Masahiro Hamasaki

National Institute of Advanced Industrial Science and Technology · Human Informatics and Interaction Research Institute

We propose a learning-based method of estimating the compatibility between vocal and accompaniment audio tracks, i.e. , how well they go with each other when played simultaneously. This task is challenging because it is difficult to formulate hand-crafted rules or construct a large labeled dataset to perform supervised learning. Our method uses s...

We present a novel concept audio–visual object removal in 360-degree videos, in which a target object in a 360-degree video is removed in both the visual and auditory domains synchronously. Previous methods have solely focused on the visual aspect of object removal using video inpainting techniques, resulting in videos with unreasonable remaining s...

For amateur creators, it has been becoming popular to create new content based on existing original work: such new content is called derivative work. We know that derivative creation is popular, but why are individual derivative works created? Although there are several factors that inspire the creation of derivative works, such factors cannot usua...

We describe the AIST Dance Video Database (AIST Dance DB), a shared database containing original street dance videos with copyright-cleared dance music. Although dancing is highly related to dance music and dance information can be considered an important aspect of music information , research on dance information processing has not yet received mu...

Audio annotation for music clips is an important task for machine-learning-based music analysis and applications. However, it is a time-consuming task because it often requires repetitive manipulations even though typical audio files often contain repetitive structures (e.g., a song often has similar phrases used multiple times). In this paper we p...

Creating new content based on existing original work is becoming popular especially among amateur creators. Such new content is called derivative work and can be transformed into the next new derivative work. Such derivative work creation is called "N-Th order derivative creation." Although derivative creation is popular, the reason an individual d...

This paper describes a music exploratory search interface called QueryShare, which provides query searching and recommendation functions for query sharing among users. Most people are not expert users who know how to use various music metadata that include automatically estimated musical features to represent their own information needs as a query....

Many amateur creators now create derivative works and put them on the web. Although there are several factors that inspire the creation of derivative works, such factors cannot usually be observed on the web. In this paper, we propose a model for inferring latent factors from sequences of derivative work posting events. We assume a sequence to be a...

We propose a novel interface that allows the user to interactively change the playback order of multiple songs by choosing one or more criteria. The criteria include not only the song's title and artist name but also its content automatically estimated by music/singing signal processing and artist-level social analysis. The artist-level social info...

Hand-over note in elderly care facilities is a very important media for the employees to share information about non-routine task procedures in the daily operation. Reviewing these memos regularly will help the facility find potential risk factors and inefficiency in routine protocols. However, hand-over notes are typically hand-written memos and c...

Many aspects of online social networks (OSN) have been studied in recent years. In this article, we focus on the question of interactions in large OSN. We propose methods to study these interactions, and apply them on a platform called Nico Nico Douga (NND), with the aim of understanding cooperative behaviors, taking the form of collective creation...

The experience and intuition gathered over many years of employment are extremely important for providing high-quality service in the fields of nursing and care. However, these experience and intuition are subjective, making it difficult to pass on experience and related know-how to a novice. To realize such field community intelligence, we have ta...

This paper describes a music browsing assistance service, Songrium (http://songrium.jp), which increases user enjoyment when listening to songs and allows visualization and exploration of a ``Web of Music''. We define a Web of Music in this paper to be a network of ``web-native music'', which we define in turn to be music that is published, shared,...

A handover support system that supports care workers to share information and knowledge on patients and nursing-care work based on information recommendation is described. A handover is time consuming work because it takes much time to write and retrieve information on patients. We investigated the handover work in a nursing home, and found that ab...

This paper describes a music browsing assistance service, Songrium (http://songrium.jp), that helps a user enjoy songs while seeing visualization of open collaboration. Songrium focuses on open collaboration for music content creation on the most popular Japanese video-sharing service. Since this open collaboration generates more than half a millio...

Innovative service operations in the healthcare field should be cooperative and proactive. However, this is often difficult because separate providers have different ideas and backgrounds and little information of others' practices. For example, we found that workers in a care facility share one notebook for communication and have no incentive to i...

Although a large number of datasets gathered as Linked Open Data (LOD) is better for data sharing and re-using, the datasets themselves become more difficult to understand. Since each dataset has its own data structure, we need to understand datasets individually. In addition, since the entities in datasets are interconnected, we need to understand...

Wakoen Care Facility workers share one notebook for communication. Most Saga University Hospital point-of-care system PDAs are not used by nurses, who must record and share information, mainly because the system interface impedes their workflow. User-oriented system development is important for nurses and care workers. First, a development style is...

