Takashi Yukawa

An expert recommendation system using concept-based relevance discernment

An expert recommendation system using concept-based relevance discernment is proposed. This system processes the description of a technical topic as input and then finds engineers who have a high level of expertise in that area. The technique employed is an extended vector space model that locates both technical topics and engineers in the same multi-dimensional space, and then calculates their relevance. This system can also retrieve engineers or documents that are related to afield matching a given engineers technical interests. Such a system can be expected to play the role of a persons professional network, and be a valuable tool for knowledge management among several organizations.

The Prefiltering Techniques in Emotion Based Place Recommendation Derived by User Reviews

Context-aware recommendation systems attempt to address the challenge of identifying products or items that have the greatest chance of meeting user requirements by adapting to current contextual information. Many such systems have been developed in domains such as movies, books, and music, and emotion is a contextual parameter that has already been used in those fields. This paper focuses on the use of emotion as a contextual parameter in a tourist destination recommendation system. We developed a new corpus that incorporates the emotion parameter by employing semantic analysis techniques for destination recommendation. We review the effectiveness of incorporating emotion in a recommendation process using prefiltering techniques and show that the use of emotion as a contextual parameter for location recommendation in conjunction with collaborative filtering increases user satisfaction.

Securing Information Sharing Through User Security Behavioral Profiling

This paper presents a method of minimizing the human-related information security problem of improper sharing of information by insiders with outsiders or unauthorized insiders. As opposed to most currently available information security solutions, this system does not rely solely on technological security measures, but adapts a mixture of social and technological solutions. The system presented through this research detects users’ observance of security best practices and behavioral patterns using both automatic and personal monitoring methods. It then creates user security behavioral profiles and thus identifies users who might potentially pose threats to the organization’s information security and determines and schedules the level and type of security education and training to be given to identified users.

A Query Classification System Based on Snippet Similarity for a One-Click Search

This paper proposes a query classification system for a one-click search system that uses feature vectors based on snippet similarity. The proposed system targets the NTCIR-10 1CLICK-2 query classification subtask and classifies queries in Japanese and English into eight predefined classes by using support vector machines (SVMs). In the NTCIR-9 and NTCIR-10 tasks, most participants used complex features or rules that depend strongly on language characteristics. The authors propose a new method that uses feature vectors created by using snippet similarities instead of the above mentioned features. In the proposed method, feature vectors have fewer dimensions, provide better generalization, lower language dependency, and reduced computer resources. This method achieved accuracies of 0.93 for a Japanese task and 0.91 for an English task.