Search

Contemporary societies face many urban issues. To address these issues, governments, corporations and individuals should disclose and share their related statistical and sensory data. However, existing published data appear in various formats and contain defects. Therefore, few problems have been solved using these data. In this research, we sought to address this problem, by considering a university campus as a microcosm of society, designed data integration schema, and consolidated data into a knowledge graph. We then, applied and modified existing completion methods. In particular, regarding the bicycle environment, we trained our knowledge graph and evaluated it with the conventional method and our proposed derivative method, respectively. Using approximately 650 parking data with various dates and times, our method correctly estimated 54.5 more bicycles than the conventional method by comparing each time's mean absolute error.

Linked Open Data (LOD) is recently attracting attention as a vast amount of distributed knowledge base on the Web. Thus, semi-structured data such as tables and hierarchical data in several domains have been triplified to the LOD. Moreover, triplification of unstructured data such as text and sensor data is actively studied in the research area. We thus developed a Web API for extracting triples from natural sentences, which is useful for the promotion of the LOD. In this paper, we first describe the service specification, its technical background, and the evaluation of the current extraction accuracy, and then introduce several use cases of the service. Although this service adopts a novel combination of a method using ontology-based rules and a machine learning method using Conditional Random Field based on a probability distribution, the main contributions are in the practical aspect, that is, mashup of several natural language processing (NLP) techniques for the text triplification, and its deployment as a public Web API, so that non-experts of NLP can easily use it.

The objective of this paper is to recognize and classify the poses of idols in still images on the web. The poses found in Japanese idol photos are often complicated and their classification is highly challenging. Although advances in computer vision research have made huge contributions to image recognition, it is not enough to estimate human poses accurately. We thus propose a method that refines result of human pose estimation by Pose Guide Ontology (PGO) and a set of energy functions. PGO, which we introduce in this paper, contains useful background knowledge, such as semantic hierarchies and constraints related to the positional relationship between body parts. Energy functions compute the right positions of body parts based on knowledge of the human body. Through experiments, we also refine PGO iteratively for further improvement of classification accuracy. We demonstrate pose classification into 8 classes on a dataset containing 400 idol images on the web. Result of experiments shows the efficiency of PGO and the energy functions; the F-measure of classification is 15% higher than the non-refined results. In addition to this, we confirm the validity of the energy functions.

In many movies, social conditions and awareness of the issues of the times are depicted in any form. Even if fantasy and science fiction are works far from reality, the character relationship does mirror the real world. Therefore, we try to understand social conditions of the real world by analyzing the movie. As a way to analyze the movies, we propose a method of estimating interpersonal relationships of the characters, using a machine learning technique called Markov Logic Network (MLN) from movie script databases on the Web. The MLN is a probabilistic logic network that can describe the relationships between characters, which are not necessarily satisfied on every line. In experiments, we confirmed that our proposed method can estimate favors between the characters in a movie with F-measure of 58.7%. Finally, by comparing the relationships with social indicators, we discussed the relevance of the movies to the real world.

Increasing levels of environmental consciousness have spurred interest in urban agriculture and greening. However, plant cultivation in a limited urban space isn't necessarily a simple matter. On one hand, a plant that's not a hardy species might die; on the other, a plant that overgrows could disrupt the balance of vegetation in the surrounding environment. The authors propose an Android application called Green-Thumb Camera, which allows users to search for plants that fit particular environmental conditions from the linked open data (LOD) cloud. Queries are made using smartphone sensor information, and the application then uses augmented reality to overlay an image of the adult plant in the specified space. The authors describe the LOD content generation method and details of the application and then evaluate the accuracy of the LOD data and the application's usability.