In this paper, we present a new solution for trend detection in user-generated content, and more particularly Web 2.0 social networks. Whereas some propositions have been published in this domain recently, we have chosen a new approach based on network analysis. We first create an evolving network of terms, which is an abstraction of the complete n...

A new trend for applications for the Internet is to create and share tiny codes for applications like site-specific codes for a broswer extension. It needs a new tools to share and distribute such codes efficiently. We built a Wiki site called Wedata which stores tiny code for a particular service. Wedata has three features: machine readability, co...

Recently we often use SNS such as Twitter or Facebook to discuss about a particular topics online. We use "hashtag" to demostrate the topic of a tweet, thus we can share the opinion about the topic. We use "like!" button to demostrate our feeling of a comment, thus our consent will be visualized. We use such technics to enact discussion smoothly. I...

In this paper, we propose an adaptive ontology-based approach for related paper identification, to meet most researchers' practical needs. By searching ontology, we can return a diverse set of papers that are explicitly and implicitly related to an input paper. Moreover, our approach does not rely on known ontology. Instead, we build and update ont...

We propose a novel style of social tagging to construct knowledge collaboratively called Social Property Tagging and introduce the prototype system Social Infobox. Structured data is useful for computer system, however defining structure of knowledge for representing data semantics is usually a costly and time consuming task. In general, data struc...

In this paper, we describe our approach for information extraction from documents, which is based on supervised machine learning and collective intelligence approach. This approach is aimed at redeeming each method, because each method has merits and demerits. It provides various ways for users to input data to improve information extraction. Users...

This paper presents an expressive activity program and a support system for collaborative storytelling based on multiple compositions: Zuzie. Zuzie’s storytelling is designed so that people can experience rich expressions and extend their interpretation of them through creation of compositions. Each composition is a deliberate arrangement of a sing...

In this paper we present a low-cost method for typing Named Entities with Wikipedia. Different from other text analysis-based approaches, our approach relies only on the structural features of Wikipidia and the use of external linguistic resources is optional. We perform binary classification of an article by analyzing of the names of its categorie...

The Web technology enables numerous people to collaborate in creation. We designate it as massively collaborative creation via the Web. As an example of massively collaborative creation, we particularly examine video development on Nico Nico Douga, which is a video sharing website that is popular in Japan. We specifically examine videos on Hatsune...

The Web technology enables numerous people to collaborate in creation. We designate it as massively collaborative creation via the Web. As an example of massively collaborative creation, we particularly examine video development on Nico Nico Douga, which is a video sharing website that is popular in Japan. We specifically examine videos on Hatsune...

The aim of transfer learning is to improve prediction accuracy on a target task by exploiting the training examples for tasks that are related to the target one. Transfer learning has received more attention in recent years, because this technique is considered to be helpful in reducing the cost of labeling. In this paper, we propose a very simple...

Studies of interactive systems in museums have raised important design considerations, but so far have failed to address sufficiently the particularities of family interaction and co-operation. This paper introduces qualitative video-based observations of Japanese families using an interactive portable guide system in a museum. Results show how une...

The Web technology enables numerous people to collaborate in creation. We designate it as massively collaborative cre- ation via the Web. It is becoming an important activity such as Wikipedia and Yahoo! QA. As an example of massively collaborative creation, we particularly examine video devel- opment on Nico Nico Douga, which is a video sharing we...

The Web technology enables numerous people to collaborate in creation. We designate it as massively collaborative creation via the Web. It is becoming an important activity such as Wikipedia and Yahoo! QA. It has an interesting characteristic, especially in an entertainment content creation. People create their own content freely and there is few e...

In this paper, we propose Modulobe: a creation and sharing platform for articulated 3D models with complex motion. Modulobe has two components: a 3D modeling application and a model sharing web site. The former is intended to achieve complex motion simulation of a D model and provide a user-friendly interface. Motion pattern of angle for each hinge...

The World Wide Web supports new styles of creative ac- tivities. For this study, we investigate massively collabora- tive creation via the Web, by which numerous people gather to evolve their works collaboratively. Nico Nico Douga is a video sharing website, where many videos are created collab- oratively. We specifically examine Hatsune Miku ,av e...

We address a new machine learning problem, taming, that involves two types of training sets: wild data and tame data. These two types of data sets are mutually complementary. A wild data set is less consistent, but is much larger in size than a tame set. Conversely, a tame set has consistency but not as many data. The goal of our taming task is to...

