In this paper, we explain a method to obtain the user's emotions in an event. Simple emotions (pleasure/displeasure) extracted by Emotion Generating Calculations (EGC) are divided into 20 various emotions based on the Emotion Eliciting Condition Theory. The theory requires the discriminate criterion to judge which an event corresponds to "Feeling for another," "Prospect and confirmation," or "Approval/disapproval" "Feeling for another" means someone's emotion (not mine) in the event. It is determined by analysis result of EGC under the condition of another's taste information. "Prospect and confirmation" is determined by extracting some aspects and adverbs about the tense to judge. "Approval/disapproval" is determined according to the utterance's case frame structure with transitive verb.

抄録全体を表示

This paper presents an application of a fuzzy integral model to an endoscopic image diagnosis and an examination of the model structure. Measures derived from human subjectivity do not usually possess additivity; thus, fuzzy measures requiring monotonicity rather than additivity can be treated as subjective measures of the person evaluating an object. In medical diagnostic imaging, even though images have no ambiguity and their items are strictly quantified, these values are more or less subjective. In the present study, the Sugeno integral model with λ-fuzzy measures for distinguishing a gastric ulcer from a gastric carcinoma was constructed using input-output data obtained from a physician. The model parameters were optimized using a method involving a genetic algorithm. As a result, the items having a high importance degree in the model corresponded with the items estimated by the physician to have a high importance degree.

抄録全体を表示

In this paper, firstly, we described a paired comparison test of foreground (character) and background colors. The color sets were selected from achromatic colors of the Web Safe Color, which is a de facto standard for the HTML color representation. Secondly, we showed a SOM (Self-Organizing Maps) analysis to estimate different color sets for visibility of same users. Since a user profiling is required beforehand for every user, we solved this problem with SOM. Finally, we proposed a Web browsing method with an automatic color correction (palette changing) for personal user's visibility. We applied the result of the test and analysis to our method, and we implemented a Palette Changing System for evaluating of usability as CGI (Common Gateway Interface) software on a HTTP server of the Internet. It was suggested that our method is useful for automatic color correction of Web browsing and is available for improving visibility and individual personal usability.

抄録全体を表示

This paper describes a technique for representing human motions employing an eigenspace technique. A video image sequence obtained from a human motion defines a multi-dimensional eigenspace in which the image sequence is represented as a curved line. The trajectory of this line should depend only on the motion and not on a person changing his/her dress in order to proceed to the recognition of unknown human motions. A different curved line is, however, created each time a person wears different dresses, since the original eigenspace technique employs initial gray-level images. To escape from this difficulty, the present technique employs blurred edge images of the original. In the performed experiment, 5 persons wearing respective clothes play 6 actions of a cricket umpire. A 6-dimensional eigenspace is created from their blurred edge images, in which one is able to see that the trajectories of the obtained 6 curved lines depend on the actions and not on the persons.

抄録全体を表示