So, I am utilizing the fragile families challenge for my dataset to see which individual and family level predictors predict adolescent academic performance (measured by GPA). Information about my dataset:
FFCWS is a longitudinal panel study in which baseline interviews were conducted in 1998-
2000 with both the mothers and the fathers. Follow-up interviews were conducted when children were aged 1, 3, 5, 9, and 15. Interviews with the parent, primary caregiver(s),
teachers, and children were conducted either in-home or via telephone (FFCWS, 2021). In the
15th year, children/adolescents are asked to report their grades in four subjects- history,
mathematics, English, and science. These grades are averaged for each student to measure their individual academic performance at age 15. A series of individual-level and family-level
predictors that are known to impact the academic performance as mentioned earlier, are also captured at different time points in the life of the child.
I am very new to machine learning and need some guidance. In order to do this, I first create a dataset that contains all the theoretically relevant variables. It is 4,898x15. My final datasets look like this (all are continuous except:
final <- ffc %>% select(Gender, PPVT, WJ10, Grit, Self-control, Attention, Externalization, Anxiety, Depression, PCG_Income, PCG_Education, Teen_Mom, PCG_Exp, School_connectedness, GPA)
Then, I split into test and train as follows:
final_split <- initial_split(final, prop = .7) final_train <- training(final_split) final_test <- testing(final_split)
Next, I run the models:
train <- rpart(GPA ~.,method = "anova", data = final_train, control=rpart.control(cp = 0.2, minsplit = 5, minbucket = 5, maxdepth = 10)) test <- rpart(GPA ~.,method = "anova", data = final_test, control=rpart.control(cp = 0.2, minsplit = 5, minbucket = 5, maxdepth = 10))
Next, I visualize cross validation results:
rpart.plot(train, type = 3, digits = 3, fallen.leaves = TRUE) rpart.plot(test, type = 3, digits = 3, fallen.leaves = TRUE)
Next, I run predictions:
pred_train <- predict(train, ffc.final1_train) pred_test <- predict(test, ffc.final1_test)
Next, I calculate accuracy:
MAE <- function(actual, predicted) {mean(abs(actual - predicted)) } MAE(train$GPA, pred_train) MAE(test$GPA, pred_test)
Following are my questions:
Now, I am not sure if I should use rpart or random forest or XG Boost so my first question is that how do I decide which algorithm to use. I decided upon rpart but I want to have a sound reasoning for the same.
Are these steps in the right order? What is the point of splitting my dataset into training and testing? I ultimately get two trees (one for train and the other for test). Which ones should I be using? What do I make out of these? A step-by-step procedure after understanding my dataset would be quite helpful. Thanks!
I have a text file which has information, like so:
product/productId: B000GKXY4S
product/title: Crazy Shape Scissor Set
product/price: unknown
review/userId: A1QA985ULVCQOB
review/profileName: Carleen M. Amadio "Lady Dragonfly"
review/helpfulness: 2/2
review/score: 5.0
review/time: 1314057600
review/summary: Fun for adults too!
review/text: I really enjoy these scissors for my inspiration books that I am making (like collage, but in books) and using these different textures these give is just wonderful, makes a great statement with the pictures and sayings. Want more, perfect for any need you have even for gifts as well. Pretty cool!
product/productId: B000GKXY4S
product/title: Crazy Shape Scissor Set
product/price: unknown
review/userId: ALCX2ELNHLQA7
review/profileName: Barbara
review/helpfulness: 0/0
review/score: 5.0
review/time: 1328659200
review/summary: Making the cut!
review/text: Looked all over in art supply and other stores for "crazy cutting" scissors for my 4-year old grandson. These are exactly what I was looking for - fun, very well made, metal rather than plastic blades (so they actually do a good job of cutting paper), safe ("blunt") ends, etc. (These really are for age 4 and up, not younger.) Very high quality. Very pleased with the product.
I want to parse this into a dataframe with the productID, title, price.. as columns and the data as the rows. How can I do this in R?
