Nuggets of Data Gold - Computerphile

Поделиться
HTML-код
  • Опубликовано: 7 фев 2025
  • Data mining, why it's better than pure statistics. Professor Uwe Aickelin explains the basics of data mining.
    JPEG Discrete Cosine Transform: • JPEG DCT, Discrete Cos...
    Altair 8800 (NOT 8080!): • Computer That Changed ...
    Prof Steve Furber on BBC Micro: • Building the BBC Micro...
    Atari: Pushing the Limits: • Pushing the Atari Limi...
    / computerphile
    / computer_phile
    This video was filmed and edited by Sean Riley.
    Computer Science at the University of Nottingham: bit.ly/nottscom...
    Computerphile is a sister project to Brady Haran's Numberphile. More at www.bradyharan.com

Комментарии • 94

  • @sight92
    @sight92 9 лет назад +43

    This guy just called Google a lookup table, that takes balls.

  • @arcooke
    @arcooke 9 лет назад +21

    That was incredible. I watch every video posted on this channel and this one really struck me. Can we get more of this guy? Very interested in what he has to say.

  • @son-tchori7085
    @son-tchori7085 9 лет назад +41

    One asks what is the value of one divided by three, then multiplied by three.
    A mathematician answers "1".
    A physicist answers "0.999999..." (alternatively "0.99 ± 1%").
    A statistician answers "¿ How much would you like it to be ?".

    • @estiaanj8425
      @estiaanj8425 9 лет назад +2

      Son-Tchor I
      Great joke, but any good scientist would still use the exact values (if the values were exact to begin with) and therefore arrive at 1. :)

    • @Measurity
      @Measurity 9 лет назад +2

      Son-Tchor I Great joke. Though I think programmer would make more sense than physicist. Because physicists often use exact math. While rounding errors are very computer specific fault.

    • @NeXtdra42
      @NeXtdra42 9 лет назад

      Son-Tchor I the last thing you say is part of what a computer algebra system like mathematica does

    • @waheedullahkhan.
      @waheedullahkhan. 3 месяца назад +1

      Son-Tchor I
      Hi

  • @Haaknes
    @Haaknes 9 лет назад +1

    This guy should teach statistics. I want more of these videos!

  • @DanteD83
    @DanteD83 9 лет назад +3

    Data mining is just applied statistics... just without a hypothesis to start with. This is why it is much more prone to mistaking random patterns for causal ones.

  • @ricardoamendoeira3800
    @ricardoamendoeira3800 9 лет назад +2

    More from him and more data mining please!

  • @Friek555
    @Friek555 9 лет назад +5

    This professors is one of the first germans whose english doesn't make me cringe. I even thought he was british for a few seconds.

  • @jam99
    @jam99 9 лет назад

    A great example of a real problem still seen with a lot of scientific studies today. Researchers feel they must have a trend and so draw in a trend line with no real support from the data. Objectivity, integrity and honesty are essential to science.

  • @bachirontzki7087
    @bachirontzki7087 9 лет назад

    This video was quite interesting. Would love some more videos on this kind of topic and videos with Professor Uwe Aickelin.

  • @AwesomeCrackDealer
    @AwesomeCrackDealer 9 лет назад

    This is exactly the guy that came to Unicamp in Brazil to talk about Nottingham's CS research in big data and DIDN'T mention Computerphile! hahaha

  • @UberAlphaSirus
    @UberAlphaSirus 9 лет назад +5

    Is there anymore info regarding the benefits of higher BP?

  • @SkonTeam
    @SkonTeam 9 лет назад +2

    WTF ? you upload this just when i am writing a paper about data mining ! thanks !

  • @NickHuntingtonKlein
    @NickHuntingtonKlein 9 лет назад +20

    This guy has a very odd idea of what "statistics" is. No way of dealing with missing or unclean data? No way of uncovering nonlinear relationships? Absolutely not true. He's clearly knowledgeable about statistics but still seems to think these things, which makes me think he just has a weird definition for the term, which is nonetheless very misleading.
    Interestingly, the causality discussion at the end of the video is currently where machine learning is weakest relative to standard statistical (and especially econometric) methods, so it's a very odd inclusion in a video with a subtitle about being better than pure statistics. Causality isn't completely unadressed in machine learning (see the work of Judea Pearl on causality in ML coming from the computer science side, or Susan Athey from the econometric side) but it's much less well established, and many of the engineering and CS people applying these methods in industry are largely clueless about these advances, or even the concept of causality outside an experimental setting.

    • @francoaraya1561
      @francoaraya1561 9 лет назад

      Dedrgt

    • @tratbagd4500
      @tratbagd4500 6 лет назад

      He might have been referring to classical statistical methods. It is well established that the field of machine learning is based on and have contributed to statistical learning theory.

