[in Japanese]

View full abstract

[in Japanese]

View full abstract

Recently, blog become an important tool for putting out the information and for communicating with other people. Then blog users are increasing. Thus, we can regard blog contents as public opinion connected with social events. Hence, we believe that we can obtain trustable information by using blog contents of important bloggers. In this paper, we define blog data model and types of important bloggers. Moreover, we propose and evaluate a method for discovering an Agitator, which we regard as the most important blogger, by analyzing real blog data. We have evaluated the recall rate of discovering Agitators and the appropriateness of blog sites which are judged as Agitator based on our method. Consequently, the results of evaluations show that our method is effective to discover an Agitator as important blogger.

View full abstract

Kohonen's Self Organizing Map (SOM) is a kind of neural networks, that learns the feature of input data thorough unsupervised and competitive neighbourhood learning. According to SOM learning algorithm, SOM learning speed is affected by learning data and initial value of feature map. By improving conventional initialization method of feature map, the influence to the learning speed is expected. In conventional initialization process, very connection weights in SOM feature map are initialized to random values, and this is also set nodes to random point of SOM feature map independently with data space. Because some linkage between them is created when learning is completed, It is expected that no linkage at initialization must be influenced to the learning speed. In this paper, here I proposed a new method, node exchange of initial SOM feature map using learning data, and a new measure of convergence, the average of the move distance of all output nodes. As a result of experiments, the effect is sufficient by performing exchange process by about 5% of the whole learning data, and the average of the move distance of all nodes was drastically shortened about 64% through about 45%, and learning speed is improved that it becomes about 55% by this method.

View full abstract

Application of travel behaviour model is highly desired to describe the daily trip pattern of the trip makers as an important analysis technique in urban transport planning. It is required that the model would describe the decision for purpose, mode, destination, and route for the trip chaining of individual trip makers. In the study, the integrated approach with fuzzy reasoning and ID3 algorithm would be proposed to formulate the decision process to establish the fuzzy modal choice model. The trip patterns can be analyzed significantly corresponding to the trip survey results for urban area. Furthermore, the effectiveness of transport policy can be discussed because the travel behaviour analysis would be carried out with considering human fuzziness.

View full abstract

Texture image mapping is a useful technique for improving quality of 3D computer graphics with low costs. When an only small image is available for texture image mapping and it is mapped on large mapping area repeatedly, discontinuity problem of intensity occurred at boundary line of neighbor texture images, and several methods proposed to solve this problem. In this paper, we propose improved image quilting method that can synthesize arbitrary size texture image with smooth boundary line by considering horizontal connectivity. Through an experiment, our method can synthesize wide area texture image with higher quality and smoother boundary line than common method.

View full abstract