accuracy of Classifying Illegal and Legal Transactions in Bitcoin Transaction Data

EHARA Takashi, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-021 ) 11 2022.09

Crypto asset (Virtual currency) transactions have long been in the limelight. Crypto assets use cryptography to achieve decentralized management by participants on the Internet. Since they can be used by an unspecified number of people and there is no intermediary in the transaction, they are often misused for money laundering and terrorist financing. In this study, we adapt various machine learning methods to ELLIPTIC DATA, which is BITCOIN transaction data used in many previous studies. As a first step, we examine parameter optimization of random forests, feature refinement, sampling methods, and so on...

The Identification of Product-Harm Crisis and its Effects on Enterprise Value via News Dataset

ZHAO Huimei, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-021 ) 02 2022.09

Product-harm crises occurred worldwide, and the effects are reflected not only on customers but also on enterprise value. In our study, we analyzed specific sites which publish scandals in Japanese and identified the keywords of product-harm crises using co-occurrence analysis. Based on the results, we can analyze news data to understand the effects of product-harm crises via event study. We will also analyze the news in different countries by natural language processing and aim at finding a difference in product-harm crisis management.

Proposal of organizational structure and performance evaluation method for Japanese IT companies using business cases

OKAZAKI Haruna, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-020 ) 10 2022.03

Business cases include a lot of information on corporate activities, including informa- tion contained in securities reports. In this study, we analyze the relationship between the orga- nizational structure of Japanese IT companies and financial performance using business cases. .. The organizational structure of Japanese IT companies can be broadly divided into pyramid-type organizations and flat-type organizations. Recently, IT companies have been called for flatten- ing, and many companies are moving to flat-type organizations. However, Japan is originally a pyramid-type organization, and it is thought that there are not many companies that are emerging as flat-type organizations. The purpose of this paper is to show a concept for exploring one of the factors.

A Study on Improving the Accuracy of Classification Models through Property Image Generation

MORIMATSU Kazuya, JIANSEN Zhao, TAMAGAWA Nozomi, OKAZAKI Haruna, YOON Sungjae, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-020 ) 03 2022.03

In the case of business development by learning information about properties, such as recognizing scratches on the floor or wall, it is difficult to find a certain pattern in the teacher data, which makes it difficult to prepare teacher data of sufficient quality and quantity. Therefore, in this study, we investigate how to improve the accuracy of the classification model by generating pseudo-images from a small number of samples to improve the quality and quantity of the teacher data.

A Study on the Impact of Information Security Incidents on Stocks of the U.S. Market

ZHAO Jiansen, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-020 ) 15 2022.03

In this study, we analyzed the impact of news related to information security incidentson the stock prices of companies listed on the NASDAQ market in the United States. Theanalysis data was based on news that was announced to the market participants by ReutersNews. For the analysis of news related to information security incidents, we used the BERTmodel of text analysis. In this study, we used the event study method to compare and analyzethe impact on stock prices in multiple periods.

Empirical Analysis of the Relationship between M&A and Innovation Activities for the Pharmaceutical and Materials Industries

TAMAGAWA Nozomi, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2022 ( BI-020 ) 07 2022.03

We conduct an empirical analysis of the relationship between innovation activity and mergers and acquisitions (M&A) by public companies in the pharmaceutical and materials industries in Japan, the U.S., and Germany. The analysis uses large-scale patent data to measure indicators of innovation activity. In addition, we attempt to quantitatively measure the technological distance between firms by analyzing patent document data using natural language processing, and apply it to the evaluation of M&A. As a result, we find a trend in the relationship between the technological distance of the acquirer and the target firm and the post-merger innovation activity. One of the novelties of this study is the application of unstructured data and information technology methods to the research field of M&A and innovation activities. This research contributes to the quantitative evaluation of intangible assets in companies.

