[ad_1]
What do you think you’re doing here?
Now that winter break is over and many of us are heading back to the office and getting back to work, I’ve been thinking a little bit about the relationship between machine learning capabilities and the rest of business. As I’ve been settling into my new role at DataGrail since November, I’ve learned how important it is for a machine learning role to know what the business actually does and what it needs. I remembered.
My thoughts here do not necessarily apply to all machine learning practitioners. The academic purists among us can probably relate. But rather than advancing machine learning for its own sake, those tasked with doing machine learning for businesses and organizations should reflect on how they interact with the organizations they work for. I think it’s worth it.
I mean, why would someone decide to hire your skill set here? Why were new headcounts required? Especially for technical positions like ours, entry-level salaries are It’s not cheap. Even if you were to fill the role of someone who left, it’s not guaranteed these days and perhaps there was a specific need. What argument was made to those holding the purse strings that they needed to hire someone with machine learning skills?
If you look into this question, you’ll find some useful things. First, what is the ideal outcome that people expect from being with you? They want to achieve data science and machine learning productivity, but they have to know what that is. However, it may be difficult to meet those expectations. You can also learn about company culture from this question. Once you know what value they think they’ll get from bringing in new ML people, are their ideas realistic about the potential contribution ML can make?
In addition to these expectations that you are putting into place, you need to create your own independent view of what machine learning can do within your organization. To do this, you need to look at the business and talk to many people in different functional areas. (In fact, this is something I spend a lot of time answering in my role right now answering this question.) What are businesses trying to do? What do they believe will lead them to success? What is your equation? Who is your customer and what is your product?
Somewhat related to this, we also need to ask about data. What data does your business have, where is it located, how is it managed, etc. This will be critical to accurately assess what efforts should be focused within this organization. We all know that having data is a prerequisite for doing data science. If your data is disorganized or missing altogether, you should be the one who needs to speak up to your stakeholders about what reasonable expectations are. It is for the purpose of machine learning based on that. This is part of bridging the gap between the business vision and the reality of machine learning, and is sometimes overlooked when everyone wants to focus on developing new projects.
Once you understand these answers, you need to bring your perspective on how elements of data science can help you into the discussion. Don’t assume everyone already knows what machine learning can do. Because this is almost certainly not the case. Other roles have their own areas of expertise, and it would be unfair to assume they also know about the intricacies of machine learning. This is a really fun part of the job because it allows me to explore creative possibilities. Is there a hint of a classification problem somewhere, or a predictive task that would actually help some departments succeed? No one has had time to dig into it, even though it could probably yield useful insights. Do you have a lot of data sitting somewhere? Maybe your NLP project is sitting in a pile of unorganized documents.
Understanding your business goals and how you expect people to achieve them will help you connect machine learning to those goals. You don’t have to have a silver bullet solution that will solve all your problems overnight, but if you can draw a line from what you want to do, you can integrate your work with the rest of the company. It will work better. Towards the goal that everyone is aiming for.
This may seem like an off-topic question, but in my experience it’s extremely important.
If your work is not aligned with the business and understood by your colleagues, it will be misused or ignored, and the value you could have contributed will be lost. If you read my columns regularly, you know that I’m a big advocate of data science literacy, and that I believe that DS/ML practitioners have a responsibility to improve data science literacy. You’ll understand. Part of your job is to help people understand what you’ve created and how it can help them. It is not the responsibility of finance or sales to understand machine learning without education (or “enablement” as it is often called these days); it is your responsibility to get the education.
This may be easier if you are part of a relatively mature ML organization within your business. Hopefully, this literacy has been understood by others before you. However, this is not guaranteed, and even large and expensive ML capabilities within an enterprise can become siled, isolated, and unreadable to other parts of the business, and this is a frightening situation.
What should you do about this? There are many options, but it really depends on your organization’s culture. Take every opportunity to talk about your work and speak at a level that the average person can understand. Defining technical terms can be difficult and take time to master, so explain them more than once. No matter what wiki or documentation system your company uses, create documentation so you can refer to it if you forget. Be honest, open, and friendly by offering to answer questions, even if they seem simple or wrong. Everyone has to start somewhere. Once you have a basic level of interest from your colleagues, you can set up learning opportunities such as lunches, learn more about ML-related topics more broadly than just your current specific project, and set up discussion groups. Masu.
Moreover, it’s not enough to explain all the great things about machine learning. You should also explain why your coworker should care and what it has to do with the business as a whole and your coworker’s individual success. What does ML bring to make their jobs easier? This question should have a good answer.
I’ve framed this in some ways as a way to get started in a new organization, but even if you’ve been working with machine learning in your business for a while, it’s useful to review these topics and look below. The way things go. Being effective in your role requires ongoing care and maintenance, rather than a one-and-done type of transaction. But if you keep going, you’ll learn that machine learning doesn’t have to be scary, that it can help your work and goals, and that your department should be collaborative and collegial rather than vague and siled. Your work will become easier as your colleagues will learn.
To summarize the main points:
- Find out why your company adopted machine learning and ask the expectations underlying that choice.
- In order to do work that contributes to (and stays relevant to) the business, it is essential to understand what the business is and its goals.
- You need to help people understand what you do and how it can help them. Because people won’t automatically and magically understand that.
[ad_2]
Source link