This paper proposes a method of identifying and aggregating literal nodes that have the same meaning in Linked Open Data (LOD) in order to facilitate cross-domain search. LOD has a graph structure in which most nodes are represented by Uniform Resource Identifiers (URIs), and thus LOD sets are connected and searched through different domains.However, 5% of the values are literal values (strings without URI) even in a de facto hub of LOD, DBpedia. In SPARQL Protocol and RDF Query Language (SPARQL) queries, we need to rely on regular expression to match and trace the literal nodes. Therefore, we propose a novel method, in which part of the LOD graph structure is regarded as a block image, and then the matching is calculated by image features of LOD. In experiments, we created about 30,000 literal pairs from a Japanese music category of DBpedia Japanese and Freebase, and confirmed that the proposed method determines literal identity with F-measure of 76.1-85.0%.

Open Data is now collecting attention for innovative service creation, mainly in the area of government, bioscience, and smart X project. However, to promote its application more for consumer services, a search engine for Open Data to know what kind of data are there would be of help. This paper presents a voice assistant which uses Open Data as its knowledge source. It is featured by improvement of accuracy according to the user feedbacks, and acquisition of unregistered data by the user participation. We also show an application to support for a field-work and confirm its effectiveness.

Recently, bicycle-related accidents, e.g., collision accidents at intersection increase and account for approximately 20% of all traffic accidents in Japan; thus, it is regarded as one of the serious social problems. However, the Traffic Accident Occurrence Map released by the Japanese Metropolitan Police Department is currently based on accident information records, and thus there are a number of near-miss events, which are overlooked in the map but will be useful for preventing the possible accidents. Therefore, we detect locations with high possibility of bicycle accidents using user participatory sensing and offer them drivers and government officials as Open Hazard Data (OHD) to prevent future bicycle accident. This paper uses smartphone sensors to obtain data for acceleration, location, and handle rotation information. Then, by classifying those data with convolutional neural networks, it was confirmed that the locations, where sudden braking occurred can be detected with an accuracy of 80%. In addition, we defined an RDF model for OHD that is currently publicly available. In future, we plan to develop applications using OHD, e.g., notifying alerts when users are approaching locations where near-miss events have occurred.

The use of thesauri and taxonomies for science and technology information in scientometrics has been attracting attention. However, manual construction and maintenance of thesauri is expensive and requires significant time, thus, methods for semi-automatic construction and maintenance are being actively studied. We propose a method to expand an existing thesaurus using the abstracts of articles from state-of-the-art technological domains with limited structured information. Specifically, we consider a method for properly allocating new terms to the hierarchical structures of an existing thesaurus using rapidly evolving word embedding. In an experiment, word vectors of 500 degrees are constructed from 567,000 biomedical articles and are clustered after dimension reduction using principal component analysis. Then, semantic relations are estimated based on the spatial relations between the new term and any of the terms in the thesaurus. We then conducted a comparison of the results obtained from three experts. In future, we will develop a recommendation system for new terms related to the existing terms to support semi-automatic thesaurus maintenance.

Illegally parked bicycles are a social problem in Japan and other countries. Illegally parked bicycles obstruct vehicles, cause road accidents, encourage thefts, and disfigure streets. In order to solve the challenge posed by illegally parked bicycles, we realized that it is necessary to collect and republish the data as reusable format. Therefore, we collected the number of illegally parked bicycles, location information, time, and factors. Then, we integrated and republished these data as Linked Open Data (LOD) on the Web. In this paper, we described a schema design of illegally parked bicycles LOD and a methodology of designing LOD schema. Furthermore, we collected data from SNS and website of municipality, and built the LOD of 21,898 triples.

The illegal parking of bicycles has been an urban problem in Tokyo and other urban areas. We have sustainably built a Linked Open Data (LOD) relating to the illegal parking of bicycles (IPBLOD) to support the problem solving by raising social awareness. Also, we have estimated and complemented the temporally missing data to enrich the IPBLOD, which consisted of intermittent social-sensor data. However, there are also spatial missing data where a bicycle might be illegally parked, and it is necessary to estimate those data in order to expand the areas. Thus, we propose and evaluate a method for estimating spatially missing data. Specifically, we find stagnation points using computational fluid dynamics (CFD), and we filter the stagnation points based on popularity stakes that are calculated using Linked Data. As a result, a significant difference in between the baseline and our approach was represented using the chi-square test.

