How ensure success of AI & Machine Learning Projects

  • A recent survey by Deloitte of “aggressive adopters” of cognitive technologies found that 76% believe that they will “substantially transform” their companies within the next three years.
  • However, when it comes to AI & ML, there remains a large gap between aspiration and reality e.g., recently Gartner estimated that 85% of big data projects fail due to one reason or another
  • Here are some practical tips on how to ensure success of AI / ML projects and convert them into results:
  • 1. Have a clear objective in mind: AI projects need to framed in the context of your business model, context and culture, with an objective in mind around what it aims to achieve
  • 2. Communicate the purpose in a clear manner: While change always has to be driven from the top, implementation is always driven lower down, and hence its important to clearly communicate the purpose and expectation
  • 3. Select the right use cases: AI is quickly automating routine tasks and helping humans shift to non-routine ones, and selecting these use cases allows to have higher success without losing customer experience
  • 4. Understand your data well: As AI and ML progress, data biases are becoming a massive problem for many organizations - it's important to understand data source and context and resultant limitations well
  • 5. Define success in a reasonable manner: It's always fancy to achieve a 99% accuracy and in that process many models end up being over-fitted. At times it's better to have aim for a simpler model which is 70% accurate
  • 6. Consider the people factor: Automation and AI initiatives sometimes create job insecurity for many and hence it's important to build reassurance and shift humans to higher value tasks after automation is achieved

Don't forget to share this pointer!

View more comments +