Harumi Murakami

A system for generating user's chronological interest space from web browsing history

We propose a method that helps users to understand their own interests by extracting terms from selected link texts and generating a new browsing history, and arranging those terms and Web-page icons on to the users interest space in chronological order. We have implemented a prototype system based on this method. The systems performance was evaluated in two experiments, which revealed that (1) Japanese interest terms can be extracted from the users selected link texts, (2) the interest-space browser displays the users interest space on browsing the Web, and (3) the users can use this system to better understand their own interests, recall their past, and reorganize previously browsed Web pages.

Assigning Vocation-Related Information to Person Clusters for Web People Search Results

Distinguishing people with identical names is becoming more and more important in searches on the Web. This research assigns useful labels to help users select person clusters that are separated into different people from the result of person searches on the Web. We proposed a method to label person clusters with vocation-related information (VRI). VRI includes broader terms that may not be considered vocations as well as terms that are useful to infer vocations, not only those rigorously defined as vocations. Our method is comprised of two processes: (a) extraction of VRI candidates using HTML structure and heuristics, and (b) VRI generation using term frequency, clustering synonyms, and calculation using a Web search engine. Experimental results revealed the usefulness of our proposed method.

Assigning Location Information to Display Individuals on a Map for Web People Search Results

Distinguishing people with identical names is becoming more and more important in Web search. This research aims to display person icons on a map to help users select person clusters that are separated into different people from the result of person searches on the Web. We propose a method to assign person clusters with one piece of location information. Our method is comprised of two processes: (a) extracting location candidates from Web pages and (b) assigning location information using a local search engine. Our main idea exploits search engine rankings and character distance to obtain good location information among location candidates. Experimental results revealed the usefulness of our proposed method. We also show a developed prototype system.

Supporting Human Recollection of the Impressive Events using the Number of Photos

We present a system to support human recollection with tag clouds, which are created from keywords generated by our algorithms from the use information of Google Calendar and Twitter. The main feature of our research is to weight words using the number of photos taken by users to recall impressive events. We evaluated tag clouds by comparing our approach and a comparative approach, and our experiment results suggest the usefulness of our approach.

People search using NDC classification system

To help users select and understand people during searches for them, we present a method of assigning Nippon Decimal Classification (NDC), which is a system of library classification numbers, to people on the Web. By assigning NDC numbers to people, we can assign not only labels to people but also build a NDC-based people directory. We developed a prototype based on this approach.

Creating user's knowledge space from various information usages to support human recollection

To support human recollection, we present a data integration method using a simple information structure called a history structure, which is constructed from time, keywords, and URI sets. We also present algorithms that generate history structures from such information usages as Web searches, twitter, e-mails, calendars, and book purchases and create a user knowledge space. Based on our approach, we developed a system called a knowledge-space browser and evaluated whether it can help users recall a particular day by summarizing that days history structure. Experimental results reveal the usefulness of our approach and our implemented system.

Constructing and sharing everyday memory for community knowledge co-evolution

The authors describe a method for supporting the process of accumulating personal everyday memory and sharing it in a community. Two ideas are used to implement the method. The first idea is intimate computing which enables people to externalize and accumulate personal memory using associative representation at any time and any place. The second idea is a talking virtualized egos metaphor that visualizes community knowledge interaction. This helps community members to find out common interests and shared knowledge. Based on these ideas, we developed the system called CoMeMo which helps people to manage personal and community memory. In experimentation, we observed the following: (a) collaborative knowledge building, (b) change of shared information, (c) creation of new human relationships.

Assigning Library Classification Numbers to People on the Web

To help users select and understand people during searches for them, we present a method of assigning Nippon Decimal Classification (NDC), which is a system of library classification numbers, to people on the web. By assigning NDC numbers to people, we can assign not only labels to people but also build a NDC-based people-search directory. We use a relative index in NDC, which lists the related index terms attached to NDC. We developed a prototype based on this approach. We evaluated the usefulness of our proposed method and directory and found that extracting relative index terms from the titles of web pages outperformed comparative methods.

Generating Wikipedia-Like Biographical Sentences from Web People Search Results

To help users understand and select people on the web, we developed a method of generating biographical sentences for the results of web people search. We extracted attribute information about people (birth date, death date, birthplace, vocation, organization, and position) from the HTML files of person clusters that were manually classified into individuals and generated biographical sentences whose style resembles the first sentence of Wikipedia. We evaluated the method using 20 queries (person names) * 50 web search results. Our experimental results show that it is easy to generate Wikipedia-like biographical sentences of people using (1) a reasonable Wikipedia-like template and (2) a simple generating algorithm that combines extracted attribute information and predefined words.

Presenting Mathematical Expression Images on Web to Support Mathematics Understanding

People cannot use a text search to find mathematical expressions because expressions cannot be replaced with words. Our research uses an ordinary text search and presents appropriate mathematical expression images (hereinafter called math-images) for input keywords. First we classify a set of the top ranking images from all the images in HTML files by scoring them. We focus on three viewpoints that are unique to mathematical expression images and mark the images by using these viewpoints. Then by adding bonus points to these marked images, the best three images are chosen from the set and presented with an explanation of the keyword and the surrounding information in the HTML files. We conducted two experiments to optimize the parameters of the expression giving the mark and to evaluate the effect of the bonus points. The rate of the average correct images of the best three was 79.5%.