Rieko Inaba

Difficulties in establishing common ground in multiparty groups using machine translation

When people communicate in their native languages using machine translation, they face various problems in constructing common ground. This study investigates the difficulties of constructing common ground when multiparty groups (consisting of more than two language communities) communicate using machine translation. We compose triads whose members come from three different language communities--China, Korea, and Japan--and compare their referential communication under two conditions: in their shared second language (English) and in their native languages using machine translation. Consequently, our study suggests the importance of not only grounding between speaker and addressee but also grounding between addressees in constructing effective machine-translation-mediated communication. Furthermore, to successfully build common ground between addressees, it seems important for them to be able to monitor what is going on between a speaker and other addressees.

Facial recognition by a compact parallel optical correlator

With the progress of information technology, the need for an accurate personal identification system based on biological characteristics is increasing the demand for this type of security technology instead of conventional systems using ID cards or pin numbers. Among other features, the face is the most familiar element and is less subject to psychological resistance. As a simple and compact recognition system satisfying the required performance, we implemented a hybrid system based on the optical recognition principle using a multi-level zone plate as a Fourier-transform lens and we report the preliminary results of their application to face recognition. In this paper, we present the design procedure and fabrication process for an improved version of a second-generation compact parallel correlator (named COPaC II), the size of which is 20 × 24 × 43 cm3 and weight 6 kg. As a result, we obtained a low error rate of 0% as the false match rate and 0.3% as the false non-match rate, thus the COPaC II significant identification security level is sufficiently stable. With the aim of further enhancing the throughput and robustness, we conducted performance tests where the system is used as a computer log-in device and as a pre-screening device for crime investigation. In both experiments, a high rate of successful recognition, such as 90% recognition and 94% rejection rate for log-in, was obtained. Experiments on twins to check the disguise recognition, and on the effects of changes in brightness and arbitrary size of images to test its robustness are also included.

Language Grid Association: Action Research on Supporting the Multicultural Society

The Language Grid is a middleware with which people can connect and use language resources such as machine translations, morphological analyzers and others created in the fields of intercultural collaboration. The Language Grid cannot exist without the collaboration of Language Grid Users who provide language and computation resources, language services, and collaboration tools. This paper overviews Language Grid Association, a user group of the Language Grid and a body promoting action research to support the multicultural society.

Usability of multilingual communication tools

Multilingual communication tools are needed to support intercultural collaboration. I describe a tool that supports multilingual communication and propose a model of the tool in which usability subjects are extracted using empirically evaluated rules.

Culturally-situated pictogram retrieval

This paper studies the patterns of cultural differences observed in pictogram interpretation. We conducted a 14-month online survey in the U.S. and Japan to ask the meaning of 120 pictograms used in a pictogram communication system. A total of 935 respondents in the U.S. and 543 respondents in Japan participated in the survey to submit pictogram interpretations which added up to compose an average of 147 English interpretations and 97 Japanese interpretations per pictogram. Three human judges independently analyzed the English-Japanese pictogram interpretation words, and as a result, 19 pictograms were found to have culturally different interpretations by two or more judges. The following patterns of cultural differences in pictogram interpretation were observed: (1) two cultures share the same underlying concept, but have different perspectives on the concept, (2) two cultures only partially share the same underlying concept, and (3) two cultures do not share any common underlying concept.

Pictogram retrieval based on collective semantics

To retrieve pictograms having semantically ambiguous interpretations, we propose a semantic relevance measure which uses pictogram interpretation words collected from a web survey. The proposed measure uses ratio and similarity information contained in a set of pictogram interpretation words to (1) retrieve pictograms having implicit meaning but not explicit interpretation word and (2) rank pictograms sharing common interpretation word(s) according to query relevancy which reflects the interpretation ratio.

Language Grid Revisited: An Infrastructure for Intercultural Collaboration

Since various communities withmultiple languagesinteract in daily life, tools are needed to support intercultural communication. However, we often observe that the success of a multi-language tool in one situation does not guarantee its success in another. To develop multi-language environments that can handle various situations in various communities, existing language resources should be easy to share and customize. Therefore, we designed the Language Grid as service-oriented collective intelligence; it allows users to freely create language services from existing language resources and combine them to develop new services to meet their own requirements. This paper explains the design concept and service architecture of the Language Grid, and our approach to user involvement in collective intelligence activities.An institutional design is also essential forcollective intelligence. We create a federated operation model to bridgedifferentstakeholders including service providers, service users, and service grid operators.

Compact parallel joint transform correlator for facial recognition

We discuss the design of a compact parallel joint transform correlator (PJTC) using four binary zone plate array as a Fourier transform lens. The external dimensions of the hybrid opto-electronic correlator is 23 cm X 15 cm X 16.3 cm, which includes the PJTC system in the upper stage and some of driving circuits and an LD for a light source in the lower stage. Experimentally, we implemented this compact PJTC system for facial recognition with database of 100 persons and obtained reliable discrimination results. Furthermore, we propose a novel method of pre-processing so that any general users could easily operate the compact PJTC system and constantly obtained the same size extracted images, by fixing three points; the edges of both eyes and the low end of nose, when normalizing the facial image size. We confirmed the ability of our system to meet the conditions for recognition by JTC.

Discrimination of portraits using a hybrid parallel joint transform correlator system

A hybrid parallel joint transform correlation system is demonstrated through the introduction of a five-channel binary zone plate array and is applied to the discrimination of portraits for a presumed criminal investigation. In order to improve performance, we adopt pe-processing of images with white area of 20%. Furthermore, we discuss the robustness.

Survey on Copyright Infringement of Digital Contents: A Case Study of Japanese University Students

This study clarified the actual use of infringed content for university students in Japan using a survey on the target video and audio contents. In this study, the following became clear. (1) The use of digital content has been pervasive. The demand for music-related content is high. (2) Most users watch and listen to digital contents without consideration for copyright infringement. Indeed, users indicated feeling no guilt and concern for the risks involved when accessing infringed materials. These results show that many users lack a proper understanding of copyright laws.