The illegal parking of bicycles is becoming an urban problem in Japan and other countries. We believe the data publication of such urban problems on the Web as Open Data will contribute to solving the problems. However, Open Data sets available for the illegally parked bicycles are coarse and in various formats, and then it is difficult to develop information services using the data. In this study, we thus build an ecosystem that generates Open Urban Data in Link Data format by socially collecting the data, complementing the missing data, and then visualizing the data to facilitate and raise social awareness of the problem. In our experiment, 747 pieces of information on the illegally parked bicycles in Tokyo were collected, and then we estimated the unknown number of the illegally parked bicycles with 64.3i¾?% accuracy. Then, we published the data as the Open Data, and also a web application, which visualizes the distribution of the illegally parked bicycles on a map.

Current crowdsourcing platforms such as Amazon Mechanical Turk provide an attractive solution for processing of high-volume tasks at low cost. However, problems of quality control remain a major concern. In the present work, we developed a private crowdsourcing system(PCSS) running in a intranetwork, that allow us to devise for quality control methods. For quality control, we introduce four worker selection methods: preprocessing filtering, real-time filtering, post-processing filtering, and guess processing filtering. In addition to a basic approach involving initial training or the use of gold standard data, these methods include a novel approach, utilizing collaborative filtering techniques. Furthermore, we collected a large amount of vocabulary data for natural language processing, such as voice recognition and text to speech using PCSS. The quality control methods increased accuracy by 32.4% in collecting vocabulary task. Then, we got 138 thousand vocabulary data. We found that PCSS is a practical system to collect data, and used for three years since 2011.

In recent years, several large-scale knowledge bases (KBs) have been constructed, such as YAGO, DBpedia, and Google Knowledge Graph. Although automatic extractio techniques that extract facts and rules from the Web is necessary for constructing such large-scale KBs, incorporation of noisy, unreliable knowledge cannot be unavoidable. Thus, Google Knowledge Vault assigns extracted knowledge with confidence scores based on consistency with the existing KBs. In this paper, we propose a new approach for associating confidence scores with knowledge based on a large amount of raw data for domains, where there is no existing KB. We first construct knowledge in a specific domain as a semantic network, and then design a probabilistic network, that corresponds to the semantic network. To associate the confidence scores with the semantic network, we train the probabilistic network with a large amount of open data, provided by the Osaka central wholesale market in Japan. We also confirm the validity of the confidence scores with the accuracy of reasoning on the probabilistic network. A semantic network associated with confidence scores, that is, a weighted labeled graph is advantageous not only for reducing the noisy, unreliable knowledge with low confidence, but also for making retrieval results ranking on the KB. In the future, probabilistic reasoning on semantic networks may also be possible.

Experiment with public deployment of the semantic service matchmaker to a UDDI registry for a year is described in this paper. UDDI is a standard registry for Web Services, but, its search functionality has been limited to a keyword search. Therefore, we propose an enhancement of UDDI, called Matchmaker, that supports semantic service capability discovery. Since September 2003, we have deployed the Matchmaker in one of four official UDDI registries operated by NTT-Communications. In this paper, we first introduce the Matchmaker, and illustrate client tools which lower the threshold of use of semantics. Then, we evaluate this experiment by benchmarks in terms of its performance and functionality. Finally, we discuss user requirements obtained by two ways: questionnaire and observation of search behavior.

UDDI is a standard registry for Web services, but so far its search functionality has been limited to keyword search. These authors propose a UDDI enhancement called Matchmaker, which supports semantic service capability discovery. Since September 2003, they have deployed Matchmaker in one of four official UDDI registries operated by NTT-Communications. In this article, they introduce Matchmaker and illustrate client tools that lower the threshold for using semantics in service discovery. Experiments show that Matchmaker search consumes about 5 seconds more time than a typical keyword-based search. The authors also collected user information through questionnaires and by observation of search behavior.

