摘要
网络广告作为新兴的广告产业正在进行快速发展的发展,内容定向广告是近几年研究的主要方向,首先分析了网页广告的特征,针对内容定向的投放算法进行了研究,通过基于VSM的TF-IDF方法发现了网页广告文本之间的不足,引入正则表达式进行网页广告和文本的匹配,通过采用树形结构进行索引和过滤提高网页和文本的匹配率,最后通过改进的BM25算法提高了网页广告文本中的检索率。通过一定数量的网页内容和广告文本实验,表明本文的算法具有很好的有效性,提高了网页内容和广告文本的匹配率。
As an emerging advertising industry,online advertisements are developing rapidly and contenttargeted advertising is the main direction of research in recent years.This paper first analyzes features ofweb advertising,studies the algorithm of distributing content targeted advertisements,and findsdeficiencies of online advertisements’text through the VSM-based TF-IDF method.Then,regularexpression is introduced to match web advertisements and texts,tree structure is adopted for indexingand filtering to improve the matching rate of webs and texts,and finally improved BM25algorithm isadopted to improve the retrieval rate of texts in web advertisements.The experiments adopt certainamount of web contents and ad texts to indicate that algorithm in this paper is rather effective,improvingthe matching rate of web contents and ad texts.
作者
蔡志荣
Cai Zhirong(Shaoxing Vocational & Technical College,Zhejiang Shaoxing,312000,China)
出处
《科技通报》
北大核心
2017年第7期94-98,共5页
Bulletin of Science and Technology
基金
浙江省高校访问学者课题
关键词
网页广告
投放
正则表达式
BM25算法
web advertisement
distribution
regular expression
BM25 algorithm