A quick and dirty approach:
mytable <- read.table(text=mytxt, sep = ":")
mytable$id <- rep(1:2, each = 10)
res <- reshape(mytable, direction = "wide", timevar = "V1", idvar = "id")
There will be issues if there are other colons in the data. Also assumes that there is an equal number (10) of variables for each case. All
I am using django1.8 and haystack 2.4 and solr 4.10. Somehow order_by is not working as expected, Please have a look at below code,
>>> sqs = SearchQuerySet()
>>> sqs = sqs.using('entry').filter(status=0)
>>> for b in sqs.filter(content="see").order_by('title'): print b.title
501 Must-See Movies
Look See, Look at Me!
Last Chance to See
1,000 Places to See Before You Die
Pretend You Don't See Her
Learning to See Creatively : Design, Color and Composition in Photography
Behavior Solutions for the Inclusive Classroom : See a Behavior? Look It Up!
See No Evil
Last Chance to See
See It and Sink It : Mastering Putting Through Peak Visual Performance
See No Evil : The True Story of a Ground Soldier in the CIA's War on Terrorism
Voice for Now : Changing the Way We See Ourselves As Women
See Jane Win : The Rimm Report on How 1,000 Girls Became Successful Women
Kaplan Medical USMLE Medical Ethics : The 100 Cases You Are Most Likely to See on the Exam
I See You
You'll See It When You Believe It : The Way to Your Personal Transformation
Body Code : Diet and Fitness Programme: Master Your Metabolism and See the Weight Fall Off
descending Order
>>> sqs = SearchQuerySet()
>>> sqs = sqs.using('entry').filter(status=0)
>>> for b in sqs.filter(content="see").order_by('-title'): print b.title
You'll See It When You Believe It : The Way to Your Personal Transformation
Body Code : Diet and Fitness Programme: Master Your Metabolism and See the Weight Fall Off
Kaplan Medical USMLE Medical Ethics : The 100 Cases You Are Most Likely to See on the Exam
I See You
Voice for Now : Changing the Way We See Ourselves As Women
See Jane Win : The Rimm Report on How 1,000 Girls Became Successful Women
See No Evil : The True Story of a Ground Soldier in the CIA's War on Terrorism
See It and Sink It : Mastering Putting Through Peak Visual Performance
Last Chance to See
See No Evil
Behavior Solutions for the Inclusive Classroom : See a Behavior? Look It Up!
Learning to See Creatively : Design, Color and Composition in Photography
Pretend You Don't See Her
1,000 Places to See Before You Die
Last Chance to See
Look See, Look at Me!
501 Must-See Movies
Why odering is not working like A --> Z and Z --> A
Recently i had same issue with haystack order_by for title. I used python lambda function to sort object list.
ascending order using title:
sqs = sqs.using('entry').filter(status=0)
sorted_list = sorted([s.object for s in sqs], key=lambda x: x.title, reverse=False)
descending order:
sqs = sqs.using('entry').filter(status=0)
rev_sorted_list = sorted([s.object for s in sqs], key=lambda x: x.title, reverse=True)
sqs.order_by works very well with Integer fields.
I have a following data frame:
reviews <- data.frame(value = c("Product was received in excellent condition. Made with high quality materials. Very Good product",
"Inexpensive. An improvement over integrated graphics.",
"I love that product so excite. I will order again if I need more .",
"Excellent card, great graphics."),
user = c(1,2,3,4),
Review_Id = c("101968","101968","210546","112546"),
stringsAsFactors = FALSE)
and I need to have desired output:
user review_Id sentence
1 101968 Made with high quality materials.
1 101968 Very Good product
2 101968 Inexpensive.
2 101968 An improvement over integrated graphics.
3 210546 I love that product so excite.
3 210546 I will order again if I need more .
4 112546 Excellent card, great graphics.
I was wondering about something like this: sent_detect(reviews$value)
But how could I combine that function to have that desired output.
