Yoshihisa Shinozawa

Proposal of the Support Tool for After-Class Work based on the Online Threaded Bulletin Board

In this paper, based on the assumption that after-class work in an exercise-based course accompanied by group work is done on an online threaded bulletin board system, the authors propose a support tool for the instructors. Specifically, while focusing on the factors that compose a discussion on the online bulletin board, the users who comment, the topics, and the items (keywords) to be discussed, the authors try to visualize the relationships among these factors as network diagrams. The authors also propose indexes, the comment degree and the activation degree, to evaluate communities formed there. Our experiments in which group work was actually implemented with the application of the proposed tool demonstrated that use of the network diagrams and the evaluation indexes served to distinguish the differences between those groups with properly-proceeding discussions and those without such discussions. The authors confirmed that this can enable the instructors to easily discover those students who do not participate in the discussion and groups with sluggish discussions.

A Proposal of an SNS to Support Individual Practices in a Voluntary Community

Widespread popularization of social networking services (SNSs) prompted, for example, a voluntary community such as an orchestra club of a university to use an SNS to support their activities. However, it is not all-purpose and lacks functions to improve members’ individual skills. Appropriate practice is a great help but we hardly find functions to motivate practices by, for instance, mutual evaluation, members’ advice, and creating competing environment. In this paper, we focus on members’ individual practices in an orchestra club and propose a SNS to foster and maintain their motivations to practice based on the analysis result of current conditions. As the first step of the system development, this paper introduces design of a prototype system and the results of a preliminary evaluation.

A Proposal of SNS to Improve Member's Motivation in Voluntary Community Using Gamification

Recently, the number of voluntary communities such as local communities and university club activities are in- creasing. In these communities, since there are various types of members and there are no binding forces, it is usually difficult to maintain and improve members motivation. To maintain and improve members motivation, most of these communities use social networking services (SNSs). However, since existing SNS offer few functions for voluntary community, it is difficult to solve this problem. This research focused on the concept of gami- fication and proposed an SNS to improve members motivation of voluntary community. First, the authors analyzed the current conditions and members of a voluntary community. Based on this analysis, the authors found that an SNS to improve members motivation of voluntary community requires functions which support members personal activities and also functions which increase social activities. Next, the authors built an SNS that had these functions by applying the concept of gamification. The authors implemented the SNS for a University clubs activities for one month and showed the effectiveness of our SNS.

Examination of the effect of social login through analysis of user's purchasing tendency

This study targets social login registrants on an EC site and aims to clarify the difference between the purchasing tendency of social login registrants and general members by analyzing product purchasing history. We focused on the golf portal site that is the subject of this research. We analyzed the purchasing data comparing social login registrants with general members. As the results, it became clear that social login registrants have less resistance to purchasing expensive products on an EC site compared with general members and golf clubs act as a bridge for purchasing.

Linguistic Productivity and Recurrent Neural Networks

Productivity is the defining property of a natural language. Any native speaker of a natural language utters a sentence that has never been heard and understands a sentence that has been heard for the first time. Chomsky claimed that the purpose of linguistics is to account for the productivity of natural languages (Chomsky, 1980). The learnability of a productive language by computational mechanisms is hindered by the inherent nature of the language. For many years, researchers have attempted to devise a learning mechanism by which productive languages can be learnt in a manner similar to that adopted by a child learning from scratch or a student learning a second language. However, there are many problems resisted to be solved. Productivity is one of their causes. This chapter is devoted to efforts undertaken to understand productivity in terms of language learning by means of simple but powerful methods such as neural networks, because neural networks are the simplest (maybe over-simplified) models of our brain mechanism we have obtained thus far. It is natural to expect that a recurrent neural network (RNN) among them is capable of learning languages, specifically a subset of a natural language, because a sentence is a sequence of words and an RNN is capable of learning sequences. The chapter consists of two parts, each of which is devoted to one of the two unmatched features of human languages—the recursive or self-embedding structure of human languages, and the syntactic or combinatorial systematicity of human languages. Both these features constitute the syntactic productivity of human languages.

A Characterization of Simple Recurrent Neural Networks with Two Hidden Units as a Language Recognizer

We give a necessary condition that a simple recurrent neural network with two sigmoidal hidden units to implement a recognizer of the formal language {anbn| n> 0 } which is generated by a set of generating rules {Si¾?aSb, Si¾?ab} and show that by setting parameters so as to conform to the condition we get a recognizer of the language. The condition implies instability of learning process reported in previous studies. The condition also implies, contrary to its success in implementing the recognizer, difficulty of getting a recognizer of more complicated languages.