According to the widespread use of Web Services, composition of multiple services is becoming a main issue. BPEL was proposed as a Web Services workflow language in this context. Then, in many cases we need much interactions between a service requester and a BPEL process execution engine. However for the end user under the ubiquitous environment where the service is executed with a cellular phone or PDA, the interaction means simply a burden. The reason is: (1)it is hard to carry out the operation manually, (2)a network connection is unstable, (3)a connection charge is relatively high. In this paper, we propose middle agents between ubiquitous users and Web Services. The agents analyze BPEL, extract the interaction which the ubiquitous users would need, then produce action operators. It makes a plan in order to achieve a service request from the user. Here the planning agent can reduce operations of the user by collecting the necessary information at the beginning which will be inputted during the IP execution. We show the architecture of the agents and an example regarding ATM service, then evaluate the number of inputs, the performance, and the quantity of data transmitted on network. Finally, we consider adaptability of this approach to generic BPEL flow.

In many movies, cultures, social conditions, and awareness of the issues of the times are depicted in any form. Even if fantasy and SF are works far from reality, the stories do mirror the real world. Therefore, we assumed to be able to understand social conditions and cultures of the real world by analyzing the movie. As a way to analyze the film, we decided to estimate the interpersonal relationships between the characters in the movies. In this paper, we propose a method of estimating interpersonal relationships of the characters using Markov Logic Network from movie script databases on the Web. Markov Logic Network is a probabilistic logic network that can describe the relationships between characters, which are not necessarily satisfied on every occasion. In experiments, we confirmed that our proposed method can estimate favors between the characters in a movie with a precision of 64.2%. Finally, by comparing the estimated relationships with social indicators, we discussed the relevance of the movie to the real world.

A wide variety of mechanical parts are used as products in the area of manufacturing. The code systems of product information are necessary for realizing Electronic Data Interchange (EDI) of business-to-business. However, each code systems are designed and maintained by different industry associations. Thus, we built an industrial parts Linked Open Data (LOD), which we called “N-ken LOD” based on a screw product code system (N-ken Code) maintained by Osaka fasteners cooperative association (Daibyokyo). In this paper, we first describe building of N-ken LOD, then how we linked it to external datasets like DBpedia, and built product supplier relations in order to support the EDI.

Linked Open Data (LOD) has a graph structure in which nodes are represented by Uniform Resource Identifiers (URIs), and thus LOD sets are connected and searched through different domains. In fact, however, 5% of the values are literal (string without URI) even in DBpedia, which is a de facto hub of LOD. Since the literal becomes a terminal node, and we need to rely on regular expression matching, we cannot trace the links in the LOD graphs during searches. Therefore, this paper proposes a method of identifying and aggregating literal nodes that have the same meaning in order to facilitate cross-domain search through links in LOD. The novelty of our method is that part of the LOD graph structure is regarded as a block image, and then image features of LOD are extracted. In experiments, we created about 30,000 literal pairs from a Japanese music category of DBpedia Japanese and Freebase, and confirmed that the proposed method correctly determines literal identity with F-measure of 99%.

In the agricultural sector, the improvement of productivity and quality with respect to such attributes as safety, security and taste has been required in recent years. We aim to contribute to such improvement through the application of Information and Communication Technology (ICT). In this paper, we first propose a model of agricultural knowledge by Linked Open Data (LOD) with a view to establishing an open standard for agricultural data, allowing flexible schemas based on ontology alignment. We also present a semi-automatic mechanism that we developed to extract agricultural knowledge from the Web, which involves a bootstrapping method and dependency parsing, and confirmed a certain degree of accuracy. Moreover, we present a voice-controlled question-answering system that we developed for the LOD using triplification of query sentences and graph pattern matching of the triples. Finally, we confirm through a use case that users can obtain the necessary knowledge for several problems encountered in the agricultural workplace.

