Understanding the practical applications of business AI | Tech Industry

There’s more published content on AI in today’s media than ever before. We read about the milestones that AI has surpassed, like Deepmind’s AlphaGo beating the world’s best Go player or Facebook’s facial recognition performing better than you at tagging photos. And to be fair, many of these articles are actually starting to cover the technological advancements without the sensationalist intrusions but I often find myself asking how is it applicable to my work? To develop an intuition for how AI can benefit our own work, we need to take all the content and look at it from the 10,000 ft. view.

It all starts with a well-defined business problem that you want to solve. You don’t want to start with machine learning and make up a business problem. But instead, think about how to really improve and grow your business. Once we have a really good problem, we can then see if AI can help.

What can AI do?

To know if AI can help solve our problem, we need to first know what it can do.

Cognitively Simple Tasks

AI can help with tasks that we can do easily without a whole lot of complex cognition. Let’s look at the picture and tweet below. Can you tell what’s in the picture? Can you determine the sentiment behind the tweet? These are tasks that humans can do pretty easily, but we can use AI to do them at scale and with great performance. Achieving this scale for AI tasks enables us to offload these basic cognitive tasks and focus on the harder, more creative parts of the problem.

Cognitively Complex Tasks

AI can also help with complex cognitive tasks. Suppose we want to predict the selling price of a house given some information about it (# of rooms, location, etc.). We can’t solve the problem by looking at descriptive tables or graphs because the task requires complex cognition. An experienced realtor might be able to process this information and estimate the price but this approach has two limitations. First, you need the relevant experience in order to use this information and predict the selling price. And second, even if you did have the experience, you can’t scale this. Imagine needing to predict the price for millions of houses in a short period of time.

With the right data, machine learning models can learn the patterns in the data and give you good predictions. You can easily scale this to predict the price of millions of houses in seconds.

When done correctly with the right data, we can combine AI that can do cognitively simple tasks (like seeing things in an image) for complex means. For example, using computer vision to find malignant tumor sites or language processing for translation. These are cognitive tasks where researchers can train the AI to pick up on complex patterns and structures required to do the task.

How can you use this AI for your work?

So we’ve seen all these cool things that AI can do but how can it specifically help your business? We’ll take a look at three different ways you can use AI, and while this isn’t an exhaustive list, they are three of the major ways AI can have an immediate impact.

Customized customer experience

The first application is customized customer experience. We can use a customer’s actions and interests to customize their experience while they use your product. For example, Netflix uses movies you previously watched and other features to recommend a new movie for you. They also use this information to decide on what new content to create. This is a great way to offer a unique experience to your users that can differentiate your product from others.

We absolutely need machine learning here because it’s very difficult to recommend something to all your different users, who have varied interests, backgrounds, etc. and doing this at scale is very difficult. But don’t just offer recommendations because it uses machine learning or because you can. Ask yourself what customization would actually be of value to your users while respecting their privacy.

Optimize internal processes

Another avenue for applying AI is to optimize internal processes. This can assume many different forms for your business but look at what your teams are doing day to day. What are some of the tedious tasks that they are doing repeatedly using data from excel or images, text, audio etc.? These tasks shouldn’t involve creativity or complex problem solving but are time-consuming tasks for your team.

Going back to our housing price example, let’s say your team of realtors is inundated with requests from online customers who want to know the selling price of their house. You can use your team’s valuable time to do this process repeatedly or use the data you are collecting to build a machine learning model to do it instead.

This offers several different advantages. First, you free up valuable time for your team to work on other things like actually selling properties. Second, with the right data, the model can make great predictions for a variety of different inputs. Let’s say for each house we have hundreds of pieces of information; this becomes a very difficult task even for an experienced realtor to process. Third, your customers are asking you this because they find value in your predictions. With models, you can respond to them in seconds instead of the delayed response you would get from your human team. In fact, offering services like this can give you a competitive advantage in your industry.

Optimizing internal processes does not mean replacing jobs. This is about removing tedious tasks and using your teams for their creativity and problem-solving skills. And it also doesn’t mean going around and trying to solve everything with machine learning. Really look at the problem and assess the worth of using machine learning to help.

Aid in decision making

The third application of AI is to aid in decision making. So far we have seen examples where AI is responsible for the entire decision. For example, using the customer’s previously-watched movies to recommend a new movie, which is then directly shown to the customer. Or using data to predict the house selling price, which is then revealed to the user. But we don’t always have to use AI like this.

Instead, we can use AI to aid in the decision-making process. For example, let’s say you are deciding what your company budget should be next year. You can use machine learning to forecast expenses, instead of just estimating using last year’s values, so you get a good idea of how much you can spend. Or let’s say you want to know if investing in a new market is a good idea. You can use models to process tweets about a specific market and see how people feel about it. This one signal alone won’t help you make your decision, and it shouldn’t, but it can help get you there.

It’s especially a good idea to use AI in this way when the decision at hand is a very sensitive and complex one. Going back to our tumor detection model, this is a very complex task and we shouldn’t trust the AI to directly report to our patients. Instead, doctors can use what the AI predicts and use it in their diagnosis, especially when it picks up on things they’ve missed. We always need to maintain control but AI can surely play a large role in helping us make a well-informed decision.

Every business can benefit from these application themes but there are few things to keep in mind. First, you should always start with the business problem and never find yourself saying things like “we need to put our data to good use”. Next, you need to come up with a problem that’s worth solving and then think about data and AI. And finally, you also shouldn’t rush to using machine learning. Don’t predict something just because you can. First ask yourself if you can solve it using simpler methods (tables, graphs, simple code, etc.) and if not, then think about using AI.

Goku Mohandas is an AI researcher in Silicon Valley with a focus on using deep learning for natural language processing. He also works on democratizing practical AI for business and strategizing scalable machine learning solutions at ExposeAI.

You might also like More from author

Comments are closed.