AI explained: an overview for busy people
A machine is “intelligent” if it can independently take actions to maximize its success within its perceived environment – without additional directives from humans. But what is considered intelligent today won’t be considered so in a decade. Modern intelligent systems don’t just act on their own, but they learn how to do so based on contextual information and new input.
Growing in popularity across the board
Cheap, accessible computing power and data resources combined with the support of big players like Microsoft, Google, Amazon and others have led to the rapid development of effective new machine-learning algorithms and groundbreaking innovation in the field of deep neural networks – known as ‘deep learning’. Training deep networks takes huge quantities of processing power and data, which is now available to players large and small. Even more, most of these algorithms and innovative approaches are open source, simplifying reuse and adaptation.
All industries have their sights set on AI
While big players with specialized research labs have the resources needed to drive new developments, small and medium-sized companies are increasingly reliant on AI. Any business or organization that recognizes opportunities to improve processes, products and customer services should absolutely consider AI as a core technology for these evolutions.
To give just a few examples of success stories, Lee & Ally is a chatbot developed by legal firm deJuristen that provides legal advice. Swarovski created an intelligent visual search system to identify pieces of jewelry, and Rolls Royce uses predictive maintenance to reduce errors and cut fuel costs.
AI is being used in diverse applications
Driven by high-quality data and applied to mature processes, AI application areas include computer vision, natural language processing, forecasting, classification recommendation engines and clustering.
- Computer vision: the intelligent gleaning of information from images, such as quality, object detection and location, alphanumeric characters and similarities.
- Natural language processing: used to power smart chatbots – matching questions to commands – and automatic document tagging, among others.
- Forecasting: predicting numerical values to gauge, for example, product demand or consumer response to marketing campaigns.
- Classification: answers yes/no questions, e.g. “will this customer churn?” or “will this employee leave the business in the next year?”
- Recommendation engines: connecting the right products with the right customers at the right times.
- Clustering: grouping similar customers together to approach them in more personalized ways based on their actions, preferences or behaviors.
Any player that sees the opportunity to improve processes, products and customer service should absolutely set their sights on AI.
Quickly build on existing solutions
Classification, forecasting and recommendation engines are mature technologies that are being used in ever more diverse and rich ways to improve all kinds of business and operational processes. Computer vision and natural language processing are tougher, but new deep learning techniques have made these solutions more accessible and easier to use.
ImageNet and COCO are examples of open-source deep learning models that have been trained using extremely large datasets. It’s possible to develop your own computer vision solution by taking the core model and retraining it using images relevant to your process.
A successful AI project needs diversity
The right skills are needed to tackle AI challenges, but these skills are more and more in demand – and becoming more and more common. These include:
- Data analysts: transform data into useful formats for decision-making though ETL work, visualization and statistical analysis.
- Data scientists: statisticians that expand on the work of the data analyst by applying the correct statistical and machine learning techniques, translating a business question into an experiment and delivering the resulting insights.
- Data engineers: tackle big data problems such as unstructured data, continuous data or huge volumes of data from a software development approach.
- Machine learning engineers: deep learning experts that grow the depth of data models to improve the accuracy (or other metrics) of a simple model. Strong in calculus and linear algebra.
Also critical to the organization, management and thus success of your project are process specialists, product owners and scrum masters. Investing in different profiles ensures a well-balanced team with the ability to handle any situation that your business might run into. However, choose curious people that enjoy learning to boost the quality of your solutions and ensure the continuous improvement of your team.
Looking for a co-innovation partner to help you add value to your business through AI? We have developed a robust methodology and 6-step approach to AI project success.