Research Workflow

This repository presents and organizes workflows for researchers conducting their studies. It includes detailed goals and the steps. These ideas come from my reflections as a PhD candidate on how to approach research effectively. Fellow researchers are welcome to share their workflows and propose possibilities for improving these processes.

The journey is long, often involving repeated steps. It can be fun at times, tedious at others, and occasionally challenging. Yet, I find myself traversing this path over and over. In this journey, I often grapple with feelings of spiritual loneliness and a sense of being lost. Yet, when I take a moment to pause and reflect on the path I’ve traveled and the progress I’ve made, a sense of relief washes over me. It’s in these moments of clarity that I find renewed courage and determination to continue moving forward.

Even with the exact same steps, different researchers will produce completely different results. Each step is a different experience for the individual. Therefore, the uniqueness of research can be reflected in all aspects of a research work. As a young researcher, I think an important lesson is to see yourself at every step.

I'm always curious about how others do research and look for possibilities to improve productivity. So, welcome to share.

📧Email: suemonsuemon@gmail.com

Basic Elements

The complete research process includes:

1. Restrictive Known Conditions
   The tasks or requirements, shown as prisms.

2. Intermediate Targets
   The detailed goal or target, shown as a rounded-corner matrix.

3. Multiple Steps
   Multiple necessary steps needed to achieve each target, represented by text on arrows.

Below is a visual representation of the basic elements:

flowchart LR
I[\inputs\]--multiple steps-->A([target])

Loading

---

Technological Innovation Workflow

For junior researchers (e.g., PhD candidates), the research work starts with some given project requirements and ends with the publication of a paper. Innovations and contributions primarily lie in technical solutions.

I reflected on and summarized my workflow, as shown below. Interestingly, the flowchart takes on a shape that’s wide in the middle and narrow at both ends. This shape reflects the essence of my work: starting with one or two compelling ideas, diving into extensive research, brainstorming, designing, and experimenting, and finally distilling everything into a polished paper.

flowchart TB

Ir[\tasks,
    requirements...\]--
1.collect magazine papers on this scenario
2.read pictures/summary texts/basic concepts
--> As([Know the general scenario])  
-- choose the suitable parts 
    and follow them
--> Aos    



Ir --
1.collect technical papers on this topic
2.read problem formulations/math notations
--> Apb([Know the general problem]) 
-- choose the suitable parts 
    and follow them
--> Aos  


Io[\"your own ideas,
    sth you want to do..."\]--
1.Collect relevant technical reports/papers
2.Read the Abstract/Intro to understand the rationale
3.If match, understand the mathematical derivation
--> Aot([Know what new techniques to use]) 


Ir --
1.Reread the technical papers on this topic
2.Read techniques and methods
3.Search and understand the rationale
4.Ignore the technical details
--> At([Know the techniques used])    


Aos -- 
1.write your own solutions
2.state the background and problems
3.summarize your contributions
4.refine the writing 
--> Aopp

Aot -- 
new solutions naturally 
emerge during integration
--> Aos([design your own technical solution]) -- 
1.choose suitable opensorce programs
2.run and understand their implementations
3.program your own parts
4.design new experiments 
   to show your advantages
--> Aopg

Aopg -->|adjust solutions
         based on results| Aos

At --
1.Reread the technical papers on this topic
2.read/summarize the comparative methods
3.read/summarize the datasets/simulation setting
4.read/summarize the main metric
5.read/summarize the experiment types
6.read/summarize the form of results
--> Ae([know the general experiment settings])
-- choose the suitable parts 
    and follow them
--> Aopg([program your own solutions 
        and conduct experiments])
    
Aopg --
1.choose the experiment results
2.show them in proper forms
--> Aopp([write your own papers])


Aopp --wait for months-->Aopps([Submit,
                     Reply to Reviews,
                     Publish])

Loading