Uncertain data mining attracts much attention recently. In this paper, as an extension of probabilistic association rules, we propose probabilistic negative association rules to capture strong relationships between presence and absence of itemsets in uncertain databases. An algorithm is developed for extracting probabilistic negative association rules, which combines probabilistic frequent itemsets stored in a postfix tree.

Information and communication industry input-output tables are published every year. If we want to compare intertemporaly these tables, first, we must transform qualitative information and communication industry input-output tables by performing a coarse graining.In this study, we investigate the impact of these coarse-graining on the network scale. As the network scale, we use network centrality measures and centaralization.

Multi-Agent Simulation (MAS) is efficient for analysis of various social mechanisms. Recently, there are many studies on massive agent model to explain more complex social phenomena. Then, we aim for implementation of large scale simulation model using Repast HPC toolkit, a platform for massive agent model. In this article, we build "Schelling Segregation Model" for spatial model using geospatial data provided OpenStreetMap, an open source project creating a free editable map. In this model, agents are located continuous space , not grid in original. When an agent is "unhappy" and migrate to new location, it costs agents some simulation time depending on distance between old location and new one. This article reports simulation results using Japanese cities and verification result about execution time.

Representative members who make laws and heads of local governments are elected by citizens in Japan. Therefore, elections are one of the most important factors in future political or economic trends. There are a lot of researchers, such as social psychologists and political scientists, focusing on voters' decision-making processes. Lazarsfield et al. argued that people called opinion leaders had a powerful effect or influence on others. However, it is difficult to observe how voters make political decisions. We proposed a simulation model based on both Latane's dynamic social impact theory (DSIT) simulation model and Riker and Ordeshook's expected utility model of voting behavior to analyze political decision-making by using multi-agent simulation. Agents in the propose model communicate with order utility that is based on ambiguity in communicating information. Agents are generated from a database of public opinion polls. Then, agents are given social attributes and values of political parties. The agents' communication space is a hierarchical network constructed by agents that have the same or similar attributes. We analyzed the effect of opinion leaders on voters' political decision-making by using hierarchical network constructed with scale-free networks in the experiment.

This paper presents a method to extract causal relationships of events from Twitter. We extracted event-speci c words, which are frequently used in a speci c period, from tweet archives. Next, we make a series of event-speci c words for each user and make a transition relationship matrix by counting their anteroposterior relationships between event-speci c words. Existence or nonexistence of causality, its direction, and its strength are determined by analyzing a transition relationship matrix. Furthermore, we simplify an extracted graph structure by removing redundant causal edges. In fact, we make a causal relationship network from tweet archive in the Great East Japan Earthquake. We analyze the network structure and show that proposed method is suitable for extracting causal relationships.

This paper describes a crowd simulation with various parameter values for each pedestrian defined in the Helbing 's model, which represents a pedestrian as a molecular and calculates a compound force of the pedestrian. This applies many parameters to the pedestrian's force model, but most researches assign the same values into all pedestrians, because of simplicity. An actual crowd, however, has so many parameters assigned to the pedestrians, that the total behavior might be much different from the behavior with the identical parameters. This research, therefore, observed actual pedestrians waking on a street corner at Kichijo-ji, analyzed the detail behaviors using the video camera, and extracted appropriate parameters in the Helbing 's model for each pedestrian. The extracted parameters were assigned to simulation agents to reproduce a crowd behavior on computer. This simulation achieved the pedestrian crowd more accurately, in which they behave diversely and dynamically.

As for the frame of the Web advertisement, dealings are conducted in real time today. It is called RTB:Real-Time Bidding. In RTB, dealings of ad spaces are conducted in the form of the auction. The auction styles of RTB differ from the former. Since RTB has just started, it is not well known about a motion of a market. So, we try to modeling of market with the feature of RTB and considering price fluctuation under the influence of Auction participant's strategy in RTB.