Analysis of News and Stock Market with FinBERT models

Yoon Sungjae, Takahashi Hiroshi

Abstracts of Annual Conference of Japan Society for Management Information (THE JAPAN SOCIETY FOR MANAGEMENT INFORMATION (JASMIN)) 202111 ( 0 ) 227 - 230 2022.01

In this research, we analyze the relationship between news announcements and trading volume in the stock market using high-frequency trading data and natural language processing technology. In the evaluation of the news, we applied the FinBERT models, which are analysis methods specialized in the financial field, and targeted the Korean and Japanese stock markets. As a result, we found that the trading volume tends to increase due to the ‘Alert' news with high immediacy. and that some companies with relatively small market capitalization have a strong connection to news and trading volume, while news about companies with large market capitalization have a limited impact. These are interesting results that contribute to the clarification of the price formation mechanism in the market.

A Customer Experience Mapping for Knowledge Extraction from Social Simulation Results

Takamasa Kikuchi, Masaaki Kunigami, Hiroshi Takahashi, Takao Terano

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies) 306 159 - 168 2022

ISSN 21903018

Various methodologies have been proposed for extracting knowledge from the output results and logs of social simulations. Such analysis methods are important not only for model developers and analysts, but also for sharing knowledge and communication among other stakeholders. In this paper, we attempt to apply the System Experience Boundary Map, which is one of the experience mapping methods used in the field of design thinking, for the analysis and formal description of the results of social simulation.

Access to Document (DOI)

Nowcasting of Corporate Research and Development trends through news article analysis by BERTopic: The case of Japanese electric company

Okazaki H, Takahashi H

International Conference on Electrical, Computer, Communications and Mechatronics Engineering, ICECCME 2022 (International Conference on Electrical, Computer, Communications and Mechatronics Engineering, ICECCME 2022) 2022

Various means exist for obtaining information about a company. However, many of them are not timely and, from the investor' s point of view, do not have enough information. In particular, it is even more difficult to find out trends of diversified companies with multiple segments. Therefore, this study aims to extract timely information on the trends of each segment of a company from news data using BERTopic. The analysis targets news headlines of diversified electronics firms listed on the Japanese stock market. The sample period was 24 years, from 1996 to 2019, and the number of news items for analysis was 26,058. As a result of the analysis, we found that (1) BERTopic can classify the target news into 46 topics, (2) it is possible to identify company segments and extract trends in company activities from the classified topics, and (3) it is also possible to visualize the time-series transition of topics related to each segment. (4) The results obtained from the analysis were used to determine the value of the company's investment in the market. In addition, the results obtained from the analysis were consistent with the descriptions in the annual reports. These results indicate the possibility of obtaining highly immediate information on corporate trends, such as R&D, through the analysis of news headlines via BERTopic.

Access to Document (DOI)

How Can We Make the Best Use of Intellectual Capital? Building an Analysis System Through Agent-Based Models

Morimatsu K, Takahashi H

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies) 306 147 - 157 2022

ISSN 21903018

It is important to utilize intellectual capital to overcome the severe business environment. However, it is difficult to grasp the causal relationship between how to utilize them and how to improve business performance. In this study, we have developed a system that can elucidate the strategies of excellent companies through the analysis of their behavior in a complex business environment using an agent-based model. The system consists of three layers of business environment: external layer, decision-making layer, and choice of action layer. In addition, there are three types of investment targets: human capital, structural capital, and relational capital. Through this system, we can optimize investment strategies according to market conditions and recommend strategies that consider the behavior of competitors, which is the strength of agent-based models. In this paper, the construction of the system and a case study of analysis are presented. One of the novelties of this research is that we have developed a structure that allows us to consider the utilization of intangible assets from a quantitative perspective.

Access to Document (DOI)

System for Analyzing Innovation Activities in Mergers and Acquisitions by Measuring Technological Distance

Tamagawa N, Takahashi H

Smart Innovation, Systems and Technologies (Smart Innovation, Systems and Technologies) 306 135 - 146 2022

ISSN 21903018

In this study, we construct an analysis system for the impact of mergers and acquisitions (M&A) on innovation activities using large-scale patent data in Japan, the U.S., and German. In the analysis, we focus on the technological similarity between the acquirer and the target companies in M&A. Specifically, we use natural language processing to analyze the patent document data and measure the technological distance among companies. In addition, we measure the innovation output in M&A using the patent data and classify the analysis targets according to the size of the technological distance. We also analyze the trend of the innovation output after the announcement. By conducting the analysis, we confirm that the innovation output after M&A announcement tends to increase in the M&A group with medium technological distance. In this result, one of the novelties is the applicability of unstructured data and machine learning methods to the research field of M&A and innovation.

