基于支持向量机的大学生网络信息偶遇影响因素研究

Study of Network Information Encountering Influence Factors for Undergraduate Group Based on Support Vector Machine

[目的／意义]研究网络环境下大学生群体的信息偶遇敏感影响因素，以指导大学生群体提高信息偶遇能力，继而提升大学生信息素养。[方法／过程]使用信息增益分析各影响因素与信息偶遇发生频次之间的相关性，构建敏感影响因素模型，并进一步利用支持向量机（SVM）建立信息偶遇频次预测模型。[结果／结论]与发生信息偶遇最相关的10个影响因素分布于信息用户、偶遇信息、网络环境、情境因素4个维度；模型分类预测精度达82．96％，说明SVM对预测信息偶遇频次有良好效果。

[Purpose/significance ] In the current Web 2.0 network environment, information encountering is one important method to get information for the undergraduate group. This study is of important significance of improving the ability of information encountering and information literacy for university students. [ Method/process ] Aiming at university students, this paper studies the sensitive influence factors of information encountering in the environment of network. Specifically speaking, this paper uses information gain to analyze the correlation between each influence factor and information encountering frequency, and then builds the model of sensitive influence factor. Furthermore, support vector machine （SVM） is introduced to establish the prediction model for information encountering frequency. [ Result/conclusion] There exists 10 most sensitive influence factors for information encountering which are located in four dimensions including information user, encountering information, network environment and situation factors. The predicted classification accuracy can reach 82.96%, which demonstrates SVM works well to predict information encountering frequency.