Bibliographic Collection

The dataset consists of bibliographic records retrieved from Scopus on April 22, 2024, using the keywords biped robot.

The search was applied to the Title (TI), Abstract (AB), and Author Keywords (DE) fields to ensure comprehensive retrieval of relevant literature.(Number of documents:8254)

The search was limited to: - Publications up to 2024(Number of documents:8079) - Articles (Number of documents:4709) and Conference paper (Number of documents:2968) - English-language publications

The final dataset comprises 7,293 records, retrieved from the Scopus interface in CSV (Comma-Separated Values) format. The exported file includes complete citation and abstract fields for further processing and analysis.

The search string used was: TITLE-ABS-KEY ( biped AND robot ) AND PUBYEAR > 1970 AND PUBYEAR < 2025 AND ( LIMIT-TO ( DOCTYPE , “cp” ) OR LIMIT-TO ( DOCTYPE , “ar” ) ) AND ( LIMIT-TO ( LANGUAGE , “English” ) )

Data loading and Converting

# 匯入Scopus資料
# Set file paths
#file_path <- "~/Documents/data_Biblio/clean_robot_article_9529.csv"  # my mac
file_path <- "C:/Users/chfawu/Documents/scopus_biped robot_7293.csv" # Set file paths(STPI)
# Creates bibliographic data frame
dta <- convert2df(file_path, 
                  dbsource = "scopus",     #資料集來源 
                  format = "csv",          #格式
                  remove.duplicates = TRUE #移除重複
                  )

Converting your scopus collection into a bibliographic dataframe

Done!


Generating affiliation field tag AU_UN from C1:  Done!

Data Overview and Missing Value Checks

# Check dimensions of each dataset
dim(dta)
[1] 7293   53

Part1：Check Missing Value (missingData(dta))

with(bibliometrix::missingData(dta), mandatoryTags) |> knitr::kable(caption="Missing Data Summary")
Missing Data Summary
tag description missing_counts missing_pct status
AB Abstract 0 0.00 Excellent
DT Document Type 0 0.00 Excellent
LA Language 0 0.00 Excellent
PY Publication Year 0 0.00 Excellent
TI Title 0 0.00 Excellent
TC Total Citation 0 0.00 Excellent
AU Author 2 0.03 Good
SO Journal 29 0.40 Good
C1 Affiliation 82 1.12 Good
CR Cited References 265 3.63 Good
ID Keywords Plus 337 4.62 Good
DI DOI 956 13.11 Acceptable
RP Corresponding Author 2716 37.24 Poor
DE Keywords 2716 37.24 Poor
WC Science Categories 7293 100.00 Completely missing

Keyword co-occurance

關鍵字共現分析(Keyword co-occurance)能識別哪些關鍵字在相同的文獻中一起出現 這種共現性表示這些關鍵字所代表的概念之間存在關聯或連結。透過分析這些共現關係，可建構網絡，其中關鍵字是節點，它們之間的連接是基於它們一起出現的頻率。對網絡進行群集分析，可將高度相關的關鍵字分組，從而揭示研究議題概念。

library(bibliometrix)

# 1.建立關鍵字共現網絡
Netmatrix <- biblioNetwork(dta, 
                            analysis = "co-occurrences", 
                            network = "keywords",
                            sep = ";"
                            )

# 2.設定 PNG 匯出位置與檔名
png(filename = "C:/Users/chfawu/Documents/data_Biblio/outputs_biped/Keyword_Cooccurrence_Network.png", 
    width = 1600, 
    height = 1200, 
    res = 200)

# 3.Plot the network
# 使用 networkPlot() 函數將這個網絡視覺化
Keyword_net <- networkPlot(Netmatrix, 
                   normalize = "association",  #標準化方法，"association"會調整邊的權重，反映節點之間的關聯強度
                   weighted = T, #設定為T時權重較高的邊會更粗或更明顯
                   n = 50,   
                   Title = "Keyword Co-occurrences",
                   type = "fruchterman", 
                   size = T,     #設定為T時，度數較高的節點會更大
                   edgesize = 5, 
                   labelsize = 0.7)

# 4. 關閉圖形設備，儲存 PNG
dev.off()
agg_png 
      2

Co-word Analysis

library(gridExtra)
library(ggpubr)
library(grid)  # 轉換圖形物件

# 匯出主成分平面圖
#png("C:/Users/chfawu/Documents/data_Biblio/outputs_connectome/conceptual_structure_MCA_plot1.png", width = 1200, height = 800, res = 150)

# 匯出樹狀圖
#png("C:/Users/chfawu/Documents/data_Biblio/outputs_connectome/conceptual_structure_MCA_dendrogram.png", width = 1200, height = 800, res = 150)

# 進行主題結構分析：使用 Author Keywords + MCA 方法
CS <- conceptualStructure(
  M=dta,                # 書目資料物件
  field = "DE",          # 使用 Author Keywords
  method = "MCA",        # 多重對應分析
  minDegree = 4,         # 節點最小連結數
  clust = "auto",        # 自動群集數決定
  stemming = FALSE,      # 不進行詞幹處理
  labelsize = 8,         # 字體大小
  documents = 10,        # 使用前 10 篇文章
  graph = TRUE          # 不繪製 igraph 對象
)

#grid::grid.draw(CS[[4]])
#grid::grid.draw(CS[[5]])
#dev.off()