Mastering Abstract Writing for Peer-Reviewed Papers Using AI Tools
Learn how to craft compelling abstracts for your research papers using ChatGPT and AI tools. Enhance visibility and citations with structured, keyword-rich abstracts.
File
How to Write an Abstract thats CLEAR and POWERFUL using ChatGPT
Added on 09/03/2024
Speakers
add Add new speaker

Speaker 1: Writing a powerful abstract for your peer-reviewed paper is so important. Firstly, it's the only thing a lot of people read before even citing the paper. And secondly, it's the only thing Google and other search engines can read before they are met with a paywall. So having something that attracts interest, has all of the important information is so important. Here is how I would write one using ChatGPT and AI tools. So the first thing we wanna do is get a structure. Structure is so very, very important, but don't come up with a structure on your own. How about using other people's structures to formulate your own? So the one thing I did was go to one of my papers here, and this was published in Solar Energy Materials and Solar Cells, and I just took this abstract. If you are planning on submitting to a particular journal, use examples from their journal articles that have been published, because it's gonna help you create a better structure for a specific journal. Okay, so the first prompt I used was, based on this abstract from a peer-reviewed paper, break down the generic structure so someone could write their own for this journal. That's very important. We're asking it for a generic structure, and I pasted it in, and this is what happened. It says, breaking down the structure of this abstract for a peer-reviewed paper, we can see that these are the elements. Introduction to the problem or challenge. So one sentence about that. Current state of affairs or limitations, second sentence. Purpose and novelty of your work. That's when we're talking about your paper. Methodology or approach. What did you actually use? One or two sentences, just so people know how you got your results. Key finding or results, super important. Keep it brief, keep it to one, and if you can, one big thing that you found, or two equally big things, but don't overwhelm people. And then we've got implications or significance. So what do your findings mean for the broader public? But we don't stop there. That's nice. Let's make it into an even more user-friendly format. So I said, turn this into a table for me to fill out. Easy, direct, that's one thing I like. And so here we've put out a table, and you can see it's got the section and the description, and then you can put in your own notes. You can paste this into Word, you can just use it as is, but ultimately, this is just the structure that you should be following. And interestingly, go check out my academic writing toolkit, which I'll put a link in the description below, where I've got this structure and others that I wrote way before AI, and this is exactly what I recommend you do, this structure, and there are a couple of the little sneaky ones in that e-book too. But I'm not happy with that. I think it can get even more powerful, because this paper is about a specific subject. What if we wanted to use it for a different subject? So I said, include examples and sentence starters for transparent electrodes. That's not what this paper is on at all, but it did a really, really good job at creating a starter sentence for each section. Check this out. Section, description, example sentence starter. Despite the growing demand for transparent electrodes, blah, blah, blah, super easy. Yes, and it mentions indium tin oxide. That is exactly what I would want to write on. Current state. While ITO has been the standard in transparent electrodes, its brittleness and high cost are major drawbacks. Yes, it knows this. You need to go find references, obviously, but this is a really, really good first draft of any abstract to do with transparent electrodes. Try it for your field. This study introduces a novel cost-effective, your innovative materials and methods here. Then it says, highlight what makes your methods unique. It really has taken all of the important things and put it into really powerful sentence starters. You don't even need to spend hours working out the best way to do abstracts for particular journals anymore. This is all done for you, but it gets even more powerful. Check out this next tip. Abstracts are only as good as the keywords that you can fit into them. It allows search engines to search for it. It allows people using semantic search to search for the appropriate things. You need to make sure you know what people are searching about in your field so you can include them in your abstract. Here is how I do that. The first thing is to head over to Google, just general old google.com, and type in what you want it to be like. Here I've got transparent electrodes. Then you push space. After space, then you know people are searching for all of these things. Transparent electrodes for efficient optoelectronics. So you may want to include efficient optoelectronics in your abstract. Amazing. You also want to mention applications. You also want to mention properties. Maybe it's based on silver nanowires. These are the things I would consider using as keywords for my abstract. I'd put them down into a list, and I would also go to Google Scholar. Exactly the same technique in Google Scholar. Here it is, transparent electrodes. Type it in, push space, and then you've got all of these. I would include the word layer, review, applications, materials. You'll notice in Google Scholar, you know, it's much less wordy, and it's just like transparent electrode, word. And that's because scientists and researchers are efficient beasts. We haven't got time for all those connecting