Motohiro Mase

Semantic Wiki Where Human and Agents Collaborate

A Wiki is a collaborative Web page authoring system. Users collaborate to build a Web site by creating and updating Wiki pages through Web browsers. However, conventional Wikis easily lose the consistency of the contents because a number of anonymous users can participate in authoring them. By introducing information agents that understand the. Wiki contents, we can keep the consistency. The agents can automatically update Wiki contents, integrate other Web contents to them, and keep them consistent cooperating with the human users. We propose KawaWiki, which is a semantic Wiki system where human users and information agents can collaborate by utilizing the semantic Web technology. To make agents and users collaborate in authoring Wiki contents, we adopt the RDF as the common representation. It is not easy for novice users to author RDF data, and we introduce KawaWiki templates to generate a Wiki page with RDF data at one time. We also introduce KawaWiki queries to make agents retrieve information efficiently from the Wiki contents. Finally, we introduce an agent description language to specify agents behavior on the Wiki.

Extracting Topic Maps from Web Histories by Clustering with Web Structure and Contents

In this paper, we propose a clustering method to extract topic maps from the Web browsing history. We improve the structure-based hierarchical clustering method using the contents similarity of the pages and the weight by the types of links and the hierarchical difference of the directories in which the pages are located. The topic maps show the topics that user has seen or not in Web browsing and the relationships between the topics. Using the Web browsing history, we experimentally extract the topic map and evaluate it

Extracting topic maps from Web pages by Web link structure and content

We propose a framework to extract topic maps from a set of Web pages. We use the clustering method with the Web pages and extract the topic map prototypes. We introduced the following two points to the existing clustering method: The first is merging only the linked Web pages, thus extracting the underlying relationships between the topics. The second is introducing weighting based on the similarity from the contents of the Web pages and relevance between topics of pages. The relevance is based on the types of links with directories in the Web sites structure and the distance between the directories in which the pages are located. We generate the topic map prototypes by assuming that the clusters are the topics, the edges are the associations, and the Web pages related to the topics are the occurrences from the results of the clustering. Finally, users complete the prototype by labeling the topics and associations and removing the unnecessary items. We incrementally use a userpsilas evaluation of the topic maps to judge whether a Web page is unnecessary or necessary and then reduce the number of unnecessary pages. We use the relevance feedback along with a Support Vector Machine (SVM) to judge the Web pages. For this paper, at the first step, we mounted the proposed clustering method and conducted experiments to evaluate the effectiveness of extracting topic map prototypes. We eventually discussed the effectiveness of our two additional points by evaluating the extracted topic map prototypes.

Natural Language Question and Answer Method for RDF Information Resource

We propose a question and answer method which responses to a natural language question about RDF information resource. This method extracts a question subgraph which represents the question from the entire RDF graph and searches the subgraph for an answer. We evaluate the performance of the method with an RDF information resource which describes our research laboratory.

Guided Input Method for Collaborative Machine Translation Systems

The quality of machine translation depends on the input sentence. Collaborative machine translation systems do not only translate the input sentence into a target language but also back-translate the translation into the input language in a reverse way, and the user repair the input sentence to improve the quality of translation referring to the back translation. However, it is not easy for novice users to repair the input sentence appropriately to be translated correctly. We propose a guided input method that assists the users to compose input sentences by presenting word candidates that may lead to a correct translation. The word candidates are generated from the sentence structure and word database that is created by storing sentences that have been translated correctly. We evaluate the performance of the guided input method depending on the number of input sentences stored in the database. As the number of input sentences increases in the database, the performance of the method improves because the number of correct translations increases and the number of repairs decreases in an experiment of Japanese-Chinese translation. On the other hand, in an experiment of Japanese-English translation, the performance of the method is not so remarkable. The guided input method works better when the users are not familiar with the target language.

Extracting Topic Maps from Web Pages

We propose a framework to extract topic maps from a set of Web pages. We use the clustering method with the Web pages and extract the topic map prototypes. We introduced the following two points to the existing clustering method: The first is merging only the linked Web pages, thus extracting the underlying relationships between the topics. The second is introducing weighting based on similarity from the contents of the Web pages and relevance between topics of pages. The relevance is based on the types of links with directories in Web sites structure and the distance between the directories in which the pages are located. We generate the topic map prototypes from the results of the clustering. Finally, users complete the prototype by labeling the topics and associations and removing the unnecessary items. For this paper, at the first step, we mounted the proposed clustering method and extracted the prototype with the method.

Semiautomatic Extraction of Topic Maps from Web Pages Using Clustering with Web Contents and Structure

In this paper, we describe a method to semi- automatically extract Topic Maps from a set of Web pages. We introduce the following two points to the existing clustering method: The first is merging only the linked Web pages, to extract the underlying relationship of the topics. The second is introducing the similarity by contents of Web pages and the types of links, and the distance between the directories in which the pages are located, to generate dense clusters. We generate the topic map by assuming the clusters as topics, the edges as associations, the Web pages related to the topic as occurrences from the result of clustering. We experimentally extracted the topic map and evaluated it.