Recently, urban greening and agriculture have been receiving increased attention, but the cultivation of greenery is not a simple matter in a restricted space. Therefore, we propose Flower Voice, a voice assistant for smartphones that answers questions users will be faced on site, and provides a mechanism for registering the work. The system uses Linked Open Data as a knowledge source, and features improvemnt of accuracy based on user feedback and acquisitions of new data by user participation. This paper presents Plant LOD and an architecture of the sytem, and then evaluates the improvement of the accuracy.

Handling a large amount of constraints and irregular cases are typical concerns of embedded software. It is the well-known problem that such concerns make embedded software complicated. Aspect oriented technologies are expected to give a solution to this problem. The advantage of these technologies are encapsulating cross cutting concerns. There has been a significant amount of research. However, these aspect-oriented modelling or programming languages have not been disseminated to embedded software development. In order to overcome this dissemination problem, the article presents an aspect-oriented state machine language. The problem is caused by hardware dependent codes, large amount of legacy codes and difficulty of education. To reduce the difficulty of the problems, the language is based on the C-language with extension of state machine concept. The weaving mechanism by mark is making clear multi-dimensional separation of concern. To evaluate the language, we will apply it into the parts for collecting data to acquire positions and communicating by I2C on a toy-airship software.

Semantic search is intrinsically suited for information retrieval in field research, where keyword search is burdensome for users. Also, exploitation of mobile and facility sensors is now prevailing, but applications are still vague. Therefore, LOD can serve as an intermediary interpreting the semantics of the users and their environmental information obtained by the sensors and connecting it to the collective intelligence on the net. In this paper, to build ecosystem of LOD used in the field, we propose a mechanism of LOD content generation and its field application. Firstly, we introduce an Android application, which searches a plant fitting the environmental conditions obtained by the sensors from Plant LOD. Then, we show that our LOD generation mechanism builds the Plant LOD from the Web with an accuracy of 85 to 97%.

Recently, it has been expected to realize privacy preserving data mining in order to acquire valuable knowledge from the combined information sources of several service providers. Therefore researches have been conducted on a distributed anonymization method, which combines the personal information and anonymize it to prevent identifying specific user records. However, in those researches, when sets of the users in the providers are not the same, there is a problem that users' presence in either provider may be revealed. Thus, this paper proposes a new indicator which represents the probability of the presence of users being revealed and introduces a modified distributed anonymization method to satisfy the proposed indicator. Also, we use U.S. census data for evaluation and calculate the relative error of its anonymized data. The results show that it is almost 10-25% in specific cases.

Service providers collect user's personal information relevant to their businesses. Personal information stored by different service providers is expected to be combined to make new services. However, specific user records risk being identified from the combined personal information, and the user's sensitive information may be revealed. Also, personal information collected by a service provider must not be disclosed to other service providers because of security issues. Thus, several researchers have been investigating distributed anonymization protocols, which combine the personal information stored by the providers and sanitize it to ensure an anonymity policy with minimum disclosure. However, when providers have different sets of the users, there is a problem that the existence of users in either service provider may be revealed. This paper introduces a new notion, δ-max-site-presence, which indicates the probability of the existence of users being revealed in a distributed environment and a new distributed anonymization protocol for hiding the existence of users. Our evaluation results show that the proposed protocol can anonymize users in accordance with the policy of hiding their existence and user anonymity without too much information loss.

In this paper, we define an activity by five basic attributes: actor, action, object, time and location. The goal of this paper is to describe a method to automatically extract all attributes in each sentence retrieved from Japanese weblogs. Previous work had some limitations, such as high setup cost, inability of extracting all attributes, limitation on the types of sentences that can be handled, and insufficient consideration of interdependency among attributes. To resolve these problems, this paper proposes a novel approach that uses conditional random fields and self-supervised learning. This approach treats the activity extraction as a sequence labeling problem, and has advantages such as domain-independence, scalability, and does not require any hand-tagged data. Since it is unnecessary to fix the positions and the number of the attributes in activity sentences, this approach can extract all attributes by making only a single pass over its corpus. Additionally, by converting to simpler sentences, the proposed approach can deal with complex sentences retrieved from Japanese weblogs. In an experiment, this approach achieves high precision (activity: 88.87%, attributes: over 90%).

