Talk:Machine learning in earth sciences
![]() | A fact from Machine learning in earth sciences appeared on Wikipedia's Main Page in the Did you know column on 13 January 2022 (check views). The text of the entry was as follows:
| ![]() |
![]() | This article is rated C-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||
|
Comments from Triton
edithi Keith,
Organization: well-defined subheadings, the flow is natural and appropriate; length of the paragraphs are mostly just right
Introduction: everything is well defined in the 1st paragraph; I think it is better to include why the technologies are important to not only earth sciences, but also everyone
Language: some sentences are too lengthy; no obvious grammatical mistakes
Illustration/visual:the flow (organization) has presented some of the scientific ideas; the problems and concepts are explained clearly; professional graphics
Reference: no observable problem--Triton Chiu63 (talk) 06:09, 13 October 2021 (UTC)
Comments from Graeme
editHi TseKiChun,
About 28 years ago I studied neural networks, and I am surprised that they are still being used today.
Anyway I will comment on your page: firstly use third person throughout, so don't use "we" or "us" (also don't say "you").
On the referencing side of things, please include a footnote at the end of each paragraph. You can just reuse the references that are earlier in the paragraph. Also in your references you seem to be missing doi's These are always good to have. You seem to have typed your references in an unformatted way; There are cite journal templates that can be easily inserted by the Wikipedia entries. Many of the references include something with "tel" in them eg tel:9276-9282, when this is probably an ISSN and not a telephone number. If someone clicks on these on a phone, it will probably dial a phone number -- pretty undesirable! Graeme Bartlett (talk) 09:29, 13 October 2021 (UTC)
Comments from Yuki
editI think the page is effective in helping me to understand this topic more and raise my interest in learning more about this area. I think the page is generally well-written and the content is quite comprehensive.
In terms of organization, I like that the headings and sub-headings are clear and precise, which allows the readers to understand the gist of the page just by looking at the content.
For the introduction, I think it neatly include the main gist of the topic, but I think you could add a few sentences to address the three main parts that you mentioned in the main text, which are significance, usage, and challenge, since the introduction should serve as a summary for the content that you would mention in the page.
For the language, I think your wordings are generally easy to understand and easy to follow. Though a few grammatical errors could be spotted and some punctuations were misplaced, but I believe that these can be dealt with quite quickly.
For the illustrations, I found that your illustrations are quite easy to understand with the aid of the main text. They bring out the gist of the concept in a very neat way.
For the science, I am a bit confused over why splitting the study area into two adjacent parts would be more useful when testing the classification models. I think it might need more explanations. I also feel like that I am not so sure of the basic principles and processes machine learning work if just by looking at your page. I think you can add one or two sentences in the Black-box operation part to explain why that is a "black-box". But I'm not sure if putting this part in the last section would help. Adding it to the introduction might help? I think you can briefly talk about how the data of the variables would change in machine learning.
For the references, I think you will need to cite the illustrations since I believe that those were modified from diagrams from research papers. I also feel like you can add more footnotes to cite the content. I would generally cite when there's information or data included in the sentence. I'm not sure if this works for you.
Comments from Rachel
editHello Keith,
I think you did a good job in summarising the key usages of machine learning in earth sciences and it is well structured.
Some improvements can be made: I think you may consider putting the "Significance" after the "Usage" because for me it is less important than the main content. You may try to put more links in your page because some concepts are quite difficult to understand.
Keep it up:)
Rachel — Preceding unsigned comment added by Rachelhunggg (talk • contribs) 15:09, 16 October 2021 (UTC)
Comments from Andy
editHi Keith,
Your figures explaining machine learning algorithms are clear and easy to understand. There are also a lot of examples showing how people use machine learning in different studies. You may also introduce the 'Simple Linear Iterative Clustering-Convolutional Neural Network (SLIC-CNN)' algorithms in Mapping section so that readers can have a better understanding in this perspective. — Preceding unsigned comment added by LklAndy (talk • contribs) 03:35, 15 November 2021 (UTC)
Comments from Jasmine
editHey Keith! I like how you structure your page, very well organised. Your visuals are easy to understand too. One thing I'd look forward to is perhaps there could be some visuals in the "Geological or lithological mapping and mineral prospectivity mapping" section too? Would be more appealing if you could add visual explanations on the performance part! But it's great in general! — Preceding unsigned comment added by Jasminesongy (talk • contribs) 16:00, 15 November 2021 (UTC)
Comments from Christy
edit1. The introduction is simple and clear. I think you can also include the challenges in the introduction briefly.
2. The organization is good. I like how you divide your sections.
3. You make good use of tables in giving examples of applications of different methods. — Preceding unsigned comment added by Christycheungkayan (talk • contribs) 07:59, 16 November 2021 (UTC)