Access to Document (DOI)

Survey on Social Simulation and Knowledge Extraction from Simulation Results - Application for Constructing Life Planning Support Frameworks -

高橋大志

International Journal On Advances in Software 15 54 - 56 2022

Accepted

Construction of a News Article Evaluation Model Utilizing High-Frequency Data and a Large-Scale Language Generation Model

Yoshihiro Nishi, Aiko Suge, Hiroshi Takahashi

SN Business & Economics (Springer Nature) 1 ( 8 ) 2021.07

Research paper (scientific journal), Joint Work, Accepted

News articles have significant impacts on asset prices in financial markets. A great number of attempts have been conducted to ascertain how news articles influence stock prices. News articles have been reported to contain sentimental and fundamental information that affects stock price fluctuations, and many studies have been conducted to evaluate stock price fluctuations using them as analytical data. However, the limitations in the number of available datasets usually become the hurdle for the model accuracy. This study aims to improve the analytical model’s accuracy by generating news articles using language generation technology. We tested whether the model that used the generated data was better than the trained model with real-world data. The model constructed in this research is a model that evaluates news articles distributed to financial markets based on the price fluctuation rate of stock prices and predicts and evaluates stock price fluctuations. This study labeled based on high-frequency trading data and generated news articles using a large-scale language generation model (GPT-2). Also, we analyzed and verified the effect. In this study, we succeeded in generating news articles using the large-scale language generation model and improving the classification accuracy. Our method proposed in this paper has great potential to improve text analysis accuracy in various areas.

Access to Document (DOI)

Analysis of the impact of crypto assets on investment portfolios before and after the outbreak of Covid-19

LIU Mengyao, JOTAKI Hiroaki, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2021 ( BI-018 ) 12 2021.03

Following the COVID-19 outbreak, unlike the plunge in traditional mainstream assets, crypto assets haveperformed very well. Indices that reflect changes in the crypto assets market have also grown in recent years. However,in terms of cryptocurrency index-related studies, there are not many, and the period analyzed is mostly three years. Inthis study, we analyze the impact of crypto assets on portfolio construction and attempt to clarify the risk-returncharacteristics of portfolios that include crypto assets as investment targets. We analyze historical data from July 2014to April 2020 based on the cryptocurrency index CRIX and six other traditional mainstream assets. By using theDCC-GARCH model, this study finds out the low dynamic correlation between the crypto assets and traditional ones.By using the mean-variance model, Cornish-fisher expansion, and T-copula CVaR approach to check the frontier lineand portfolio performance, we find out that crypto assets have the potential to improve the risk-return characteristics oftraditional portfolios.

A Study on Determining Factors of Amount of Gas Used in Japan — Analysis using multi-layer perceptron model —

HEO Myeongseok, MURAKAMI Eiji, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2021 ( BI-018 ) 14 2021.03

Gas and electric power are one of the essential social infrastructure for social and economicactivities. The purpose of this research is to construct a gas' consumption model suitable for thedistribution and sales market of LP gas in Japan by utilizing temperature. As a method of predictingdemand, we would like to first create a model of power consumption and then apply it to LP gas. Analysisof customers' gas consumption can infer from what standpoint they were consuming public goods.Through this, we would like to offer users' reasonable price and usage to expand their choices andcontribute to the efficient use of social resources. Furthermore, analysis of LP gas usage through deeplearning is possible.

Classification and Prediction of Stock Price Using Summarization and Machine Learning

NOYA Jun, TAKAHASHI Hiroshi

JSAI Technical Report, Type 2 SIG (The Japanese Society for Artificial Intelligence) 2021 ( BI-018 ) 08 2021.03

　 Previous page - Next page 　