The authors’ group has been developed AI-based Online Consensus Building Support System called COLLAGREE and used around our university since 2010. The goal of our research is to spread this system to other local governments. However, new technology which directly concerns to society might make a conflict with current society. This problem has been commonly known as ELSI. In this paper, we analyzed our past experiments by using our system, developed the reason of success, and ELSI problems that need to solve. To achieve this purpose, we launched ELSI Committee on December 2017. Our approach should be useful for other AI-based systems, when researcher has been conscious on ELSI issue.
Many kernels for RDF graphs have been designed to apply to machine learning such as classification and clustering. However, the performances of these kernels are affected by the variety of RDF graphs and machine learning problems. For dealing with the lack of robustness, this study proposes a generic kernel function called skip kernel that is a generalized of the existing PRO kernel. We formalize a feature extraction in the skip kernel that replaces some edges and nodes (corresponding to predicates and objects) of each resource with variables in a RDF graph. The skip kernel is effectively computed by (i) a recursive process of constructing each set of resources from RDF graphs and (ii) a size calculation of the intersection of two sets of skip structures for resources. We show that the time and space complexities of computing the skip kernel are reduced from O(d(2MN)d) and O(d(M +1)d-1MN) to O((M +1)d-1MN2) and O(M +dN), respectively. In our experiments, several kernels (skip, hop, PRO, walk, path, full subtree, and partial subtree) with SVMs are applied to ten classification tasks for resources on four RDF graphs. The experiments show that the skip kernel outperforms the other kernels with respect to the accuracy of the classification tasks.
This paper addresses character expression for humanoid robots that play a given social role such as a lab guide or a counselor via spoken dialogue so that the character matches to the social role. While most conventional methods of character expression aim to change the style of utterance texts, this study focuses on dialogue features that may affect the impression of spoken dialogue. Specifically, we use five features: utterance amount, backchannel frequency, backchannel variety, filler frequency, and switching pause length. We adopt three character traits of extroversion, emotional instability, and politeness for a character expression, and investigate the relationship with the dialogue features. A statistical analysis of subjective evaluations shows that the dialogue features except for the backchannel variety are related to either of the traits. By using the subjective evaluation scores on the relevant traits, we can train models to control the dialogue features and behaviors according to the desired character. An experimental evaluation demonstrates the feasibility of character expression with regard to the traits of extroversion and politeness.