The recently held Wrangle event hosted at the ADAX office in Bangsar South, saw an impressive list of international speakers gracing the event. Being its 2nd installment with the first held last year, the event was well received with a packed room at the ADAX HQ.
Big Community was privileged to have a quick chat with Hong Cao, Head of Data Science at Ernst and Young and pick his thoughts on the topic of AI, Machine Learning and data Wrangling and how its effecting the modernising digital society.
Hong Cao is responsible for setting up and managing EY’s ASEAN data science team, and leads the data science projects through the pipeline starting from presales, scoping to delivery. He is also a data science evangelist and advises the higher management on current data science technologies and the undertaken strategy. He has a research background with prominent scientific publications in top-tier venues of machine learning, data mining and signal processing. His work in data-driven image forensics received the best paper award in IWDW 2010 and an honorary mention in ISCAS 2010. He has also won multiple international data science competitions, such as GE Flight Quest in 2013 and OPPORTUNITY activity recognition challenge in 2011. He currently chairs the IEEE signal processing society, Singapore chapter.
“I believe AI and Machine Learning still have a long way to go to become advanced. But perhaps this is a good time because there’s more and more attention put on the AI and Machine Learning. I think especially with some of the cases like people have demonstrated in (the) past,” he said adding that platforms such as IBM Watson showed their mettle by beating the human counterpart in games.
People’s interest are definitely peaked in this domain. And not always for positive reasons as Hong reveals.
“People are worried about losing their jobs to AI. But I think its how you think about it. AI is best used to do something repetitive about work.” He went on to explain that AI can assume work with some intelligence, but work that has little or no interest for workers.
Granted those are good job scope targets to be replaced, he believes that AI needs to work together with human beings to do a much better job. Jobs that take time to go through massive amounts of data might be a good place to start.
“Because most of the jobs, especially at certain levels like specialists, they really need a lot of help to do their job much better.”
With AI tasked to augment certain processes, this leaves the task of completing or adding the finishing touches by human hands much simpler and reducing the time to completion by many fractions.
“People are worried AI is getting superior. Its like in the movies. Skynet,” he joked. But he believes with the right regulations, people’s fears will be addressed and this will no longer be an issue.
He shared that the biggest hurdles today is no longer new technology, but the regulation of that technology. Once the security measures are put in place to protect peoples interest, then there will be reduced fear as technology such as autonomous vehicles will have the proper guidance and regulatory mandates in place to protect human beings from any untoward incidents.
Hong also shared with us the importance of having good data to work with.
“How do we ensure we get good quality data so we can trust in it more, is a very important aspect. In technology, people have to look at the source (as to) where the data has been acquired and how they have rules and controls there. Also with some organisations, especially the user groups (can) be responsible to take care of certain sets of data. They need to be data stewards to identify (between good and bad data) and be responsible to maintain (the data) so it can be useful.”
- January 2018(59)
- December 2017(71)
- November 2017(72)
- October 2017(75)
- September 2017(65)
- August 2017(97)
- July 2017(111)
- June 2017(87)
- May 2017(105)
- April 2017(113)
- March 2017(108)
- February 2017(112)
- January 2017(109)
- December 2016(110)
- November 2016(121)
- October 2016(111)
- September 2016(123)
- August 2016(169)
- July 2016(142)
- June 2016(152)
- May 2016(118)
- April 2016(60)
- March 2016(86)
- February 2016(154)
- January 2016(3)
- December 2015(150)