摘要
成功的网络小说IP凭借庞大的受众群体,拥有巨大的商业价值。准确预测网络小说隐含潜在的价值量,掌握行业潜在规律对指导网络小说运营商开展运营计划具有很重要的指导作用。选择总推荐量等指标运用八爪鱼在起点网收集数据,使用WEKA平台构建M5模型树预测模型,对网络小说的隐含价值量进行预测,并与传统线性回归和梯度下降分类树这两种算法预测结果进行了对比。结果发现网络小说总推荐量、总字数与网络小说的点击量成负相关,并针对此种现象进行了分析和探讨。
Intellectual property of successful online novels has great commercial value since it can attract large number of readers.Thus,it is significant for operators to understand the potential industry rules,and forecast the hidden value of online novels,which can help them to make the appropriate commercial plans.In this paper,we choose the total recommended.quantity and construct M5 model tree based on WEKA platform to forecast the hidden value of online novels,and then compare the accuracy of the model with traditional linear regression and gradient descent classification tree.The results demonstrate that the total recommended quantity of online novels is negatively correlated to total numbers of words and click rates.Detailed explanation is given in this paper.
作者
赵礼强
姜崇
唐金环
ZHAO Li-qiang;JIANG Chong;TANG Jin-huan(College of Economics and Management,Shenyang Aerospace University,Shenyang 110136,China)
出处
《沈阳航空航天大学学报》
2018年第1期66-73,共8页
Journal of Shenyang Aerospace University
基金
国家自然科学基金(项目编号:71702112)
教育部人文社科规划基金(项目编号:17YJA630139)
辽宁省教育厅重点项目(项目编号:L201706)
