Review of: "Feature Selection and Classification of Type II Diabetes on High Dimensional Dataset"
Résumé
I suggest the following structure: Problem, Aim, Methods, Results, and Conclusion.>> The Introduction section is poorly written; it needs to be expanded to add the problems in the context of existing work and how the proposed algorithms and approach can be used to overcome them.>> The structure of the manuscript is poorly distributed; there is no synchronisation between the different parts of the paper.After the introduction section, I suggest putting the related works with more clarifications and highlights about the research gaps, then the proposed methodology, which details the proposed approach including the description of the dataset and the proposed algorithms, followed by the results and discussions section, and finally the conclusion.>> The methodology is not well written and detailed, highlighting the proposed approach, since the problems of this work are not clearly stated.There is ambiguity in statement understanding.>> I feel that more explanation would be needed on how the proposed method is performed, on what basis you reduced the features, what algorithms were used.You mentioned that you used all the ML algorithms (KNN, Nb, SVM, LR, RF, GBS), but in the manuscript, you gave a large part to NB and no comparison was made with the others.In this case, explain the choice of NB, the parameters used... >> For the figures, put the titles and give a brief explanation for each one.For example, you need to explain the blocks used in the first figure .>> References are missing and should be cited throughout the manuscript.>> The results obtained must be interpreted and clearly explained.Additional analysis is required at each experiment to show its main purpose.A comparison with the state of the art in the form of a table should be added.>> The conclusion should not include results, figures, or tables.Rewrite the conclusion section so that it clearly highlights
Citer ce document
Accès au document
Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter
Voir l'article sur le site de la revueAuteur(s)
Statistiques
Consultations : 1
Téléchargements : 0