Agent-based Social Simulationは現実に近い領域のモデルであるほどエージェントの行動モデルのパラメータ同定が難しいという課題がある．我々はBayesian Network（BN）を用いてこの課題を解決することを目的とする．実際にモデルを構築する対象として民間医療保険市場を設定する．民間医療保険の消費者行動モデルを構築し，そのパラメータに関わるアンケートを採り，アンケートデータに基づいて行動モデルに整合するBNを構築する．BNの確率推論を用いてパラメータ取得を行う．構築したモデルでシミュレーションを行い，民間医療保険市場における新たな知見を得る．

This paper reports the experiment of multi-agent simulation of the "Black Swan" appeared in the high-tech industry. The event does not occur only in rare case, event leading to catastrophic damage is called "Black Swan Events". And, the mutation that causes a "Black Swan event" is "Black Swan". Such "Black Swan" exists as a disruptive innovation in the high-tech industry.With the emergence of the Black Swan, there are cases where high-tech industry collapsed. Therefore, whether there is a countermeasure high-tech industry can take to the emergence of Black Swan. Based on the problem consciousness of them, has been developed (MAS) simple multi-agent simulation of this paper. The suggestions obtained from the simulation experiments are reported.

Purpose of promotion is to win users attention and interest, and to increase the con-tractor of service. Promotion includes various activities such as event participation, CM delivery,and information spread through SNS (e.g., twitter and Facebook). In this study, we try to analyzethe effect of promotion through twitter by counting the number of new follower. Our study follows\AISAS" framework. \Tweet" from company spreads through \retweet" and wins attention ofnew users who follow the original followers of company's account. In this study, we try to analyzethe in uence of retweet on new followers by 1) Poisson regression analysis, 2) Percentage of newfollowers who receives retweets, and 3) a decision tree analysis. The results of analysis which use4 selected twetter IDs are reported in this paper.

Downtown Dynamics, Artificial Society, Jane Jacobs, City Diversity Generator, Constructive Approach.

バリアフリー化や国際化の進展により、高齢者、車いす利用者、外国人観光客などの移動制約者がより自由に外出できるようになっているが、一方で災害発生時に路上や観光地にいる移動制約者を円滑に一時滞在施設へ誘導する必要性も増している。移動制約者は、移動速度の個人差が大きいこと、段差等が障害になること、土地に不慣れなこと等の行動特性を有しているため、群集に移動制約者が高い比率で含まれる場合は避難に要する時間が長くなる。そこで、これらの行動特性を持たせたエージェントを含むエージェントベースモデルを作成し、シミュレーションを行って、避難を円滑にする誘導方法や街づくりのあり方を検討する。

Systemic risk that propagates through financial systems causes insolvency or failure of particular financial institutions such as Bankruptcy of Lehman Brothers or European debt crisis. Although many researchers have challenged to find the propagation mechanism of the crisis in the inter-bank network, it is not clear completely yet. Namatame (2013) shows four fundamental models, Eisenberg-Noe model, GK model, NYYA model, and May model. We focus on May model which uses mean-field approximation methods of a network structure, and try to extend the model to Agent-based modelling with a realistic network structure.Considering liquidity risk and the inter-bank transaction structure of Japan, this paper also examines a possibility of an effect of a capital infusion. The purpose of this study is to find a suggestion to help systemic risk reduce by reviewing and simulating several cases of defaults in financial institutions.

when patent examiner indicates a refusal to patent application,"Notice of Reason for Refusal" is issued. We have created a synonym dictionary of compound nouns using the information extracted from the Notice of Reason for Refusal, and extracted the dependency relationship of compound nouns and verbs in the claims full text written in patent applications.Between this patent application and the patent publication cited, we measured a similarity by using created formula.Then, when using the synonym dictionary, we investigated whether this synonym dictionary is effective.

Decision-making in business is one of key issues.Any decision-making should align with business goal. Maximization of profit may not be the goal. When we think about a decision-making by a computer instead of human in business, we need to definewhat "better" decision-making means. Consideration on business, business goal, and business values is discussed.

In recent years, a method of machine learning technology has been advancing. Getting the distributed representation of words by the Neural Probabilistic Language Model (NPLM)has attracted interest. In this study, we attempt to create a new dictionary on the basis of oneswidely used in finance.In this analysis, we useNews Feed Direct of Thomson Reuters Corporation provides (NFD) that focus onthe market trends ofthe Japanese stock marketin order to make the dictionary.As a result of analysis, we could succeed in creating a new financial dictionary. However, we also found problems to besolved to get better performance.

The amount of logistics in Japan has been increasing in recent years. For this reason, logistics companies need to improve its efficiency. Although optimization of both human resources within a base and trucks has been already done, seasonal variation of the logistics itself has not been considered. This study firstly analyze the data of the amount of logistics in each base to overcome it.

This study analyses the relationship between textual information and nancial marketsin Japan, focusing on headline News, a source of information that has immediate in uence on themoney market, and also which is regarded as an important source of information when makinginvestment decisions. In particular we propose the objective way to estimate news articles' negative-positve by using mechine learning and statistics.

This study proposes a simulation model of a new type of infectious disease basedon smallpox and Ebola hemorrhagic fever and a health policy Game. SIR(Susceptible, Infected,Recovered) model has been widely used to analyse infectious diseases such as in uenza, smallpox,bioterrorism and so on. On the other hand, Agent-based model or Individual-based model beginsto spread recent years. The model enables to represent behaviour of each person in the computer.It also reveals the spread of a infection by simulation of the contact process among people on themodel. The study designs a model based on Epstein's model in which several health policies aremade a decision such as vaccine stocks, antiviral medicine stocks, the number of medical staff toinfection control measures and so on. Furthermore, infectious simulation of Ebola hemorrhagicfever which has not any effective vaccine yet is also implemented in the model. As results ofexperiments using the model, it has been found that vaccination ability per day and the numberof medical staff are crucial factors to prevent the spread. In addition to that, a health policy gameagainst a new type of infectious disease is designed as a serious game.