Can a layman use AI or Machine Learning?

Machine Learning (ML) is considered to be a subset of Artificial Intelligence (AI), and it applies computer algorithms and mathematical optimization methods to build predictive statistical models. In the hands of those with the right know-how, it is an incredibly powerful tool. However, for a layman or a beginner who have just been introduced to Machine Learning, it can be a daunting prospect.

Use of Machine Learning

Any person who regularly uses the internet will have no doubt come across the term “Machine Learning.” Programs like ChatGPT or stable diffusion can take tasks most people find difficult, such as writing or drawing, and trivialize it with a single input. The ramifications are enormous; after all, having a computer complete arduous work in a fraction of the time you would have taken would be a godsend for productivity. But what makes it work, and more crucially, how can it be used to its fullest potential?

When discussing Machine Learning, it is easy to attribute much more capability to it than is currently possible. A newcomer or layman may only have a vague idea of what Machine Learning is. Thus, to effectively use Machine Learning, it is important to first understand what it is. Machine Learning algorithms “learn” patterns it observes in your data, and then produces a prediction based on this. This can be a known pattern that you are trying to extrapolate from, or an unknown but inherent pattern that you are looking for. Knowing this is vital as it will help inform you of whether it is a suitable tool for your use.

Examples of Machine Learning Application in Business

The underlying concept behind Machine Learning has been around for many decades. What really propelled the development was the advancement of computing power; this allowed these theoretical concepts to be widely used by data scientists in practice. Can Machine Learning solve all our problems? Probably not, but it is very helpful in a lot of areas. To have a sense, let’s take a look at the following possible Machine Learning applications:

• Application #1: An insurance company is seeking to identify the claimants who are more likely to become costly (e.g. high settlement amount for auto insurers, mental health issue for workers’ compensation providers). They are looking for some good predictors (e.g., age, gender, previous claim history, income level, etc.) in order to intervene the claims earlier to reduce costs.
  
• Application #2: A credit card company would be interested in pre-screening likely fraudulent transactions for further investigation in order to reduce losses.
  
• Application #3: For a retail business, identify customers who are more likely to return and purchase again, which would help the marketing department to understand their target customers, and design a more efficient and focused marketing strategy.

Solving some business problems could be very labour intensive (e.g., Application #2 above), and AI or Machine Learning could really provide a much more efficient solution without the need of a team of experienced staff. So, can AI or Machine Learning replace a lot of jobs? Well, may be some, but on the other hand, they may also create new jobs. Still sounds scary if you are not a mathematician, statistician, or a computer scientist? Yet, there is definitely a demand for people who can understand mathematical formula, apply statistical theories, and write computer programs. But what about a layman who don’t have these backgrounds?

How to Start Machine Learning Project with Right Tool

In fact, when using Machine Learning, the technical details are not initially necessary. The most crucial aspects are to:

• understand the problem you are trying to solve
• assess if the problem is something Machine Learning can help
• formulate your problem and determine the type of Machine Learning method(s) to be used
• know the data required to solve your problem

These will create the foundation for your entire project, so it is essential to iron them out before moving forward. 

With the above in mind, the next step is to find a tool you are comfortable with. There are a lot of tools out there. Some requires coding, and some provide a more user-friendly interface. However, most, if not all, of them require some basic knowledge of Machine Learning. At the minimum, you will need to formulate your problem in a way which you can apply Machine Learning methodology.

Still too much? Okay, as a layman or a beginner of AI and Machine Learning, you would want to find a budget & user-friendly analysis tool which can:

• help you to identify and formulate the business problem
• give you some idea on what you can do about Machine Learning
• require no programming
• provide a step-by-step guidance from the beginning to the end
• produce an understandable interpretation in a documentation/reporting format

It is probably very hard to find an affordable tool in the market as the above sounds more like the service of a consulting firm would provide and build a customized solution for their clients. But never say never, a tool like Aidy (www.aideaye.com) can do all the above and it works well with the commonly used application Excel.

Okay, so we now have a Machine Learning tool to use which does not require a lot of technical background. But what about data? Do we need to collect a lot of data before we can use Machine Learning? An interview with Dr. Andrew Ng, an A.I. pioneer, said the following:

I feel like we need to shift in mindset from big data to good data. If you have a million images, go ahead, use it—that’s great. But there are lots of problems that can use much smaller data sets that are cleanly labeled and carefully curated.

Dr. Andrew Ng further said:

One other suggestion: it’s more important to start quickly, and it’s okay to start small.… So I think I see more companies fail by starting too big than fail by starting too small. It’s fine to do a smaller project to get started as an organization to learn what it feels like to use AI, and then go on to build bigger successes.

Jump in. AI is causing a shift in the dynamics of many industries. So if your company isn’t already making pretty aggressive and smart investments, this is a good time.

A Final Note

So don’t wait if you don’t want to be the last one using Machine Learning in your peer group. There is really nothing to lose to get started sooner rather than later. In particular, when a simple tool such as Aidy is already available so that you don’t have to get started until you have someone with a very technical background. It doesn’t seem to be too painful to be an early bird and you may have great potential, so why not? Want some more basic ideas of AI and Machine Learning, and what they can do, take a read at the blog article “2 Broad Types of Machine Learning: 5 Real-life Problem Types”.

Reference: MIT Technology Review, “Andrew Ng: Forget about building an AI-first business. Start with a mission.