More than 85% of AI projects fail. This high failure rate is a main reason why data science is still a “science.”
In this talk, we illustrate that domain insights are crucial for successful AI/ML initiatives, discussing three areas of concern: clarification of business context, awareness of nuances of data sources, and navigating organizational structure.
To clarify business context, we discuss the importance of understanding organizational vision and mission, the maturity stage of the organization, product, and feature, as well as techniques for crystalizing a project’s description, meaning, relevance, value, and purpose.
To be aware of the nuances of data sources, we discuss the biases, inaccuracies, incompleteness that inevitability exists in data, and provide examples in web session data, geolocation data, and financial transaction data.
To navigate organizational structures in specific industries, we interpret the stage of product development, the maturity of the data science infrastructure, and examples for being flexible to adapt to industry norms.
By rigorously applying domain insights in AI/ML initiatives, we can reduce the failure rate and improve teams’ confidence in executing successful AI projects.
MITCNC Members. Please log in for access to Registration Link
Offices of Mayfield Fund
2484 Sand Hill Rd
Menlo Park, CA, 94025
South BayMITCNC Event Sun, Nov 17 / 9:00 am PT
4900 Marie P. DeBartolo Way
Santa Clara, CA, 95054
San FranciscoMITCNC Event Wed, Jan 15 / 6:00 pm PT
First Republic Bank
One Front Street
San Francisco, CA, 94111