If your data really are so tidy, you can just use cSplit from my "splitstackshape" package.
library(splitstackshape)
cSplit(reviews, "value", ".", direction = "long")
# value user Review_Id
# 1: Product was received in excellent condition 1 101968
# 2: Made with high quality materials 1 101968
# 3: Very Good product 1 101968
# 4: Inexpensive 2 101968
# 5: An improvement over integrated graphics 2 101968
# 6: I love that product so excite 3 210546
# 7: I will order again if I need more 3 210546
# 8: Excellent card, great graphics 4 112546
As a relative novice in R and programming, my first ever question in this forum is about regex pattern matching, specifically line breaks. First some background. I am trying to perform some preprocessing on a corpus of texts using R before processing them further on the NLP platform GATE. I convert the original pdf files to text as follows (the text files, unfortunately, go into the same folder):
dest <- "./MyFolderWithPDFfiles"
myfiles <- list.files(path = dest, pattern = "pdf", full.names = TRUE)
lapply(myfiles, function(i) system(paste('"C:/Program Files (x86)/xpdfbin-win-3.04/bin64/pdftotext.exe"', paste0('"', i, '"')), wait = FALSE))
Then, having loaded the tm package and physically(!) moved the text files to another folder, I create a corpus:
TextFiles <- "./MyFolderWithTXTfiles"
EU <- Corpus(DirSource(TextFiles))
I then want to perform a series of custom transformations to clean the texts. I succeeded to replace a simple string as follows:
ReplaceText <- content_transformer(function(x, from, to) gsub(from, to, x, perl=T))
EU2 <- tm_map(EU, ReplaceText, "Table of contents", "TOC")
However, a pattern that is a 1-3 digit page number followed by two line breaks and a page break is causing me problems. I want to replace it with a blank space:
EU2 <- tm_map(EU, ReplaceText, "[0-9]{1,3}\n\n\f", " ")
The ([0-9]{1,3}) and \f alone match. The line breaks don't. If I copy text from one of the original .txt files into the RegExr online tool and test the expression "[0-9]{1,3}\n\n\f", it matches. So the line breaks do exist in the original .txt file.
But when I view one of the .txt files as read into the EU corpus in R, there appear to be no line breaks even though the lines are obviously breaking before the margin, e.g.
[3] "PROGRESS TOWARDS ACCESSION"
[4] "1"
[5] ""
[6] "\fTable of contents"
Seeing this, I tried other patterns, e.g. to detect one or more blank space ("[0-9]{1,3}\s*\f"), but no patterns worked.
So my questions are:
Am I converting and reading the files into R correctly? If so, what has happened to the line breaks?
If no line breaks is normal, how can I pattern match the character on line 5? Is that not a blank
space?
(A tangential concern:) When converting the pdf files, is there code that will put them directly in a new folder?
Apologies for extending this, but how can one print or inspect only a few lines of the text object? The tm commands and head(EU) print the entire object, each a very long text.
I know my problem(s) must appear simple and perhaps stupid, but one has to start somewhere and extensive searching has not revealed a source that explains comprehensively how to use RegExes to modify text objects in R. I am so frustrated and hope someone here will take pity and can help me.
Thanks for any advice you can offer.
Brigitte
p.s. I think it's not possible to upload attachments in this forum, therefore, here is a link to one of the original PDF documents: http://ec.europa.eu/enlargement/archives/pdf/key_documents/1998/czech_en.pdf
Because the doc is long, I created a snippet of the first 3 pages of the TXT doc, read it into the R corpus ('EU') and printed it to the console and this is it:
dput(EU[[2]])
structure(list(content = c("REGULAR REPORT", "FROM THE COMMISSION ON",
"CZECH REPUBLIC'S", "PROGRESS TOWARDS ACCESSION ***********************",
"1", "", "\fTable of contents", "A. Introduction", "a) Preface The Context of the Progress Report",
"b) Relations between the European Union and the Czech Republic The enhanced Pre-Accession Strategy Recent developments in bilateral relations",
"B. Criteria for membership", "1. Political criteria", "1.1. Democracy and the Rule of Law Parliament The Executive The judicial system Anti-Corruption measures",
"1.2. Human Rights and the Protection of Minorities Civil and Political Rights Economic, Social and Cultural Rights Minority Rights and the Protection of Minorities",
"1.3. General evaluation", "2. Economic criteria", "2.1. Introduction 2.2. Economic developments since the Commission published its Opinion",
"Macroeconomic developments Structural reforms 2.3. Assessment in terms of the Copenhagen criteria The existence of a functioning market economy The capacity to cope with competitive pressure and market forces 2.4. General evaluation",
"3. Ability to assume the obligations of Membership", "3.1. Internal Market without frontiers General framework The Four Freedoms Competition",
"3.2. Innovation Information Society Education, Training and Youth Research and Technological Development Telecommunications Audio-visual",
"3.3. Economic and Fiscal Affairs Economic and Monetary Union",
"2", "", "\fTaxation Statistics "), meta = structure(list(author = character(0),
datetimestamp = structure(list(sec = 50.1142621040344, min = 33L,
hour = 15L, mday = 3L, mon = 10L, year = 114L, wday = 1L,
yday = 306L, isdst = 0L), .Names = c("sec", "min", "hour",
"mday", "mon", "year", "wday", "yday", "isdst"), class = c("POSIXlt",
"POSIXt"), tzone = "GMT"), description = character(0), heading = character(0),
id = "CZ1998ProgressSnippet.txt", language = "en", origin = character(0)), .Names = c("author",
"datetimestamp", "description", "heading", "id", "language",
"origin"), class = "TextDocumentMeta")), .Names = c("content",
"meta"), class = c("PlainTextDocument", "TextDocument"))
Yes, working with text in R is not always a smooth experience! But you can get a lot done quickly with some effort (maybe too much effort!)
