Decoding AI and how it can help your business

Senior business executives and CEO’s are grappling to understand the difference between the interchanged terms; Artificial Intelligence (AI), Machine Learning (ML) and Natural Language Processing (NLP) and more importantly: how it be used to improve business, provide a better customer experience… And, generate operational efficiencies?

Definition of Artificial Intelligence

AI is simply any computer system that exhibits intelligent behaviour – one that isn’t programmed, it learns for itself.

It’s a subset of computer programming in general, though there are many computer systems that you wouldn’t really describe as being intelligent – a database for instance, arguably, does not exhibit “intelligence”, it just stores and retrieves data. But, what constitutes “intelligence” is a grey area and the boundary between AI and non-AI systems isn’t a sharp one – which is why it is difficult to tell which tools are truly powered by AI and not deep learning or machine learning (more on these definitions later).

How is AI beneficial to a business?

AI is the premium approach to analysis, offering these benefits:

• Programmatically achieving outcomes speeds up the process of analysis
• Frees up analysts to work on other projects that require more human intervention and insight
• Saves money through more accurate, prompt decision making

Here’s how it can work for specific sectors:

• Marketing campaigns. To create much more relevant messaging offers and creative to match the interests of potential customers.
• Online recommendations. To tailor products or services to customers based on their previous viewing or purchase history
• Credit and loan application decisions. Faster, more accurate lending decisions (improving the customer experience and protecting the lender from potential bad debt by lending to someone that cannot afford a loan, whilst also protecting the consumer)
• Healthcare. In some areas of medicine, AI is being used to generate automated diagnoses, most notably in the area of tumour detection. While AI can perform these tasks at a superhuman level, it is critical to understand how the systems produce their answers both because the cost of false negatives is so high and also to inform on treatment needs.

Watch the film here from our experts on which areas AI are delivering the best results:

Word of caution

• If sourcing an external specialist for support, make sure they are using genuine AI and not claiming to but actually using machine learning or another approach below.
• Some tools out there are not suitable and could be damaging. For example, in credit risk, it is fundamental that models are explainable so that a lender can explain to the regulators why they made a lending decision.

Definition of Machine Learning

Machine Learning (ML) is the application of algorithms that learn by themselves through experience. Common machine learning algorithms include decision trees, which derive a series of “if then else” rules automatically (and so in a sense are like automated expert systems), and regression models which essentially identify lines of best fit through the data.

To contrast with expert systems, instead of an industry expert deciding the logic, ML finds the variables and values that are of greatest importance and which can be used to most accurately predict the outcome variable. It analyses the relationships between the variables in the data and uses this to produce a model in effect playing the same role in this process as the industry expert does in an expert system.  The model is then used to make decisions.

How is it beneficial to a business?

• Improved efficiency - reduces the time taken to build complex models, compared to using industry expertise

Word of caution

• We need experts to select the correct ML path to pursue.
• Achieving good results also sometimes requires careful and creative data preparation to produce good transformations of the raw input data that are useful in the model – a process known as “feature engineering”.
• This requires manual effort and expertise – though still considerably less than manual review or even creating an expert system.

Definition  of Natural Language Processing

Natural Language Processing (NLP) is a sub-field of machine learning that is focused on enabling computers to understand and process human languages, to get computers closer to a human-level understanding of language. It's everywhere. You probably use it now without a second thought it’s that transformative – Amazon’s Echo and Google Assistant.

A quirk of language models is that they can be used to generate text, mimicking the corpus from which they were trained.  You do this by feeding it a starting word (or sentence), then taking the predicted word as its next input and repeating the process. When you do this something magical happens. Take the below diagram an example of what might happen for a model trained only on Shakespeare…

How is it beneficial to a business?

Here’s some business questions that you could answer through the use of NLP:

• What do my employees think of my company? (based on articles from Glass Door)
• How should I group this content on my website to optimise for search engines?
• Which judge is likely to be more sympathetic to my client’s legal case?
• Which of my marketing content doesn’t match our brand ‘tone of voice’
• Which of my overdue debtors are likely to respond to litigation?

Using LEGO to understand where AI fits

There are numerous ways to play with LEGO. We can follow the instructions included in the box, step by step. Or we can ignore the manual and instead make up our own process keeping in mind the aim is to build a spaceship. Or we can go totally free format and try to figure out what is the best way to use available pieces.

These three approaches are analogous to three key categories of machine learning algorithm:

• Supervised Learning models learn from datasets that contain the right answers for every example (where the next brick should go, given how the previous bricks have been used) and the training process aims directly to reproduce those answers.
• Reinforcement Learning also has a target outcome in mind, but instead of case-by-case, guidance is based on a reward system that quantifies an overall goal (Are the bricks configured as a spaceship?) and the algorithm must learn which steps to take to maximise reward.
• Unsupervised Learning the goal is to explore patterns and features of the data (what configurations of the bricks are possible).

Top tips from our experts

We asked our experts and here’s the key things to consider for AI:

• Don’t use AI for the sake of it
• Run a small prototype test first before jumping into a large project
• Involve your employees in a workshop for ‘light bulb moments’ to problem solving
• Don’t invest heavily in day one – be cautious
• Consider your strategy, how AI can help now and, in the future.

Want more? Watch this short video...

Summary

The possibilities are endless. And because of this, AI and machine learning is set to become fundamental in the future of business. So, whatever your background, having a clear understanding of the terminology and concepts is key to navigating through all the hype to help with a range of tasks and prevent you get left behind competitors.