登录查看更多内容

How do you work with others in a data science framework?

由人工智能和领英社区提供技术支持

Data science is a collaborative field that requires working with others from different backgrounds, roles, and perspectives. Whether you are part of a data science team, a cross-functional project, or a client-facing service, you need to communicate effectively, share your insights, and learn from others. In this article, you will learn how to work with others in a data science framework, a structured approach to define, execute, and deliver data science solutions.

此文章中的业界达人

由社区从 10 条内容中精选。了解更多

Ziyang Zhou

Staff Data Scientist @ LinkedIn | UC Berkeley Stats
Somnath Ojha

Artesian of Insights, Navigator of Data Realms, Delivery Manager: Data Science & AI@MathCo | Ex Lead Data Science…
Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author

1 Define the problem

The first step in any data science framework is to define the problem you are trying to solve, the scope of the solution, and the success criteria. This is where you need to work closely with your stakeholders, such as business managers, domain experts, or end-users, to understand their needs, expectations, and constraints. You should ask relevant questions, listen actively, and validate your assumptions. You should also document the problem statement, the objectives, and the metrics to measure the outcomes.

添加您的观点

Somnath Ojha

Artesian of Insights, Navigator of Data Realms, Delivery Manager: Data Science & AI@MathCo | Ex Lead Data Science Consultant at Accenture | Ex CIMB | Ex Mindtree | Ex ITC Infotech | Data Science | Machine Learning
举报内容
"In our data science framework, collaboration is key. We gather diverse minds like puzzle pieces, each bringing unique expertise. Through brainstorming and shared analysis, we assemble a complete picture. The result? A masterpiece woven from the threads of teamwork."

已翻译

赞
Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
In my experience, this stage requires a balance between technical understanding and business acumen. Engaging the stakeholders with empathy, along with domain expertise, fosters trust and alignment. It’s essential to translate business needs into data-driven questions while keeping in mind the constraints and ethical considerations.

已翻译

赞
Bharat Saxena

Taking LLMs from PoC to Production | Agents | Graphs | NLP | Prompt Engineering | MTech - Data Science and Engineering (BITS, Pilani)
举报内容
urthermore, fostering clear and consistent communication channels between data scientists, domain experts, and IT teams facilitates a shared understanding of technical requirements and potential roadblocks, leading to more effective problem-solving and solution deployment. Regular check-ins and progress updates help maintain alignment with evolving stakeholder priorities and ensure the project remains on track for achieving the intended outcomes.

已翻译

赞

2 Explore the data

The next step is to explore the data you have or need to collect, to get a sense of its quality, structure, and patterns. This is where you need to work with your data engineers, data analysts, or data providers, to access, clean, and transform the data. You should use descriptive statistics, visualizations, and exploratory analysis techniques to summarize, inspect, and discover insights from the data. You should also communicate your findings, challenges, and assumptions to your stakeholders and collaborators.

添加您的观点

Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
Exploring the data isn't just a technical endeavour. Collaboration with domain experts can unearth valuable context, influencing the cleaning and pre-processing phases. An iterative approach, where insights are continuously shared and validated, ensures alignment and creates a shared understanding.

已翻译

赞

3 Build the model

The third step is to build the model that will address the problem and generate the solution. This is where you need to work with your fellow data scientists, machine learning engineers, or researchers, to select, design, and implement the appropriate algorithms, methods, and tools. You should use best practices, such as version control, code review, and testing, to ensure the quality and reproducibility of your code. You should also evaluate your model's performance, interpretability, and robustness, and compare it with alternative or baseline models.

添加您的观点

Ziyang Zhou

Staff Data Scientist @ LinkedIn | UC Berkeley Stats
举报内容
There are definitely circumstances that you will have to build a model, but I don't think *every* problem needs to be solved by a model. In fact, building a model to answer questions can sometimes over-complicate the problem and solution. Also, the more fancy your model is, the more difficulty you are adding to yourself when you explain the results to your business partners. Most of the time, I tend to just find the simplest solution whenever possible that just have the right amount of information to inform decisions.

已翻译

赞
Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
Model building is where the art and science of data science converge. Engaging with peers for code reviews and insights helps in nurturing creativity while adhering to best practices. Collaboration here often means embracing diversity of thought, leading to robust and innovative solutions.

已翻译

赞

4 Deploy the solution

The fourth step is to deploy the solution to the target environment, where it can be used by the intended users or integrated with other systems. This is where you need to work with your software engineers, devops engineers, or IT staff, to package, deploy, and monitor your solution. You should use frameworks, platforms, and pipelines that facilitate the deployment process and ensure the scalability, security, and reliability of your solution. You should also document your solution's features, requirements, and usage instructions.

添加您的观点

Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
Deployment is where theory meets practice. It necessitates close collaboration with IT and DevOps to ensure the model's efficiency and alignment with organisational infrastructure. Understanding their constraints and concerns, and clear documentation, can streamline this phase, ensuring smooth implementation.

已翻译

赞

5 Communicate the results

The fifth step is to communicate the results of your solution to your stakeholders, such as executives, managers, or clients, who will make decisions based on your findings. This is where you need to work with your data storytellers, data journalists, or data visualizers, to craft a compelling and clear narrative that highlights the value, impact, and implications of your solution. You should use charts, graphs, dashboards, or reports that convey your message effectively and persuasively. You should also anticipate and address potential questions, feedback, or objections.

添加您的观点

Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
Communication is as vital as the analysis itself. Collaborating with visualisers and storytellers helps in crafting a narrative that resonates with the audience. Simplicity and clarity are key, and practicing the art of ‘storytelling with data’ ensures that the insights are not just understood but acted upon.

已翻译

赞

6 Learn and improve

The final step is to learn and improve from your experience, feedback, and results, and to apply your learnings to future projects or iterations. This is where you need to work with yourself, your peers, and your mentors, to reflect on your strengths, weaknesses, and opportunities for improvement. You should solicit and accept constructive criticism, celebrate your achievements, and identify your areas for development. You should also keep up with the latest trends, technologies, and best practices in data science, and seek new challenges and opportunities to grow your skills and knowledge.

添加您的观点

Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
Continuous learning is a personal journey, but one that thrives in a community. Engaging with peers for feedback, partaking in knowledge sharing sessions, and contributing to the data science community fosters a culture of growth. Recognising achievements, however small, fuels motivation and innovation.

已翻译

赞

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?

添加您的观点

Iain Brown Ph.D.

Head of Data Science | Adjunct Professor | Author
举报内容
The journey of a data science project is rarely linear. Being adaptable and embracing the iterative nature of this process allows for continuous alignment with business needs and emerging insights. Cultivating a culture of curiosity, resilience, and ethical consideration elevates the overall impact.

已翻译

赞

Data Science

+ 关注

给文章评分

我们借助人工智能创建了此文章。您认为这篇文章怎么样？

很棒不太好

举报此文章

查看全部

How do you work with others in a data science framework?

1

2

3

4

5

6

7

1 Define the problem

2 Explore the data

3 Build the model

4 Deploy the solution

5 Communicate the results

6 Learn and improve

7 Here’s what else to consider

Data Science

给文章评分

感谢您的反馈

更多Data Science相关文章

更多相关阅读内容

How do you work with others in a data science framework?

1

2

3

4

5

6

7

1 Define the problem

2 Explore the data

3 Build the model

4 Deploy the solution

5 Communicate the results

6 Learn and improve

7 Here’s what else to consider

Data Science

给文章评分

感谢您的反馈

查看其他技能