Analyzing and modeling complex and big data | Professor Maria Fasli | TEDxUniversityofEssex
HTML-код
- Опубликовано: 4 ноя 2014
- This talk was given at a local TEDx event, produced independently of the TED Conferences. The amount of information that we are creating is increasing at an incredible speed. But how are we going to manage it?
Professor Maria Fasli is based in the School of Computer Science and Electronic Engineering at the University of Essex. She obtained her BSc from the Department of Informatics of T.E.I. Thessaloniki (Greece). She received her PhD from the University of Essex in 2000 having worked under the supervision of Ray Turner in axiomatic systems for intelligent agents. She has previously worked in the area of data mining and machine learning. Her current research interests lie in agents and multi-agent systems and in particular formal theories for reasoning agents, group formation and social order as well as the applications of agent technology to e-commerce.
About TEDx, x = independently organized event In the spirit of ideas worth spreading, TEDx is a program of local, self-organized events that bring people together to share a TED-like experience. At a TEDx event, TEDTalks video and live speakers combine to spark deep discussion and connection in a small group. These local, self-organized events are branded TEDx, where x = independently organized TED event. The TED Conference provides general guidance for the TEDx program, but individual TEDx events are self-organized.* (*Subject to certain rules and regulations)
What an elegant and intelligent lady. I've watched a dozen videos about Data Science and this is one of the best and most in-depth ones I can find
🤣
prof, I was feeling sleepy but your voice made me woke up !!
Fascinating presentation. Thank you
This is very insightful. I had no idea it was common practice to stop at theory. Makes sense in the real world vs school.
Interesting! Thanks Professor!
Thank you professor it is very useful :)
Love her presentation...
This was a really inspiring talk, I came up with more ideas on how to build meaningful KPI's than the last 10 data talks I watched combined!
That is interesting talk about big data and marchine learning
That is interesting talk about big data and marchine learning
its the year 2018 and she pointed out an accurate prediction: "By 2018 there will be about ~ 4 Billion Internet Users" by December 31 2017 there was ~4,156,932,140 Internet Users
Cisco predicted it ;)
Perfect ending for this time near Christmas :)
It boosts our skills in Big data science software engineering.
She's awesome! 💗
Now that's an amazing talk on Big Data.
Data is unprocessed information, while information is processed data.
Interesting ... Thanks
WOWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW. I've never had a ted talk make me say WOW out load. That was beautiful, the part about social networks
Thanks
What program did she use to create the visual at 17:10
Very interesting Talk
Good !!!
Very nice 👍🏻
How do I get in touch with Santa? I would like to know what I want.
The adds I get are always for what I have, not what I want to get.
Me Enamoré
For software engineering freshers what would be better profession data analyst or developer?
16:50
The lord of the rings!
I have 2600 watch later RUclips videos. How do a manage?
Damn control freaks!
“I will leave it up to you whether your smart phone can take you to the moon.” 🥁🎭
Yes, her name was moon. My phone took me there but didn't told me her brothers were son of sun. 🤣
i'd like to sneak some big data somewhere too hehehe
There are so many issues with big data that models and analysis are pointless. People type information in the wrong fields, they type the wrong information in the right fields, they duplicate data and they fail to enter data. Data based on surveys, even if entered 100% correctly, are still suspect because people may not answer honestly, or not have an answer from the possible answers provided, or not understand either the question or possible answers and the survey itself may be designed with a bias. For example, car owner surveys. People who drop $30,000 on a car are less likely to say the car sucks because they're worried people will think they're morons for dropping $30,000 on a car that sucks. So look at car reliability surveys to see what cars get repaired the most! You think doctors will perform unnecessary surgeries and prescribe unnecessary medicines for kick backs but car dealers won't misdiagnose an issue for return business? How can you build a model or do an analysis that have any true value when the information collected and entered has no integrity?
rwfrench66 great analysis
I mean this just misses the point completely
probability theory + econometrics + economics = and you can deal with every issue you just mentioned with very reasonably chance of error.
Statistics my friend... Statistics...
Statistics is often a form of organised lie ....
no ps ta cabron
data gived you an indian accent
It is not an Indian accent. She is from Greece. So she has greek accent. Even it is, so what?
She definitely haved an accent...
See I believe that big data is not so complex its the way we see it makes it complex. According to me break the big data into small fragments and small patterns and understand the correlation. Let us take an example let's consider we are taking sample of a single person buying behavior in e-commerce just take his history of purchases like minimum of 10 purchases and you ll find a pattern look for the similar patterns in the whole data you can find similarities of it now do try to understand what is that which is correlating one and another you would end up in a particular factor influencing segregate that into that particular factor like wise if you keep taking little samples and following up the same processes I believe that you can atleast find a pattern behind it which may or may not be accurate but one thing is for sure you could understand the behavior or the factor that influencing which if enhanced or marketed to the right group and right audience might influence better shopping experience and more over it will make their shopping easy rather than looking at everything that doesn't mean nothing to them.
This is what I feel break big data into small fragments.
cax
Does anybody has the feeling that she does not really understand what she is saying ? Just an example: 2.5billion GB of data - > we are talking about millions of bytes...
No shit woman!
She knows quite well what she talks about, and she also said it right: 2.5billion GB of data - > "that's millions trillions of bytes.." check it out carefully my friend
Christos Alexiou Yeah it must be that-it couldn't possibly be because the best presentations clarify concepts and occasionally emphasize things so all people in the audience can follow along.
BERNARD MARTY my bad mate.
RDB Totally with you. But I am having a hard time thinking of a reason that somebody who does not have a clear image of what 2.5B Gb of data is, would be interested in such a presentation.
Christos Alexiou True, but the majority of TED viewers are online. I myself am fuzzy on the respective sizes and how they relate, I got here on a path through other vids (as you do) even though I don't know. That clarification is for people like me