We proposed BaggTaming to boost the prediction accuracy by exploiting additional data whose class labels are less reliable. This algorithm is successfully applied to the personalized tag predicition for the data collected from the delicious. To check whether our method is generally effective, we test the data crawled from the hatena bookmark.

本論文では，ソーシャルマッチングにおいて人による紹介を用いることを提案し，その有用性とそれを助ける支援システムについて考察を行う．紹介は，私たちが日常生活においてごく一般的に行う行為であり，あまりに日常的なため深く考えることは少ないが，それによって出会いの支援や効率化が行われている．しかし「紹介」は人にとっても容易なタスクではない．そこで本研究ではソーシャルマッチングのための紹介支援システムについて考察を行う．我々はプロトタイプシステムを開発し，運用実験を行った．運用結果からは，人による紹介がユーザにとって興味のある相手にコンタクトするのを支援する手法として有効であること，紹介支援として紹介文のドラフトを自動生成することが有効であること，紹介者にとって紹介を希望するユーザがなぜ相手に興味があ...

We analyzed the user behavior of Modulobe, a creation and sharing platform for articulated 3D models with complex motion for sup- porting social creativity. Modulobe has two components: a 3D modeling application and a model sharing web site. The former is intended to achieve complex motion simulation of a D model and provide a user- friendly interf...

To realize a new collaborative communication system, we propose qwikWeb which has a design philosophy based on communication patterns where a user can easily build a system that matches the group's activity. Adequacy and efficacy of this system were demonstrated by designing, implementing, operating, improving, and conducting analysis of operation...

Concepts of ‘culture’ are often invoked in analysis of humancomputer interaction, notably in attempts to refine or adapt systems to differing cultural contexts, such as in the process of internationalization or in creating systems and processes that can adapt to user’s cultures. This paper takes ethnographic research in this area to the study of cu...

This paper proposes integration of a social network with the tripartite model of ontologies by P. Mika. That model is based on three dimensions, i.e. ac-tors, concepts and instances, and illustrates ontol-ogy emergence using actor-concept and concept-instance relations. However, another important in-gredient is the actor-actor relation. For example...

One of the challenges for ubiquitous computing is to design systems that can be both understood by their users and at the same time understand the users themselves. As information and its meaning becomes more associated with the communities that provide and use it, how will it be possible to build eective systems for these users? We have been exami...

A social networking service can become the basis for the information infrastructure of the future For that purpose, it is important to extract social networks that reflect actual social networks which users have already had Providing a simple means for users to register their social relations is also important We propose a method that combines vari...

Social networks play important roles in the Semantic Web: knowledge management, information retrieval, ubiquitous computing, and so on. We propose a social network extraction system called POLYPHONET, which employs several advanced techniques to extract relations of persons, detect groups of persons, and obtain keywords for a person. Search engines...

In this paper, we propose a Weblog system called ActionLog, which can associate Weblog entries to real world contexts. The real world context is not only useful for Weblog authors themselves, but also beneficial to communication among people, because people with the same or a similar context can easily find each other. ActionLog collects user actio...

Social networks play important roles in the Semantic Web: knowledge management, information retrieval, ubiquitous computing, and so on. We propose a social network extraction system called POLYPHONET, which employs several advanced techniques to extract relations of persons, to detect groups of persons, and to obtain keywords for a person. Search e...

Social networks are important for the Semantic Web. Several means can be used to obtain social networks: using social networking services, aggregating Friend- of-a-Friend (FOAF) documents, mining text informa- tion on the Web or in e-mail messages, and observing face-to-face communication using sensors. Integrating multiple social networks is a key...

There are many situations where individuals who belong to large social networks meet in smaller communities on an 'ad hoc' basis. This paper reports on the development and use of a system to support the micro-communities that form in these settings. The system combines the benefits of automatic social network visualization with the ability for memb...

Social Network Systems have become highly applicable to everyday life, but continue to remain on the desktop for most users. This paper reports on initial analysis of groups interacting with a social network system in the real-world, in this case a conference setting. The system, Polyphonet Conference, and its RFID card interface together allows ri...

本論文では, Webマイニングとソーシャルネットワークサービス, そして実世界でのICカードを用いた学会支援システムについて述べる. 我々は, 人工知能学会で2003年, 2004年, 2005年と3年間, 学会支援システムを運用している. 本稿では, 特にシステムの中で得られた3つのデータに着目し, その分析を行う. 3つのデータとは, Webマイニングによって得られた参加者の関係性のデータ, システム上でソーシャルネットワークとして知り合い登録を行った知り合い関係のデータ、そして実世界でユーザが情報キオスクを利用したというインタラクションのデータである. この3つのデータを比較した結果, Web上の情報が知り合い登録の生成に, 知り合い登録情報が実世界でのインタラクションに, それぞれ関...

