Keynote Talk

Sensing-based Information System in the Era of IoT, Cloud Services and Smartphones

Prof. Hiroshi Tanaka
Department of Information and Computer Sciences
Kanagawa Institute of Technology

The number of computers is increasing exponentially, and they are becoming smaller and smaller. At the same time, their processing performance is also increasing, and many kinds of wireless communication systems are being developed to improve both speed and reliability. MEMS technologies are also advancing, and tiny sensors with low power consumption are being introduced into the computing environment. Smartphones are a representative example of the application of such technologies. If a smartphone is able to connect to cloud services, the capacity to provide many more information services is greatly expanded.

In this talk, some practical examples and demonstrations using a smartphone, including its integration with other devices/appliances and cloud services, are introduced by using demonstration videos. The information system the speaker is presenting in the conference can be divided into three categories:

  1. A new positioning system using smartphone functions that achieve greater accuracy compared with that using radio systems
  2. A home electrical appliances operation system based on a platform comprising a smartphone and cloud services
  3. Newly developed systems that are now being used on our campus as a lecture tool and for ECO activity

All systems have been researched and developed in cooperation with laboratory students. The speaker will be particularly delighted if these applications can be of references to professors and enhance the motivation of students.


Developing Language Resources for Under-Resourced Languages in Indonesia

Dr. Totok Suhardijanto
Department of Linguistics
Faculty of Humanities
University of Indonesia

Language resources plays a key role in human language technology especially NLP and computational linguistic works such as machine translation, sentiment analysis, automatic sound recognition, and text-to-speech. Many NLP applications has been built based on existing language resources. Over the past decade, major languages such as English, French, and German have experienced a significant progress and have a tremendous language resources for building application, linguistic research, and future use. Unlike these major languages, many other languages of the world are still neglected from the perspectives of language engineer. As a result, it is difficult to find any existing language resources in minor languages including languages og Indonesia. With regard to Indonesia, it has 742 languages that most of them are under-resourced languages. In this paper, we present an analytical review of existing problems and approaches in our effort to develop language resources for many languages lacking speech and text resources in Indonesia.

This paper describes the creation of useful resources for languages that currently lack them, taking resources for Indonesian and Javanese treebanks as case studies. Three different paradigms are discussed for creating language resources, namely: (1) using crowdsourcing to produce a small resource rapidly and relatively cheaply; (2) translating an existing gold-standard dataset, which is relatively easy but potentially of lower quality; and (3) using manual effort with appropriately skilled human participants to create a resource that is more expensive but of high quality.