Self-supervised capturing of users' activities from weblogs. (1st January 2012)

• Record Type: Journal Article
• Authors: Nguyen, The-Minh; Kawamura, Takahiro; Tahara, Yasuyuki; Ohsuga, Akihiko
• Is Part Of: International journal of intelligent information and database systems. Volume 6:Number 1(2012)
• Page Start: 61
• Page End: 76
• Publication Date: 2012-01-01
• Subjects: human activity -- semantic network -- weblogs mining -- conditional random fields -- self-supervised learning -- CRFs; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

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.