Article: How to create a research project by mining the datasets of open science sharing?
Author: Liu Huan (1983-), Chief Editor, Master of Science (First Class Honours, 2009), The University of Auckland.
Copyrights Registered by Block Chain, and the Block Chain ID (6921cb6d84b3a74c435d45d7575bad612f3a67262c5233bbc61e1af6bf5c5439). Register time: 2025-02-03 16:49
All copyrights reserved. Latest revised on 2025-2-3
In recent years, due to the construction and development of open research data sharing platform, such as Zenodo and Science Data Bank, the traditional review papers have not meet the requirement of the era development any more, especially under the background of the increasingly intensive competition and more strictly regulatory environment. The traditional review papers are only suitable for textbooks, which must be changed to Dissertation papers to meet the requirements of international journal publishing.
Dissertation obtains the empirical data by searching and collecting data (mainly by quoting the published data of others and obtaining the use license) in an organized way. Then re-editing and re-calculating data creates new original research papers as dissertation, which are suitable for the first year postgraduate study (Research-based) or for the Master dissertation (Coursework-based). The main requirements are as follows:
1.Relevant empirical data sourcing from at least three published papers in the database should be cited. Among them, the empirical data sourcing from each paper shall not exceed one-third of the dissertation re-edited. Otherwise, the originality of this dissertation will be reduced;
2.The data must be estimated according to the ‘regression discontinuity designs’, as indicated by the ‘observational study’ project of Open Science Framework. This means that the data used in your study should not be exactly the same as the original data, and this can be estimated by establishing the discontinuity regression according to the original data (for example, if the original data is the pollutant concentration at water depth of 1000m, then your data must be estimated at water depth of 900m by establishing the regression analysis based on the original data). Pictures and tables are converted into each other. For example, if the result data of the cited paper is described in picture format, it must be converted into table form and placed in your own paper. In summary, the data and its expression method should not be exactly the same as the original data, which can be estimated according to the regression analysis established on the basis of original data;
3.Clearly quoting the source near each empirical data; you must pay for the license for data use if needed;
4.Re-describing the results of the cited empirical data with different sentences, or re-calculating the cited empirical data with new methods, getting different results as to show your own innovation;
5.In the discussion section of dissertation, the conclusions obtained by comparing and contrasting with other papers must be significantly different from the original cited papers to show theoretical innovation. This is very important. It is often the key difference between grade A and B. For publishing in my journal, it is also required to add AI & Applied Maths programming (C++) into research project, in order to achieve higher grade.
6.For publishing in my international journal, the total number of words must exceed 20 000 English words in an article, requiring systematic discussion and deeply exploring the natural science findings to form a monograph.
版权申明：禁止复制、转载！
Copyrights Registered by Block Chain, and the Block Chain ID (6921cb6d84b3a74c435d45d7575bad612f3a67262c5233bbc61e1af6bf5c5439). Register time: 2025-02-03 16:49
近年来由于开放型研究数据共享平台的建设和兴起，比如Zenodo、Science Data Bank之类研究数据共享平台的发展，传统的综述类论文已经显得不再符合时代发展，尤其是在竞争与监管环境日益激烈和严格的背景下，传统的综述类论文仅仅适合教科书，必须变更为Dissertation论文才能符合国际期刊的发表要求。
Dissertation论文是通过有组织地收集资料的方式来获取实证性数据（主要通过引用他人已经发表的论文数据，并且取得相关数据使用许可），自己重新编辑、计算，创作出新的原创型实证性论文。可以适用于研究生一年级阶段中（研究类）科研基础性训练或是硕士论文（授课类）。要求如下：
一、至少要引用三篇以上相关的、已经发表在数据库中论文的实证性数据，其中每篇原论文中被转载的实证数据在自己重新创作、编辑的原创型论文中不得超过三分之一。否则自己重新创作的论文原创度则会打折扣；
二、根据Open Science Framework平台中观察型研究项目的要求，论文中所使用的数据必须根据“非连续型回归分析设计”进行估算。这意味着你的论文中所使用的数据不应当与原始数据完全相同，而这可以通过基于原始数据建立非连续型线性回归分析来估计数据（比如，如果原始数据是在1000米水深的污染物浓度，则论文使用数据可以是基于原始数据、通过回归分析估算900米水深的污染物浓度）。另外就是图表之间相互转换。比如如果原论文的结果数据是图片格式描述，则必须转换成表格形式放在自己的论文中。总而言之，论文中的数据和表达方式都不能与原始数据完全一致，但是可以在原始数据基础上，通过建立回归分析来进行估算新数据；
三、在每个实证数据旁清晰标注引用出处；如果需要购买许可，则必须付费取得许可；
四、用不同的语句重新描述原始实证数据的计算结果，或者用新的方法重新计算原始实证数据，得出不同的结果，以示自己的创新；
五、在discussion一节中，通过与其它论文进行对比对照得出的结论必须与原始被引用的论文显著不同，以示理论创新，或者提出新的自然科学规律。这点是很重要的，往往是A等与B等的区别；为了在我的期刊上发表，还需要在研究项目中添加人工智能和应用数学的编程项目（C++），以达到更高的成绩。
六、对于我自己主办的国际期刊，一篇论文篇幅的总字数必须超过20 000个英文字数，需要系统而深入对某一命题进行发掘，形成一本专著。
修改于2025-2-3
发表于 2021-11-7 12:04:53
收藏