Keywords are phrases or words that sum up your research paper. They increase the number of citations you receive and make your paper searchable (and easily discoverable!). Therefore, it is crucial to incorporate domain and methodology related keywords of your research work to aid in the discovery of your paper.
For a researcher, making his/her research paper available to the research community, on the Google search engine within first few search page results is a challenging task. Appearance of an artice in search results will push the research scholars to download , refer and cite articles in their work. This will help the researcher to increase his citation counts and improve his research profile to a greater extent.
Though the the researcher do not have ditrect control over the journal serach ranking on the google search engine in which he/she has published paper, still he/she can make his article rank higher with one simple option. Keyword listing is one such section within research paper, where the researcher has privilage to push the ranking of the paper on google search engine.
Keywords, therefore, are vital for filtering the abundant amount of resources available. Keywords are one of the parameters used for searching an article in a database or a search engine, that in turn retrieves a bunch of results ranked according to relevancy.
The appearance of the document higher in the order is directly proportional to the number of relevant keywords used in that document. Then how to make our research paper rank higher in the search results and not be lost in the ocean of the documents stored on the net? The answer is of course choosing the right and relevant keywords.
The purpose of keywords in a research article is to help other researchers find your paper when they are conducting a search on the topic. Picking the apt keywords is crucial because these are used for indexing purposes. Well-picked keywords help your article to be more easily searched and cited.
In a research article, abstracts are usually followed by a list of keywords selected by the author. Keywords define the domain, subdomain, topic, research objective, etc. that are covered by the article. Most search engines, citation indexing databases, or journal websites refer keywords to decide where and when to display your article to the researchers who are searching for articles related to your article.
Keywords make your article easily searchable and ensure that your article gets more citations. Hence it is essential to include and select relevant keywords and filter out the large body of unwanted material.
Let us take an example to see why keywords are useful. A paper titled ” New approaches in Leaf Image processing using Machine Learning Techniques” describes, how some Machine Learning Algorithms will help in identifying the leaf spot diseases like Bacterial Blight at the earlier stage using Machine learning based Image Processing techniques.
Suitable keywords for such a paper can include the following
- Support Vector Machine, Reinforcement Learning ( From Machine Learning concepts)
- Image Segmentation, Image Clustering (From Image Processing concept)
- Leaf Spot disease names Bacterial Blight, Anthracnose etc. ( From Domain concept)
Search for any of these keywords will lead a research scholar to this paper.
Tips for Writing Keywords for a Research Paper
Here are a few tips that will help you create relevant and effective keywords for your research paper:
Start thinking about the terms you use to search for research papers related to your topic. Possibly these are the terms used by other researchers for searching the topic of your interest. These terms can be ideal keywords for your research paper.
Most of the time it is assumed that a keyword means a single word. However, as per the research, search engine users are becoming more specific and they have understood that a single keyword is probably going to be too broad of a search to return the articles they’re looking for.
A good example is what happens when you do a search for the keyword “security”, you may be in need of articles written on network security for cloud computing but doing a quick search on Google with the keyword security gives you results as varied as articles on security article in Wikipedia, Security of nation, the Social Security number, security jobs in your local area or a recently released movie with title “security”.
Using the keyword phrase “network security for cloud computing”, returns a couple of research articles about network security for cloud computing. Keywords should ideally be phrases of 2-3 words. Ideally can give single-word keywords but it may lead to many false matches. Note that actually, keywords are not simply a set of words instead they are phrases.
Examples: Soft Computing for Leaf Image Analysis, Support Vector Machine for Fruit quality Identification, Neural Network for Leaf Image Extraction
Identify the generally used alternate terms for the words written in your title. That is, include significant abbreviations, acronyms, and other short-form or substitute names for your paper. But care should be taken while using acronyms that may have other meanings. WWW would be an abbreviation since most hits would relate to the Internet. Similarly, other abbreviations like i) SVM for Support Vector Machine, ii) NN for Neural Network and iii) Soft Computing for Neural Network, Fuzzy Logic and Genetic Algorithm can be used
Do not use words or phrases from the title as keywords. Keywords should contain a list of words that supplement your title’s content. This is because most of the search engines and journal databases use Research Title for indexing purposes.
TITLE: “Leaf Image Analysis for Pathological Issues Using Soft Computing Techniques”
KEYWORDS: Unsupervised Neural Networks, Leaf Spot disease Identification, Disease spot extraction using Image Processing.
Keywords should contain words and phrases that suggest what the topic is about. Also include words and phrases that are closely related to your topic. (For example, if the paper is about Image Processing for Leaf Spot Identification and Disease Classification use words like Plant Pathology, Bacterial Blight in pomegranate, Leaf Disease spot Identification etc.
If your research revolves around a key method or technique, make sure the term for it is located in your keyword. Example: If the method uses Multi-Spectral Camera for image capturing and processing then Keywords must contain terms like Multi-Spectral Image Analysis.
If the paper focuses on a particular region use that as a keyword. If the paper is on Voice Analysis of Southern Indian Community for Age Identification then Keyword can be “Southern Indian Community” If the paper is Pomegranate Leaf Spot Identification using a Support Vector Machine and if the pomegranate considered are from the Indian sub-continent then the keyword can be ” Pomegranate plant of the Indian subcontinent“.
If your article is about developing applications, check whether potential applications, issues or phenomena can serve as keywords (Smart City Construction, Plant Pathology, Medical Image Processing for Skin Cancer, Sensor for Tsunami).
The most important experimental techniques used in your article are worth considering as keywords. For example, Stereo Vision for Robot, X-ray analysis for pomegranate fruit etc.
Some sites provide keyword generators or keyword planners to help you think of other terms you could include.
Keyword Identification : An Example from a Research Paper
Let me show you an example of a Research Paper with a Title, Abstract and set of possible Keywords:
Title: Leaf Disease Feature Identification and Extraction Using Deep Neural Network
Extraction of meaningful leaf disease features by applying image processing techniques is a problem that has been studied by the image processing community for decades. Image processing research for leaf spot disease identification has matured significantly throughout the years, and many advances in image processing techniques continue to be made, allowing new techniques to be applied to new and more demanding pathological problems.
In this paper, a method for the detection and classification of leaf spot diseases affecting Pomegranate crops is developed using Deep learning Neural networks. Throughout, we have presented tables and charts to compare the performance of the proposed method with state-of-the-art techniques. Experimental results show that Deep Neural Networks handle uncertainty effectively and they can be trained with limited data sets. The paper has also made suggestions for future research directions.
Image Processing for plant pathology, diseased leaf spot extraction, leaf spot disease classification, segmentation of leaf spot disease, Machine learning for Leaf disease identification, Pomegranate leaf disease identification, Pomegranate leaf disease classification.