Over the course of this project I have mostly been focusing on reading research papers and presenting my findings to the group. This has taught me a few techniques and strategies when researching in an academic sense.

One of the most important things I learned about is how to read research papers. When I first started off very little of what I read made sense. It took me many hours to just understand the first research paper I was given. However, as time went on and I read more and more papers comprehending the papers became easier. As I read more papers from the subject, I picked up more and more vocabulary terms and theories that are prevalent to the subject. This allowed me to recognize and understand those terms in the new papers I read, which would in turn make the papers much easier to digest. Contrasted to the first paper I read, which was filled with many terms and theories I didn’t know or understand, which made the paper much harder to understand.

Another strategy I picked up to help make the papers easier to read, was reading through the papers multiple times. When I first started out I would try to get through and understand an entire paper the first time I read it. This would take a long time and leave me mentaly drained afterward. Recently, however, I realized if I read through the paper once and didn’t worry about fully understanding it, I could come back later, after I had time to digest what I read, to give it a second readthrough. This second readthrough would go much smoother and I would understand a lot more. This would not only take less time overall, but it would also be less taxing and make the reading more enjoyable.

I also learned which sections of the paper to focus on, based on what I was looking for. If I was looking for the experiments and algorithms the paper discovered or proposed I would only read the abstract and the place where they explain what they did, and possibly the conclusion. I would skip over the introduction, as I was already introduced to the topic since I already read many papers about it, the related work section, as I probably already knew a lot of the related work, and the methodology and results, because the vast majority of the time the proposed algorithm would perform better than the current state and the authors would have also summarized the results in the abstract and conclusion. If, for instance, I wanted to try and find a wide range of ideas and theories the related works section is a great place to read because it has, like the name implies, a bunch of related work and a variety of algorithms to look deeper in.

I also learned how to keep track of the research and how to organize it, or at least how I like to organize it. I prefer to keep a document with all the papers I have read so far and below the titles a brief description of what they cover. I don’t like to go into a lot of detail because if I use the paper I will have to re-read it and too much detail will make the document hard to search through. I tend to just write down the main algorithms mentioned that way I can search by algorithm, and a brief sentence or 2 about what the paper found.

When it comes to presenting my findings I make sure to understand the key topics of the paper and find a way to explain them as simply as possible. I make sure not to try and explain the math, because most of the time I don’t understand it well enough and it’s not that important to the presentation. The important parts are how it works in general and how well it works, relatively, if the people I’m presenting to want to know more about why it works or how exactly the math works they can read the paper.