words. Thank you very much. So in here, we're going to say, maybe we want to put solar cell, maybe we want to put substrate, and this is where the power is. So using all of these that you think are suitable for your research field and your abstract, you go to chat GPT, and then I put in what this kicks out, right? After you've done this first thing, you've created your first draft, I would then head on over and say, rewrite what you have drafted, and include as many of these keywords as possible. In here, you can see I've got transparent electrode layer, transparent electrode applications, film pattern, substrate properties. These are the things I know people are searching for, which is going to make your research much more searchable on things like Google Scholar and Web of Science, et cetera, et cetera, et cetera. And this is what the result is. Transparent electrodes consisting of film, blah, blah, blah. So it's squeezed it in, and I read through it, and it really did a great job at subtly putting them in. We know people are searching for them. We know this abstract contains it. It's going to be a much better situation for your abstract and your paper to be discovered than if you didn't do this step. Not many people actually do this, and I'm amazed at how powerful this is, and no one does it at all. I know it's powerful, because this is what one of my most highly cited papers was able to do, is bring in a load of traffic from Google Search. It is one of the most highly cited things I've ever been a part of, and it's all because the right keywords are in that abstract. The last way, you can use ChatGPT or whatever AI tool, Perplexity, Bing, whatever you're using, it doesn't matter. But you can actually, if you're brave enough, put in your entire paper that you have produced and say, write an abstract for this. We will actually use the structures we've generated in the first step to help guide this process. But the first step is to say, okay, just generate an abstract. Let's see what comes out. And then, it's a little bit of a conversation. That's what you gotta do. Check this out. So, I've said, this is the content from my peer-reviewed paper. Can you write a powerful abstract using the content of the paper? And this was the paper here. Highly conductive, interwoven, carbon nanotube, blah, blah, blah, silver nanowires. I included everything from the introduction all the way down. And I just copied and pasted. Like, I didn't worry about copying it in separately. I just sort of like, went from here, and I just said, home, and then I said, yeah, that's what I want. Everything, all the way up to here. Then I said, yeah, write a powerful abstract. And this is what it did. Oh, what a lot of stuff I pasted in. This wasn't possible just sort of like a few months ago. They've increased the token limit, which is so cool, which means you can now paste in entire papers without it saying, sorry, split up the text, please. All right, this was the abstract that it spat out. And I'm not super happy with it, to be honest. It's a bit long, and it says this paper. Okay, but it has got all of the important information in there that I wanna talk about. So now I've got this first draft. I'm gonna use the structure that I created in the first step from someone else's or my previous abstracts and say, hey, use this structure to create an abstract. And that's what I did. So I said, rewrite it using this structure. And I just copied and pasted the structure in. And then I said, use these keywords. So I'm combining everything together in this step to make sure that it is mega powerful and it's getting sort of like the most important input so it can make the best abstract for me. And then what it did is under each section, introduction to the problem, it gave me a kind of little paragraph, which is good, but it's not quite what I wanted. And this is what you've gotta do with ChatGPT in any AI you're using. Just tell it what you do actually want. So I've got here, current state or affairs, purpose and novelty. It's done a really good job at working out what I'm actually trying to do. But then I said, combine all of them into one abstract that is 200 words long, and then it did this. I think that is better than anything I would come up with as a first draft. And all it did was take me a few copied and pastes to get it into this format. Incredibly powerful, and you should be doing this for every paper that you produce. And you should change the abstract for the journal you want to publish in. That way, it's gonna be the most powerful, most relevant, and it's gonna satisfy what the editor clearly wants because you're using examples from past papers. Why did I go, ooh, ooh, ooh, ooh? Because you're using papers from past journals, lovely. So that's how you make sure you create the most powerful abstract ever for your peer-reviewed paper. If you love this video, remember to go check out this one where I talk about how to use ChatGPT to easily write research articles. It's an amazing watch. It's super easy. I'm amazed that you can do it this powerfully. Go check it out. I think it's an awesome next watch for you.

ai AI Insights
Summary

Generate a brief summary highlighting the main points of the transcript.

Generate
Title

Generate a concise and relevant title for the transcript based on the main themes and content discussed.

Generate
Keywords

Identify and highlight the key words or phrases most relevant to the content of the transcript.

Generate
Enter your query
Sentiments

Analyze the emotional tone of the transcript to determine whether the sentiment is positive, negative, or neutral.

Generate
Quizzes

Create interactive quizzes based on the content of the transcript to test comprehension or engage users.

Generate
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript