Published papers

本研究は、飲食店のPOSデータと顧客データを用いて企業の経営力向上に繋げるシステムの提供を目的としている。昨今のビッグデータ活用時代において、個人のデータを蓄積し利活用することの重要性が増してきている。個人が頻繁に利用しデータ取得可能な業態として飲食店が挙げられ、そこで蓄積されたデータを分析することでCRMに有用な示唆を得ることができると考える。本分析では、顧客の購買行動の特徴を明らかにし、在庫管理と販売政策について検討を行った。

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.

本研究は製薬企業が保有する特許が企業業績にどのような影響を与えているかを、特許の抄録集を抽出しベクトル表現を獲得することで企業の技術距離を年次的に算出、その関連性を分析している。日本の社会保障の構造上、製薬企業は「長期収載品に依存するモデルから、より高い創薬力を持つ産業構造に転換」に迫られている。製薬企業にとって、特許は莫大な研究開発費用の投資による結果であり、これを分析することで企業の技術戦略の方向性を窺い知れる可能性がある。分析の結果、技術的多角化は(1)企業のファンダメンタルに負の影響を与える可能性が示唆された。(2) 株式市場に負の影響を与える可能性を示せた。以上の結果は技術的進歩を常に追い求める製薬企業の企業経営にとって重要な示唆を与える。

There are many previous studies on IPO open prices. In particular, underpricing,where the public price is lower than the initial price, is an important issue from themanagement's point of view, as it causes loss of opportunity for issuers. Various hypothesesabout the cause of underpricing have been made by previous research. In this research, wefocus on the calculation of corporate value when determining the open price. Specifically,SCDV was created based on patent documents owned by the company. Next, a similarcompany comparison method was performed by selecting similar companies based on thecreated document vectors, and the validity of the open price was attempted.

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.

In Europe and the United States, appearance is considered to be an important factor in business,and not a few studies have examined the effects of managerial appearance on corporate management andindividual success and performance. In this study, we attempt to analyze quantitatively the faces ofmanagers in Japan, which is one of the most important components of appearance, in order to elucidatewhether a similar phenomenon is found in Japan. Photographs of managers' faces are collected fromannual reports and other sources, and feature vectors are extracted. This will be clustered using theK-means method using FaceNet's trained model VGGFace2 to analyze whether there are differences inthe behavior of the companies in each cluster, i.e., how the face of the manager affects the company'sbehavior.

There are also many studies that suggest that the characteristics of managers, such as theirbackground, orientation, and age, have some influence on the performance and behavior of companies. Inthis paper, we analyze the relationship between managerial face and corporate behavior, especially R&Dinvestment, after focusing on face as one of the important components of appearance. We classified the facetypes of managers into several patterns and analyzed the impact of these types on corporate behavior.as aresult, it was confirmed that managers classified in one cluster tend to invest more actively in R&D thanthose classified in the other clusters.

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.

Patents in pharmaceutical companies are the result of R&D investment, and can be regarded as one of the indicators that represent the results of exploratory research for the creation of new drugs, as well as reflect the company's strategy for the technological domain. Therefore, if patent documents are quantified, it is possible not only to objectively grasp the degree of concentration or expansion of technological fields, but also to make comparisons between companies. In this study, we attempted to visualize the transition of the technological domain of Japanese, European, and American pharmaceutical companies by quantifying the technology travel distance using vector expressions extracted from patent abstracts.

This paper proposes a method of building a polarity dictionary using news articles and stockprices in the Chinese market by textual analysis in finance. In order to measure the degree of polarity, weassociated the news articles' sparse composite document vectors to a score. The score is calculated by themethod of event study with the abnormal change rate of stock prices on the publication date. Weconducted support vector regression (SVR) and built a polarity dictionary with polarity data from learners.Furthermore, we made a comparison on accuracy to traditional ways of calculating word polarity inwhich news articles are represented by a one-hot wordlist. The comparison of the existed polarity is made.

The location is important for the real estate management business, but it is difficult to bequantified. As a case study of the Shimokitazawa station, the correlation analysis of retail rent was examinedwith building and location measures. The location measures include KDDI's pedestrian data in each samplepoint from GPS of mobile phone because the recent IT development has enabled us to get pedestrian dataeasily. The result showed positive correlation between retail rent and the pedestrian data can be seen.

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.

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.

Pedestrian traffic is reported to be a component of rental price. However, rental price analysiswith an agent-based simulation has not been conducted. In this article, shop rental around Shimokitazawastation is analyzed by conducting an agent-based simulation. The result shows that there is a positivecorrelation between shop rental and the number of footprint, which is the output of the agent-basedsimulation.