International Journal of Artificial Intelligence and Machine Learning
|
Peer review is defined as obtaining advice on individual manuscripts from reviewer’s expert in the field who are not part of the journal’s editorial staff. Peer review exists to ensure that journals publish good science which is of benefit to entire scientific community. The journal aims to provide the best possible service to authors of original research articles, and the fairest system of peer review, with subject experts and dedicated Editorial Board of internationally renowned scientists.
Peer reviewers may point out gaps in your paper that require more details or further experiments. If your paper is difficult to understand, reviewers will give you suggestions, so that you can fix them. Peer reviewers also consider the importance of your paper to others in your field and can make suggestions to improve or better highlight this to readers.
In addition to offering authors advice, another important purpose of peer review is to make sure that the manuscripts published in the journal eventually high quality.
Types of peer review
There are different forms of peer review used by the journal, although the basis is always the same, field experts providing comments on a paper to help improve it. The most common types are..
Single blind – where the reviewers are aware of the authors’ identities but the authors do not know who reviewed their manuscript.
Double blind – in this case neither authors nor reviewers know each other’s identities.
Open – where there reviewers are aware of the authors’ identity and the reviewers’ identity is revealed to the authors. In some cases journals also publish the reviewers’ reports alongside the final published manuscript.
SvedbergOpen usually practices double blind peer review process.
Key peer-review process includes the following:
There are two stages of review process. The first stage is internal review. In the first stage, our screening team will perform the initial technical check, plagiarism and the suitability of the paper. If found any plagiarism, we shall return the paper to author asking him/her to modify it.
The second stage of review is the external refereeing. Under this the paper will be sent to external referees for double blind refereeing. The referee has three options; he can give the feedback as accepted/rejected/need to be revised. If the referee accepts the paper, it would be published automatically. If the referee suggests any modifications or revisions in the paper, the author has to revise the paper and resubmit to us. If the referee rejects the paper, we shall return to the author with comments. You can read more about the peer review process here.
Copyright © SvedbergOpen. All rights reserved