In the web research, ontologies are widely used these days. However, several difficulties in ontology building have been pointed out so far and its costs are raising problems currently. In this paper, we propose ONTOMO that enables Internet users to take part in building ontologies as a part of collective intelligence. After introducing an overview of ONTOMO, an instance and property recommendation mechanism based on the editing history of multiple users will be presented together with experimental evaluations.

The goal of this paper is to describe a method to automatically extract all basic attributes namely actor, action, object, time and location which belong to an activity, and the transition between activities in each sentence retrieved from Japanese CGM (consumer generated media). Previous work had some limitations, such as high setup cost, inability of extracting all attributes, limitation on the types of sentences that can be handled, and insufficient consideration of interdependency among attributes. To resolve these problems, this paper proposes a novel approach that treats the activity extraction as a sequence labeling problem, and automatically makes its own training data. This approach has advantages such as domain-independence, scalability, and unnecessary hand-tagged data. Since it is unnecessary to fix the positions and the number of the attributes in activity sentences, this approach can extract all attributes and transitions between activities by making only a single pass over its corpus.

In our definition, human activity can be expressed by five basic attributes: actor, action, object, time and location. The goal of this paper is describe a method to automatically extract all of the basic attributes and the transition between activities derived from sentences in Japanese web pages. However, previous work had some limitations, such as high setup costs, inability to extract all attributes, limitation on the types of sentences that can be handled, and insufficient consideration interdependency among attributes. To resolve these problems, this paper proposes a novel approach that uses conditional random fields and self-supervised learning. This approach treats activity extraction as a sequence labeling problem, and has advantages such as domain-independence, scalability, and does not require any human input. In an experiment, this approach achieves high precision (activity: 88.9%, attributes: over 90%, transition: 87.5%).

The aging of society is emerging as a global problem. Participation of elderly people in social activities is highly desirable in several perspectives of the society. Thus, we propose a mobile service that supports elderly people wishing to go out and promotes their participation in social activities. This system collects information on the existence of nearby barriers and provides timely information to users via a mobile phone equipped with GPS. The nearby barriers include not only permanent barriers such as stairs, but also temporary ones such as crowds and parked bicycles. Further, we propose functionality that selects and informs barrier information suitable for each user. This paper illustrates the public experiment that we conducted in Tokyo, and confirms the accuracy of the information filtering.

Consumer Generated Media (CGM) has a significant impact on on companies' product marketing strategies. This paper illustrates development of opinion analysis service for marketing research, WOM Scouter. Then we present an algorithm of associated topic extraction, which is one of main features in WOM Scouter. Associated topic extraction finds out competitive products from blog entries commenting on a specified product. The main feature is to apply product ontology in addition to natural language processing. By looking up a term on product ontology, the product domain is identified in blog entries, and general nouns are excluded. Another feature is to evaluate an importance of each product by means of two kinds of smoothing functions based on link popularity and document frequency. The experimental evaluation shows that the proposed algorithm is closer to blog readers' impression than TF-IDF.

In this paper, we introduce a service where ontology summarizes blogs to get useful in the real stores In ubiquitous computing environment, it would be desired for users to bind their real world situation and useful information on the Internet However, the current typical device for ubiquitous computing like a cellular phone has a small display, limited operations, and narrow-band network Therefore, semantics use to extract only the necessary information and services is for the ubiquitous computing Ubiquitous Metadata Scouter is for the user to scan products barcodes by cameras of cellular phones It gets the corresponding metadata to the product from the Internet, and collect the related blogs Then, it analyzes the contents of each blog referring ontologies, and indicates the total reputation Also, it shows other related products which are much talked about This paper illustrates each function of this service and our public experiment at the real consumer electronics store and book store in Tokyo since March 2006 It would provide an instant benefit as a semantics usecase in the ubiquitous computing.

