文献计量学系列33: 关键词时间分布规律

导读

本期我们将介绍关键词随时间的变化。bibliometrix包中有一个时间相关的统计函数keywordGrowth。该函数可以统计每年关键词的出现的次数（词频）。很多与时间相关的分析均可从该函数开始。

<<<<<<<<<课程推荐>>>>>>>>>

R文献计量学基础篇已上线！

内容涵盖文档、作者、期刊、研究机构和国家等相关文献计量学指标分析

更多自定义函数

一次性获取较多文献计量指标

让学习更轻松！

学习力，才是最大的竞争力！扫码约我吧！

<

一、keywordGrowth函数简介

keywordGrowth函数用于计算每年关键词或术语的词频次数或累加词频次数，结果返回一个数据框。

二、加载包和数据导入

pacman::p_load(bibliometrix, rio, tidyverse, Hmisc)m1_TE <- import(file = 'E:/精鼎统计/m1_TE.xlsx')rownames(m1_TE) <- m1_TE$SR

三、关键词描述统计

关键词词频：

kwg <- KeywordGrowth(m1_TE, Tag = 'DE_TM', sep = ';', top = 10, cdf = FALSE)#提取累加词频排名前10的关键词的词频head(kwg)# Year CATCHMENT STABLE-ISOTOPE RUNOFF GROUNDWATER PRECIPITATION RUNOFF-GENERATION MODEL TRACER BASIN# 1 1991 1 0 1 0 0 2 1 1 0# 2 1992 1 0 2 0 0 0 0 0 0# 3 1993 3 0 3 4 0 1 1 0 1# 4 1994 2 0 1 1 1 2 2 0 0# 5 1995 2 0 3 2 1 2 1 0 1# 6 1996 3 0 2 3 0 1 1 2 0# RIVER# 1 0# 2 0# 3 0# 4 0# 5 0# 6 0#figurekwggather <- gather(kwg, key = 'keywords', value = 'Freq', -Year)#宽数据框转长数据框kwggather$Freq[which(kwggather$Freq == 0)] <- NA#频率为0变空值NAkwggather$Freq_min <- ifelse(kwggather$Freq >= 5, kwggather$Freq, NA)#最小展示频率kwggather$year1 <- ifelse(is.na(kwggather$Freq), NA, kwggather$Year)#添加线的x轴值kwggather$keywords <- factor(kwggather$keywords,levels = names(kwg)[-1])#字符格式转因子格式kwgth <- ggplot(kwggather, aes(x = Year, y = keywords))+ geom_line(aes(x = year1, y = keywords, group = keywords), size = 0.8, color="firebrick", alpha = 0.3)+ geom_point(aes(size = Freq),color = "dodgerblue4", alpha = 0.5)+ geom_text(aes(label = Freq_min), size = 3)+ scale_y_discrete(limits = rev(levels(kwggather$keywords)), labels = rev(unique(capitalize(tolower((kwggather$keywords))))))+ scale_x_continuous(limits = c(1991,2019),breaks = seq(1991,2019,1))+ labs(x = '', y = '', size = 'Frequency')+ theme_bw()+ theme(panel.grid = element_blank(), axis.text = element_text(size = 12), axis.text.x = element_text(angle = 90, vjust = 0.4), legend.text = element_text(size = 14), legend.title = element_text(size = 20))+ scale_size_continuous(breaks = seq(5,35,5))kwgth

关键词累加词频：

kwg1 <- KeywordGrowth(m1_TE, Tag = 'DE_TM', sep = ';', top = 10, cdf = TRUE)kwggather1 <- gather(kwg1, key = 'keywords', value = 'cumFreq', -Year)kwggather1$keywords <- capitalize(tolower(kwggather1$keywords))kwggather1$keywords <- factor(kwggather1$keywords,levels = capitalize(tolower(names(kwg1)))[-1])
kwgth1 <- ggplot(kwggather1, aes(x = Year, y = cumFreq, color = keywords))+ geom_line()+ scale_x_continuous(limits = c(1991,2019),breaks = seq(1991,2019,1))+ labs(x = '', y = 'Accumulative Frequency')+ theme(axis.title = element_text(size = 14), axis.text = element_text(size = 12), axis.text.x = element_text(angle = 90, vjust = 0.4), legend.text = element_text(size = 14), legend.title = element_text(size = 20))kwgth1


四、小结

keywordGrowth函数除了对关键词的进行分析外，还可以对其他的字段标识进行分析，比如作者（AU），国家（AU_CO）等，感兴趣的同学自己可以试一试。

---