3

If a paper takes up an approach to solve a problem, isn't that what makes up the primary research questions it is trying to address? I have been assigned to read a paper and answer a couple of questions based on it. One of the questions is to identify the problem the paper is trying to solve, and another question asks about the primary research questions it answers. I am confused as to what's the difference between these two.

Edit:

The paper is from the domain of artificial intelligence. The problem domain it belongs to is concerned with natural language processing. This domain poses a lot of problems, the paper just picked up one of them and I get that that's the "problem" it is trying to solve. How is a primary research question different from this?

dhanshreeA
  • 197
  • 1
  • 1
  • 8
  • 6
    The questions we try to answer are not always the questions we actually answer. – JeffE May 14 '17 at 14:39
  • 2
    This could be field-dependent. The word problem is not always synonymous with research question. For example, the research question could be what are the sleeping habits of a certain bird, and the problem would be that these birds are extremely rare. – Bitwise May 14 '17 at 19:36
  • @Bitwise: Thanks! I'll edit the question to include that this paper is specifically from the domain of artificial intelligence research. I hope that would give some clarification. – dhanshreeA May 14 '17 at 23:09
  • You say, "I have been assigned to read a paper and answer a couple of questions based on it." You seem to be asking a very generic question. If I were you, I would clarify from the person who gave you the assignment exactly what THEY mean rather than relying on general answers from Stack Exchange. – Tripartio May 15 '17 at 20:43
  • @Tripartio good advice, I did that and got my answer, however I also wanted to know if the community had something more to offer. Thanks! :) – dhanshreeA May 18 '17 at 05:57
  • #alannaC, so, what was the answer you received? – Tripartio May 18 '17 at 08:25
  • For me is more like Lucky dep, for me is black magick. – Dysk Zewnetrzny Nov 16 '18 at 19:02

1 Answers1

2

As the saying goes,

There's more than one way to cook an egg.

In the research context, this means that sometimes a problem is presented, but there are numerous ways to approach the larger problem. It's possible - in fact, it's likely - that most large problems are addressed from multiple angles by different papers.

As a concrete example (in computer science), consider the problem of sorting a list of integers. There are many, many different ways of doing this (some of which you probably don't want to use), but the assumptions you make have a major impact on how you solve the problem. So a single problem might lead to many research questions:

  • How do I sort a list of random integers?
  • How do I sort a list of integers of unknown size?
  • How do I sort a list of partially-sorted integers?
  • How do I sort a list of integers with no worse than O(x) performance, where x is some upper bound?

And so on. Each of these research questions is just a smaller, more digestible piece of a larger problem. But because the problem is so large, you can't answer every question in a single paper -- hence, the discrepancy you pointed out.

tonysdg
  • 6,052
  • 2
  • 26
  • 47