Web services is already an infrastructure for practical e-Business. At the same time, mobile devices such as smart phones and PDAs are widely used by end-users these days. In this paper, we focus on interactions between Web services and the mobile devices during execution of a service flow. Frequent interactions decrease the usability of the service flow for the mobile users, and increase the total processing time and data amount. Thus, we propose an user agent who executes the service flow instead of the users. We first illustrate how the agent is automatically created for the service flow, then evaluate it on the user interaction, the processing time and the data amount. Finally we discuss its applicability to a standard Web services flow language, BPEL.

Metadata have been already given to most of the data and objects in the real world, such as books, foods, digital contents like movie, electric devices, and so forth. Further, they can be accumulated electronically by barcodes and RFIDs, which is expected to spread explosively in 2005. On the other hand, web services are getting popular in the internet, and UPnP services and ECHONET are penetrating into the home network. In our project, we propose a new handheld application called Ubiquitous Service Finder, in which user can intuitively browse as icons the metadata around him/her in a cellular phone, then invoke the services semantically related to the metadata by simple drag and drop operation.

Integrating web sites to provide more efficient services is a very promising way in the Internet. For example searching house for rent based on train system or preparing a holiday with several constrains such as hotel, air ticket, etc... From resource view point, current web sites in the Internet already provide quite enough information. However, the challenge is these web sites just provide information but do not support any mechanism to exchange them. As a consequence, it is very often that a human user has to take the role to "link" several web sites by browsing each one and get the concrete information. The reason comes from a historical objective. Web sites were developed for human users browsing and so, they do not support any machine-understandable mechanism.Current researches in WWW environment already propose several solutions to make newly web sites become understandable to other web sites so that they can be integrated. However, the question is how to integrate existing web sites to these new one. Evidently, redeveloping all of them is an unacceptable solution. In this paper, we propose a solution of Web Service Gateway to "wrap" existing web sites in Web services. Thus, without any efforts to duplicate the Web sites code, these services inherit all features from the sites while can be enriched with other Web service features like UDDI publishing, semantic describing, etc. This proposal was developed in Toshiba with Web Service Gateway and Wrapper Generator System. By using these systems, several integrated-applications were built and they are also presented and evaluated in this paper.

Public deployment of the semantic service matchmaker to a UDDI registry for half a year is shown in this paper. UDDI is a standard registry for Web Services, but if we consider it a search engine, its functionality is limited to a keyword search. Therefore, the Matchmaker was developed to enhance UDDI by service capability matching. Then, we have deployed the Matchmaker to one of four official UDDI registries operated by NTT-Communications since September 2003. In this paper, we first introduce the Matchmaker with UDDI, and illustrate client tools which lower the threshold of use of semantics. Then, we evaluate this deployment by benchmarks in terms of performance and functionality. Finally, we discuss user requirements obtained by two ways: questionnaire and observation of search behaviour.

The public experiment of the semantic service search with the public UDDI registry is shown in this paper. The UDDI is a standard registry for Web Services, but if we consider it as a search engine, the functionality is restrictive, that is, based on keyword retrieval only. Therefore, the Matchmaker was developed to enhance the UDDI search functionality by using semantics such as ontology and constraints. However, for Web Services as e-Business platform, compliance with the standard specification such as SOAP, WSDL, UDDI is the key. Thus, our goal of this experiment is to seamlessly combine semantic search with those standards, and investigate the feasibility of using semantics with Web Services. This paper firstly shows the overall architecture where the Matchmaker can be located between UDDI and service developers/users. Then, the Matchmaker and our semantic service description which can be complementary used with the standard WSDL and UDDI Data Structure are shown. Also, we illustrate some tools which generate the service description and support the use of the Matchmaker. Finally, we qualitatively evaluate this experiment on the response from business users.

