Machine learning in production: Human error is inevitable, here’s how to prepare.
Blog #4
Website :
Student Name : Vanessa 黃妮莎
Student ID : D0726551
Machine Learning in Production :
Human Error is Inevitable,Here's How to Prepare
Machine learning and AI are likely to make $2.6 trillion in new business value for marketing and sales alone by 2022. Not to mention another $2 trillion for manufacturing and supply chain planning. This value of money is a big number therefore you need to guarantee the models you are working for will not fail you. To maintain the machine learning operational there are many factors need to be considered.
Of course, models are programmed by humans. But remember, humans aren’t perfect, there is this popular term called “human error. One small human error might result in a complete disruption and loss for any company. There are some standard factors needed throughout the continuous monitoring and governance policies to help your team steer the expected errors (human or others) and even inspire them to find new ways or opportunities for future model. The standard factors are :
· Maintaining visibility into each model’s data lineage
· Limit who can access and administer changes to models or environments
· Observe the technical, mathematical and business performance of every single model in production
Mistakes can happen and with the right decision, it can quickly be fixed. Your company must prioritize on understanding the capabilities of Machine learning and make sure to visualized reports possible and increase the ability to manage and report results. Reporting to C-level executives is significant for proving success and keeping investments into your initiatives a priority. Lastly, don’t forget to exploit analytics and outcomes in them.
In the end it’s all about visibility. Machine learning models only stand to improve from rough examination and transparent communication is essential to guarantee that you are maximizing visibility to them.
Agree! The machine benefits to human from human error. But we still have to check the visibility of the machine to be correct like what we thought. So that, it will be very greateful and more conveninet for human.
ReplyDeleteWilly Erico謝洪標
ReplyDeleteD0726476
I’m agree with this article, because Human error is inevitable, that’s why we(Human) need machine or AI(Artificial Intelligence), AI and Machine Learning have changed the ways different business activities are performed today. In the past, when companies used to perform all their activities manually & their production rate was quite low. Due to the increasing demand of customers, companies have started to use artificial intelligence and machine learning to create an artificial mechanism that can mimic a real person and give solutions to customers easily and quickly. AI and Machine Learning affect different areas of business and help them grow fast. Also, beside to help the Human error, AI and machine learning also can reduce the cost, because AI is a working tool that provide information and automate many workflows. It should be included as an assistant in solving daily tasks without having any fear of incorrect results. But over time, they easily understand the production process and start working quickly without making any error. This eliminates the need for too several employees on the company. Companies save a lot of money by introducing AI in daily business activities as it performs the work of many employees. Also, AI and machine learning can run the business effective and efficient but it doesn’t mean, in the company just only AI and machines, because AI and machines cannot run without humans, so still employees are needed and AI and machine learning are just supporting system for human to avoid Human error. Thankyou vanessa for publish this article, it really useful in business world.