Reducing paper results by joining keywords"

In petro.One: Statistics and Text Mining for Oil and Gas Papers from OnePetro Metadata

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

This is a demonstration on how the number of papers can be reduced using additional keywords to control the number of results returned.

library(petro.One)

test #1

We start with these keywords: water injection water flooding machine-learning artificial intelligence neural networks

# provide two different set of keywords to combine as vectors
major  <- c("water injection", "water flooding")
minor  <- c("machine-learning", "artificial intelligence")
lesser <- c("neural networks")

result_object <- join_keywords(major, minor, lesser, get_papers = TRUE)
result_object

# save findings
# save the three objects as one
papers <- result_object
wat_inj_ml_1 <- petro.One:::as_named_list(major, minor, lesser, papers)
save(wat_inj_ml_1, file = paste0("wat_inj_ml_1", ".rda"))

# load previous save
load(file = paste0("wat_inj_ml_1", ".rda"))
papers <- wat_inj_ml_1$papers
papers

Using a new function to replace multiple commands

paper_results <- run_papers_search(major, minor, lesser,
                                   get_papers = TRUE,       # return with papers
                                   verbose = FALSE,         # show progress
                                   len_keywords = 4,        # naming the data file
                                   allow_duplicates = FALSE) # by paper title and id

names(paper_results)

test #2

Then, we increase the number of keywords:

water injection     water flooding
machine-learning    machine learning    intelligent
neural network      SVM                 genetic
algorithm

# provide two different set of keywords to combine as vectors
m  <- c("water injection", "water flooding")
n  <- c("machine-learning", "machine learning", "intelligent")
p  <- c("neural network", "SVM", "genetic")
q  <- c("algorithm")

paper_results_9 <- run_papers_search(m, n, p, q,
                                   get_papers = TRUE,       # return with papers
                                   verbose = FALSE,         # show progress
                                   len_keywords = 4,        # naming the data file
                                   allow_duplicates = FALSE) # by paper title and id

paper_results_9$papers

petro.One documentation built on May 2, 2019, 3:10 p.m.