Conducting the Survey of Japanese National Character every 5 years since 1953, Chikio Hayashi has led the structure of thinking of the character of the Japanese using Quantification Method Ⅲ. In this paper, the author re-analyses the structures of thinking called ‘giri-ninjo’, ‘traditional vs. modern’ and ‘opinions about science and technology’ using data collected from 1953 to 2008. In short, the tendency was for the responses pertaining to ‘giri-ninjo’ to encompass abstract situations before 1988, and a difference in such situations was perceived after 2003. Regarding the structure of ‘traditional vs. modern,’ the responses considered as traditional thinking in 1953 were all very similar up until 1973. The cluster gradually loosened and the relationship has changed in regard to the thinking about science and technology, as well. Furthermore, a new movement is found in the structure of these responses. To compare the different cultures, it is not only important to compare each response, but also the general structure of responses that is not so specific as each response. Quantification Method Ⅲ is suitable for addressing these aspects.

In investigation of dynamics or quantitative assessment of trees in a forest, an analysis ofeach partitioned forest is conducted. For such an analysis, it is desirable to use a method fordividing a forest on the basis of similar ecological characteristics, rather than to division atregular intervals. Such areas having common ecological attributes are called patches. Currentlyavailable techniques for identifying patches are based on the limited tree species ortree composition in a grid. However, they do not fully reflect the fact that a forest comprisesvarious species of trees, besides the results are dependent on size of a grid. Therefore, it isnecessary to develop a new method for dividing a forest into patches.

In this article, we develop a technique for identification of patches in a forest using echelonanalysis, which can provide topological structure of space in hierarchically-representeddendrogram. We show that use of echelon analysis enables patch identification based on theinformation of all tree species and forest succession. Furthermore, we apply this techniqueto the actual data of the Ogawa forest in Ibaraki prefecture in Japan to demonstrate itsusability with the results showing the hierarchical structure and heterogeneity of the forest.

Biplots (Gabriel, 1971; Gower & Hand, 1996, Gower, Lubbe & Roux, 2011) provide thetwo-dimensional configurations of data matrices in which the rows (objects) and the columns(variables) of the matrices are plotted jointly. The biplots would be difficult to capture fordata matrices with a large number of objects and variables. In this paper, we consider suchmatrices of large sizes and propose a new biplot procedure. In this procedure, the objectsand variables are simultaneously classified into a small number of clusters using K-meansclustering (MacQueen, 1967), followed by bilplotting the resulting clusters of objects andvariables. This procedure allows us to capture configurations easily and further to grasp thememberships of objects and variables to clusters. A numerical simulation and a real dataexample are given to illustrate the effectiveness of proposed biplot procedure.

This paper proposes a method to recommend spots according to user preference inferredfrom the results of integrated analyses of GPS log and browsing log. Many content-searchor recommendation services are now available to handle the large volume of contents heldby the Internet. Typically, in the absence of very precise user requests, these services needlarge amounts of browsing or search log data before they can offer suitable contents. This iscalled the cold-start problem. Of particular note, it is extremely difficult for spot recommendationservices to get adequate amounts of data. For example, users do not use restaurantrecommendation services for regular meals taken at home. Our approach is based on thefact that most portable devices, including mobile phones, have GPS and it is easy to accumulatea GPS log. The proposed method computes distributions of tags for each user usingGPS log of the user and the spot-data gathered from the Internet. The method analyzes auser’s GPS log and browsing log with the distributions. The method is based on statisticalhypothesis testing and uses Stouffer’s Z-score method for the integration. We evaluate themethod through realistic GPS and browsing logs.