Shulou(Shulou.com)06/01 Report--

In this issue, the editor will bring you about how to reproduce the GISer paper. The article is rich in content and analyzes and narrates it from a professional point of view. I hope you can get something after reading this article.

one

Introduction

When writing scientific research papers, reproducing other people's papers is also a very important step, and sometimes many students will get stuck in this step directly. So today let's talk about the recurrence of the paper.

The reproduction of the paper should first consider the following issues. (1) can I have the data that the author has? How much is the difference in quality? (2) can I understand the methods used by the author? Can my programming skills solve the problem? (3) what if we do it again and the result of the original text is contrary to that of the original text? Who's wrong? (4) even if the recurrence is successful, what should I do next? In view of these problems, I will give my opinion.

two

Data and methods

When I was reproducing other people's papers, I was now in another study area. Because I think, even if I get exactly the same result as the author, can I write a paper with this experimental result? Such being the case, you might as well choose another study area at the beginning. The original text is Beijing, and you can write about Shanghai, Guangzhou and Shenzhen.

Sometimes you can't get the data of the original text at all, for example, the original text uses mobile phone signaling data, can your research group get it? Is the funds allowed? It is not allowed to use some public data, such as those written in my previous official account article. In addition, I would like to say that the statistical yearbook is the eternal god!

Flickr Geographic tagging Photo Research papers and data download

Data welfare | Shenzhen mobile signaling data download and processing (including ETC/ bus card / taxi track, etc.)

Download and summary of taxi track data in major cities across the country

Data benefits | Microsoft City Computing, AI for Earth

Data welfare | | official sharing of short-term rent data of Airbnb accommodation

As for the method part, are you sure you can reproduce the complex algorithms in the paper? We need to be able to recognize our own strength, machine learning, deep learning CNN, KNN can you do it? If not, don't be a copycat. Honestly add your own understanding and a little improvement to the classical methods, which is already innovation and progress.

three

How to treat the result

If we really choose data and methods that are different from those of the author, then we don't care whether the author's result is correct or not, we just need to write our own paper wholeheartedly. The reproduction of the paper is not an experimental assignment in class. We must follow the experimental instructions. References are not imperial edicts, we only need to treat the reproduction of papers with a normal mind, we can start another stove to complete our own research papers. This is how the GISer shared by the editor reproduces the paper. If you happen to have similar doubts, you might as well refer to the above analysis to understand. If you want to know more about it, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.