The next phase in performing an SLR is searching for research, which entails using the search terms and databases established during the preparation stage to find pertinent studies. This is commonly done by searching electronic databases, but it is also possible to search other sources like conference proceedings and grey literature.
Electronic database searches are a useful tool for finding many studies on a particular subject. It's crucial to use numerous databases and the search terms determined during the preparation step when performing a database search. In order to focus their searches, researchers should be careful when utilising Boolean operators like AND and OR
Process of conducting an electronic database search as part of the Searching for Studies stage in an SLR
| Steps | Description | Computer Science Example |
|---|---|---|
| 1. | Use multiple databases | Use multiple databases such as Google Scholar, IEEE Xplore, and ACM Digital Library to search for relevant studies. |
| 2. | Use search terms identified during the preparation stage | Use search terms such as “machine learning,” “computer vision,” and “deep learning” to locate relevant studies. |
| 3. | Refine searches using Boolean operators | Use Boolean operators such as “AND” and “OR” to refine searches, for example: (“machine learning” AND “computer vision”) OR “deep learning.” |
This table provides an overview of the key steps involved in conducting an electronic database search as part of an SLR, using a computer science example. This process can help researchers to locate a large number of relevant studies on a given topic and is an important step in the overall SLR process.
Finding the studies and screening them for eligibility are the following steps. This entails comparing each study to the inclusion and exclusion standards established during the planning phase. Studies that don't fit the requirements for inclusion are taken out of the review, while those that do are kept in for additional examination.
Screening the Studies
| No | Study Title | Inclusion Criteria | Exclusion Criteria | Eligibility |
|---|---|---|---|---|
| 1 | "Artificial Intelligence in Healthcare" | Studies published in English | Opinion pieces and non-research articles | Included |
| 2 | "Machine Learning Approaches for Predictive Maintenance" | Studies published in the past 10 years | Studies focused on other industries | Included |
| 3 | "Neural Networks in Image Processing" | Studies using neural networks as the main approach | Studies using other techniques | Included |
| 4 | "The Impact of Deep Learning on Speech Recognition" | Studies evaluating the impact of deep learning | Studies that only compare deep learning to other techniques | Included |
| 5 | "Applications of Reinforcement Learning in Robotics" | Studies applying reinforcement learning in robotics | Studies focused on other domains | Included |
Note: The above table is just an example, and the inclusion and exclusion criteria will vary depending on the specific research question and scope of the systematic literature review.
The research that is part of the review should be organized to make evaluation and retrieval simple. To do this, you might use a spreadsheet, a database, or reference management software.
Here's a hypothetical example of a table that could be used to manage the studies in a computer science-related systematic literature review:
Managing the Studies
| Study ID | Author(s) | Year | Title | Method | Participants/Samples | Results |
|---|---|---|---|---|---|---|
| 1 | Smith et al. | 2020 | "Evaluation of Machine Learning Algorithms for Image Classification" | Comparative Study | 100 | Algorithm X was found to have the highest accuracy for image classification |
| 2 | Johnson et al. | 2021 | "A Study on Deep Learning Techniques for Speech Recognition" | Experimental Study | 50 | Technique Y showed significant improvement in speech recognition compared to traditional methods |
| 3 | Patel et al. | 2022 | "Comparison of Reinforcement Learning Approaches for Gaming Applications" | Comparative Study | 75 | Approach Z was found to perform the best in terms of speed and accuracy for gaming applications |
This table provides a concise overview of the key information for each study, including the study ID, author(s), year, title, research method, number of participants/samples, and results. By organizing the studies in this manner, it becomes easier to assess the quality and relevance of each study, compare the results of different studies,
In conclusion, searching for studies is an important step in conducting an SLR. By using electronic database searches, screening the studies for eligibility, and managing the studies in an organized manner, researchers can ensure that they have located all relevant studies on their topic of interest and that they are ready to move on to the next stage of the SLR process.
Please create an Issue for any improvements, suggestions or errors in the content.
You can also contact me using Linkedin for any other queries or feedback.