All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
The revised article is ready for acceptance.
[# PeerJ Staff Note - this decision was reviewed and approved by Claudio Ardagna, a PeerJ Computer Science Section Editor covering this Section #]
Addressed Issues are improved , no comment
no comment
no comment
I am pretty much satisfied with the revised version of the paper. The authors carefully address my concerns. I suggest accepting the research work in its current form.
Well, the revised version of the paper looks more suitable for publication than the first one. I think now the paper's quality is good enough to get published. Therefore, my suggestion is to accept the paper.
See above
See above
Dear Authors,
Reviewers are suggesting some improvements especially related to adding more details in the Introduction and experiment parts. The novelty of the work also needs to be highlighted.
[# PeerJ Staff Note: Please ensure that all review comments are addressed in a rebuttal letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. It is a common mistake to address reviewer questions in the rebuttal letter but not in the revised manuscript. If a reviewer raised a question then your readers will probably have the same question so you should ensure that the manuscript can stand alone without the rebuttal letter. Directions on how to prepare a rebuttal letter can be found at: https://peerj.com/benefits/academic-rebuttal-letters/ #]
There are several concerns with this manuscript in its current form.
This paper proposes a ML framework for sentiment analysis based on Tweets which are related to COVID-19 vaccines. Additionally, different feature engineering approaches were tested toward providing the suitable one in terms of accuracy.
The abstract should be revised and connected to the introduction. When the title is read as the introduction, they are related to the same purpose, which is: to explore feature engineering techniques considering sentiment analysis regarding COVID19 vaccination. However, it does not happen in the abstract, it is not clear what is the real meaning of the manuscript. The abstract should be revised as it does not enough chiefly introduce the area of research along with the research question.
The introductory part should provide more information and contributions to the proposal instead of defining theoretical issues.
Acronym table needed to be added at the end of the paper.
In LSTM detail, reference is not properly cited. It is showing up (?) on line 293.
The python packages used should be included with reference.
Main theme of paper is sentiment classification and diverse feature engineering. Authors need to cite some state-of-the-art approaches based on these two topics in related work to make it interesting. How much work is done on these topics and still continue.
Any reason why deep learning model not performs well?
Limitations of proposed work needed to be added.
Suggestions:
- In general, the learning models seems little scared. They should be in structured way.
- The novelty and contributions of your work should be better explored.
The paper presents sentiment analysis of tweets on different COVID-19 vaccinations. The dataset "COVID-19 All Vaccines Tweets" is downloaded from Kaggle. A bunch of classical machine learning models including Random Forest, Naive Bayes, logistic regression, SGD, ETC, ANN, and CNN were analyzed and compared. Overall, the paper is well organized and straightforward to follow. However, considering the dataset adopted and the learning methods evaluated, the paper is more of a fundamental application of machine learning models on a publicly accessible dataset. Then to make something interesting authors manually labelled the dataset which is a good approach. Still for that, I have the following concerns:
Concerns:
(1) No details are found how the authors manually labelled the dataset in the dataset description section.
(2) Based on this paper, "TF", "TF-IDF", and "word-embeddings" are the three data representation schemes. I strongly recommend to add a table showing advantages and disadvantages of the three.
(3) All the machine learning methods introduced in this paper are traditional and fundamental, no need to explain them in detail by adding equations. In order to make paper compact, Add them all in a two-column table with proper referencing.
(4) For the deep learning models presented in this paper, the authors used a 2x2 max-pooling layer in CNN. Since the tweet data is text-based and sequential, one-dimensional convolution is often recommended to process it.
(5) Also, the authors should add the limitations of proposed approach.
See the above section.
See the above section
Minor format and writing issues:
The dataset is divided into two classes: ’against’ and ’in favor’. Tweets classified into the former class present positive opinions of users", 'against' and 'in favor' should be switched to make the meanings consistent with the explanation. They should need to strict with one label convention either positive, negative or against, in favor.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.