One target of a ubiquitous computing environment is to create and activate communities in which people interact in a lively manner based on their interests and situations. To achieve that goal, we have been developing a "ubiquitous community support system (UbiCoSup)" for event spaces since 2002 because event spaces have rich contents and interacti...

In this paper, we propose the communication supporting system with location-based information such as landmarks. Our system provides cooperative aggregation environment for location-based information. The user can also communicate each other with.

In this paper we propose a social bookmark system using sev- eral metadata and personal network to construct a community-based on- tology. Our system allows users to browse friends' bookmarks on his/her personal network. The user can map their own tags onto more than one tag from different friends so that they are linked to the user. The matchmaker...

We propose a personal knowledge publishing system called Semblog is realized with integration of Semantic Web techniques and We- blog tools. Semblog suite provides an integrated environment for gath- ering, authoring, publishing, and making human relationship seamlessly to enable people to exchange information and knowledge with easy and casual fas...

Hierarchical categorization is a powerful and convenient method so that it is commonly used in various areas, such as ontologies. Although each hierarchy is useful, there are problems to manage multiple hierarchies. In this paper, we propose an alignment method between concept hierarchies by using the similarity of the categorization and the conten...

When we have a large amount of information, we usually use categories with a hierarchy, in which all information is assigned. The Yahoo! Internet directory is one such example. This paper proposes a new method of integrating two catalogs with hierarchical categories. The proposed method uses not only the contents of information but also the structu...

With the rapid advance of information technology, we are able to easily and quickly obtain a great deal of information on almost any topic. One method by which to managing such large amounts of information is to utilize catalogs which organize information within concept hierarchies. However, the concept hierarchy for each catalog is different becau...

When we have a large amount of information, we usually use categories with a hierarchy, in which all information is assigned. This pa- per proposes a new method of integrating two catalogs with hierarchical categories. The proposed method uses not only the contents of informa- tion but also the structures of both hierarchical categories. We conduct...

In this paper, we discuss importance and utilization of personal network in a community system through the result of management and analysis of the scheduling support system for academic conferences. The important feature of the system is generation and utilization of personal network to support information exchanging and information discovery amon...

In this paper, we discuss importance and usefulness of interpersonal network in a community support system. We built a scheduling support system for an academic conference. Our system supports information exchange among participants and information discovery with generating participants’ interpersonal network. This system was used in an academic c...

We propose an egocentric search method which evaluates the importance of information by the distance between the user and the information. Personal networks around the user can be ex- tracted by using Weblog and RSS. We per- formed an experiment and found this distance is related to similarity between documents. We applied the method to a Weblog ed...

In this paper, we propose an algorithm called Neighborhood Matchmaker Method to optimize personal human networks. Personal human network is useful for various utilization of information like information gathering, but it is usually formed locally and often independently. In order to adapt various needs for information utilization, it is necessary t...

In this paper, we propose a re-configuration method for personal human network and show the information sharing system using it. Personal human network is useful for various utilization of information like information gathering. Since a personal network is usually ad-hoc network, it is necessary to extend and optimize it. Using the neighborhood mat...

In this paper, we propose a reconfiguration method for personal human network and show the information sharing system using it personal human network is useful for various utilization of information like information gathering. Since a personal network is usually an ad-hoc network, it is necessary to extend and optimize it. Using the neighborhood ma...

In this paper, we propose an algorithm called Neighborhood Matchmaker Method to optimize personal human networks. Personal hu- man network is useful for various utilization of information like informa- tion gathering, but it is usually formed locally and often independently. In order to adapt various needs for information utilization, it is nec- es...

In this paper, we describe the outline and the current progress of our project called "Collection and Integration of Concept Systems (CICS)" that aims to support development of seman-tics in the Internet. Since semantics is not always chang-ing, providing fixed ontologies is not stuffiest. Our project aims to support dynamics of semantics by collec...

In this paper, we propose Neighborhood Matchmaker Method as a re-configration method for personal human network. Personal human network is important to realize the Seman- tic web because trust should be obtained and used with the personal network. Since a persnal network is usually ad- hoc network, it is necessary to extend and optimize it. Us- ing...

In this paper, we propose shared topic networks as a model of human network to organize Internet Information, and developed a system called kMedia that can generate shared topic networks by using WWW bookmark files. We also evaluate the system with experiments to know how shared topics network can help uesrs especially to know each other. A shared...