I've been a couple of weeks dealing with a problem i haven't been able to resolve. I'm preparing tagged twitter data to be trained with skip gram, for training subject referencing in phrases.
Let me explain a bit better with an example, let's say i have a root (original) tweet:
"the scientist was reading the paper" where i tag every reference with the tag CAD just like html:
"the <CAD=1> scientist</CAD> was reading the <CAD=2>paper</CAD>"
and also every reply to that tweet is tagged the same way, if it refers to the scientist or paper:
so yes its making the comparison but its reading by columns, i need it to be read by rows so it compares one reply at a time, all of this so i can automate the creation of this vector. I've also tried using mapply with for loops but this has only made things much more complex, maybe is my lack of experience but there must be a simpler, easier way to achieve this.
This is a very detailed example. I am novice in R, only studying it at university and gaining knowledge in research paper writing, but I aim to use R for my sociological research in a field of digital communications. Hope this explanation will help to build a research strategy.