Research Of Interest Search Mechanism Of SNS Learning Communities

With the popularization and development of the Web2.0technology, the Internet has entered the era of SNS at full speed. Meanwhile, great changes have happened in the educational field and more chances of communication and interaction are needed between teachers and students. In this background, a new kind of network learning environment--SNS learning community arises. The prime advantage of the community is that the teachers and students can freely communicate and interact with each other. In this community, the important criterion of measuring the learners’ learning efficiency is the efficiency of searching required resources. It will raise the learners’ learning efficiency if the learners can search the required resources in the first time.Compared to the traditional and the common network teaching modes, SNS learning community is no longer dominated by teachers. On the contrary, students work as the center of the whole teaching process. Meanwhile, besides "teaching" alone, it pays more attention to the students’"learning" ability. Certainly, how to improve students’learning efficiency is an important problem that can not be ignored in this community. In this paper, firstly, it deeply analyzes the efficiency of searching required resources, and then, combined with the characteristics of SNS learning community, reasonably improves the existing similar interests finding algorithm to put forward a way that the learners search effective resources based on their own interest. The technical route of this paper is that we make preprocessing according to the learners’ documents to get the interest vector, and then calculate the interest similarity to find corresponding interest domain, and finally, students can search resources in the improved hybrid network topology based on their interest.This paper improves the algorithm of learners finding others with similar interest. Firstly, we need to set a interest vector for each learner. The content of the vector can be acquired by processing their own documents according to the fuzzy set theory and the topic model technology. Then, we calculate the relevant degree of interest between learners by interest vector. We select the learners with similar interest to make up the interest domain. Finally, they can search for similar interests through hybrid network topology. This paper divides the searching process of the learners similar interests into three parts. We realize learners interest searching in SNS learning community by carrying out system designation in JXTA platform. Ultimately, we make four groups of simulation experiments by using Matlab. The experimental results show that the proposed improved method provides higher efficiency in learners searching needed resources, and also has higher performance in the scope and the success rate of searching.