We propose a new concept called Rental Service. Recently, Rental Application has been applied as ASP (Application Service Provider). For the next step in advancing ASP, we would prefer to provide Service which creates new functions, coordinating and facilitating already-existing functions, provided by many applications in the network. In this paper, we firstly made a list of problems for adding such service into the current ASP systems. Then, we developed a framework as a solution to the problems, in which the service itself is regarded as an agent. Moreover, we introduced a practical system developed by the framework, and finally examined its effectiveness for the above problems through that practical case.

The web is moving from being a collection of pages toward a collection of services that interoperate through the Internet. A fundamental step toward this interoperation is the ability of automatically locating services on the bases of the functionalities that they provide. Such a functionality would allow services to locate each other and automatically interoperate. Location of web services is inherently a semantic problem, because it has to abstract from the superficial differences between representations of the services provided, and the services requested to recognize semantic similarities between the two.Current Web Services technology based on UDDI and WSDL does not make any use of semantic information and therefore fails to address the problem of matching between capabilities of services and allowing service location on the bases of what functionalities are sought, failing therefore to address the problem of locating web services. Nevertheless, previous work within DAML-S, a DAML-based language for service description, shows how ontological information collected through the semantic web can be used to match service capabilities. This work expands on previous work by showing how DAML-S Service Profiles, that describe service capabilities within DAML-S, can be mapped into UDDI records providing therefore a way to record semantic information within UDDI records. Furthermore we show how this encoded information can be used within the UDDI registry to perform semantic matching.

The goal of this paper is to describe a method to automatically extract all basic attributes namely actor, action, object, time and location which belong to an activity from Japanese weblogs. Sentences retrieved from weblogs are often diversified, complex, syntactically wrong, have emoticons and new words. There are some works that have tried to extract users' activities in sentences retrieved from web and weblogs. However, these works have several limitations, such as inability of extracting infrequent activities, high setup cost, limitation on the types of sentences that can be handled, necessary of preparing a list of object and action. To resolve these problems, we propose a novel approach that treats the activity extraction as a sequence labelling problem, and automatically makes its own training data. This approach can extract infrequent activities, and has advantages such as scalability, and unnecessary any hand-tagged data. Since it does not require to fix the positions and the number of the attributes in activity sentences, this approach can extract all attributes, with high recall.

In this paper, we introduce a mobile service that extracts reputations of a product from weblogs by cellular phones during shopping. In a mobile computing environment, it would be desirable for users to bind their real-world situation and useful information on the internet. However, the current typical devices for mobile computing, such as cellular phones, have small displays, limited operations, and narrow-band network. Thus, semantics can be useful in order to extract only the necessary information depending on the users' situation. Word-Of-Mouth (WOM) Scouter is an application for this purpose. If the user takes a photo of a product barcode on the package with a cellular phone camera, WOM Scouter first gets the product metadata (name, manufacturer, etc.) from the internet and collects blogs that review the product. Also, it analyses the blog contents with NLP techniques and ontologies. Then, it indicates the overall reputation (positive or negative), and other related products that are the subject of much discussion in the blogs. This paper illustrates each function of this service and a public experiment and evaluation at a real consumer electronics store and book-store in Tokyo in March 2006. This is a case of semantics use to provide an instant benefit in a mobile computing environment.

The Web is moving from being a collection of pages toward a collection of services that interoperate through the Internet. The first step towards this interoperation is the location of other services that can help towards the solution of a problem. In this paper we claim that location of web services should be based on the semantic match between a declaritive description of the service being sought, and a description of the service being offered. Furthermore, we claim that this match is outside the representation capabilities of registries such as UDDI and languages such as WSDL. We propose a solution based on DAML-S, a DAML-based language for service description, and we show how service capabilities are presented in the Profile section of a DAML-S description and how a semantic match between advertisements and requests is performed.