If you could share one of your PDF files or the output of dput(EU), that might help to identify exactly how to capture your page numbers with regex. That would also add a reproducible example to your question, which is an important thing to have in questions here so that people can test their answers and make sure they work for your specific problem.
No need to put PDF and text files in separate folders, instead you can use a pattern like so:
EU <- Corpus(DirSource(pattern = ".txt"))
This will only read the text files and ignore the PDF files
There is no 'snippet view' method in tm, which is annoying. I often use just names(EU) and EU[[1]] for quick looks
UPDATE
With the data you've just added, I'd suggest a slightly tangential approach. Do the regex work before passing the data to the tm package formats, like so:
# get the PDF
download.file("http://ec.europa.eu/enlargement/archives/pdf/key_documents/1998/czech_en.pdf", "my_pdf.pdf", method = "wget")
# get the file name of the PDF
myfiles <- list.files(path = getwd(), pattern = "pdf", full.names = TRUE)
# convert to text (not my pdftotext is in a different location to you)
lapply(myfiles, function(i) system(paste('"C:/Program Files/xpdf/bin64/pdftotext.exe"', paste0('"', i, '"')), wait = FALSE))
# read plain text int R
x1 <- readLines("my_pdf.txt")
# make into a single string
x2 <- paste(x1, collapse = " ")
# do some regex...
x3 <- gsub("Table of contents", "TOC", x2)
x4 <- gsub("[0-9]{1,3} \f", "", x3)
# convert to corpus for text mining operations
x5 <- Corpus(VectorSource(x4))
With the snippet of data your provided using dput, the output from this method is
inspect(x5)
<<VCorpus (documents: 1, metadata (corpus/indexed): 0/0)>>
[[1]]
<<PlainTextDocument (metadata: 7)>>
REGULAR REPORT FROM THE COMMISSION ON CZECH REPUBLIC'S PROGRESS TOWARDS ACCESSION *********************** TOC A. Introduction a) Preface The Context of the Progress Report b) Relations between the European Union and the Czech Republic The enhanced Pre-Accession Strategy Recent developments in bilateral relations B. Criteria for membership 1. Political criteria 1.1. Democracy and the Rule of Law Parliament The Executive The judicial system Anti-Corruption measures 1.2. Human Rights and the Protection of Minorities Civil and Political Rights Economic, Social and Cultural Rights Minority Rights and the Protection of Minorities 1.3. General evaluation 2. Economic criteria 2.1. Introduction 2.2. Economic developments since the Commission published its Opinion Macroeconomic developments Structural reforms 2.3. Assessment in terms of the Copenhagen criteria The existence of a functioning market economy The capacity to cope with competitive pressure and market forces 2.4. General evaluation 3. Ability to assume the obligations of Membership 3.1. Internal Market without frontiers General framework The Four Freedoms Competition 3.2. Innovation Information Society Education, Training and Youth Research and Technological Development Telecommunications Audio-visual 3.3. Economic and Fiscal Affairs Economic and Monetary Union Taxation Statistics