    • @biel96
      @biel96 5 лет назад +1

      @@tratbagd4500 That's unlikely. Classical statistics does include missing data correction and non linear models, at the very least.

  • @FYIitsnotme
    @FYIitsnotme 9 лет назад +1

    DRINKING GAME:
    Drink every time he says Data
    Drink every time he says Artificial Intelligence

  • @122mlb
    @122mlb 9 лет назад

    Wow, excellent talk about data mining/big data/machine learning. Good job guys!

  • @power-max
    @power-max 9 лет назад

    Once you figure this out, even more complex forms of those Capta's will be needed (the annoying things you have to type when attempting to register for things)

  • @PeterWalkerHP16c
    @PeterWalkerHP16c 9 лет назад +1

    A common problem that analysts encounter is where managers ask for this and that mean or average. They come to us with their solution, not their problem. Thus, they often get what they asked for and not what they wanted;

  • @JustPingo
    @JustPingo 9 лет назад +7

    What is hidden behind that square over the computer's screen?

    • @dsgreat3
      @dsgreat3 9 лет назад +10

      ***** A lot of data analysis and mining is done under NDA also if you are working with things like medical records or other personal information you have to worry about the data protection act. He probably forgot to turn off his screen before they did the interview.
      A place i used to work doing something thing similar although less complex you had to lock you pc even if you were getting up for just a few seconds.

    • @GabrielHawkPot
      @GabrielHawkPot 9 лет назад +6

      ***** It's a spoiler for the end of the video.

    • @PerMortensen
      @PerMortensen 9 лет назад

      GabrielHawkPot Why not just turn off the monitor or minimize the window or something though.

    • @PerMortensen
      @PerMortensen 9 лет назад

      GabrielHawkPot Why not just turn off the monitor or minimize the window or something though.

    • @ThoolooExpress
      @ThoolooExpress 9 лет назад +4

      His rarerest rare pepes...

  • @hungbahuynh7666
    @hungbahuynh7666 8 лет назад +2

    This video make remind me of tv series person of interest

  • @aidano1773
    @aidano1773 9 лет назад +1

    Loved this video, thanks :)

  • @OwenPrescott
    @OwenPrescott 9 лет назад +18

    There are some free MIT lectures on algorithms and AI worth watching.

    • @ivanaslamov
      @ivanaslamov 9 лет назад +3

      DJTripleThreat78 Just search for MIT Open Course Ware

  • @AstAMoore
    @AstAMoore 9 лет назад +2

    The true mystery is: What’s behind the gray rectangle mask over the computer display? We may never know . . .

  • @Bigwebwasher
    @Bigwebwasher 9 лет назад

    Wow, that thing with blood pressure is quite a cool example!

  • @tsjoencinema
    @tsjoencinema 9 лет назад +2

    Uwe Aickelin sound like just the man for James Harris "Jim" Simons company.

  • @Marius-vw9hp
    @Marius-vw9hp 5 лет назад

    The salt example blew my mind

  • @nosuchthing8
    @nosuchthing8 9 лет назад +1

    Amusing that a big data video appears a few days after some big data operations in the US are ruled illegal. Or alarming.

  • @integlangs
    @integlangs 6 лет назад

    In data mining how does one ensure that you aren't a victim of p-hacking, if you gave a sufficiently large data set?

  • @stavyadatta9650
    @stavyadatta9650 4 года назад

    Thatssss my professor!!!!!!!!!!!!!!!!!!!!!!!!!!!!

  • @Great.Milenko
    @Great.Milenko 9 лет назад +4

    professor Uwe has the most awesome accent ever........ im wondering where hes from.... anyone?

    • @TimOkolowski
      @TimOkolowski 9 лет назад +5

      AwesomeVindicator He is from Germany, but in my opinion his accent doesn't sound stereotypically german.

    • @Braminam
      @Braminam 9 лет назад +2

      Tim Okolowski it did sound typical german to me though, the way he said "dat" instead of "that". Im from the netherlands though so i might be a bit biased.

    • @cretin76
      @cretin76 9 лет назад

      He sounds like a friend of mine who is Austrian.

    • @devluz
      @devluz 9 лет назад

      AwesomeVindicator Sounds German to me because that is how I sound as well :p

    • @abeedhal6519
      @abeedhal6519 9 лет назад

      Braminam haha yea its obvious :D the dutch have the same accent when speaking english as the germans

  • @MikaelLevoniemi
    @MikaelLevoniemi 9 лет назад +2

    Google and facebook do big data mining in much larger scale than professor here actually indicates. To get that data is easy, to piece it up is hard. That's one reason Facebooks aquisition of whatsapp is already profitable for facebook. They got the one piece they didn't have. Phone numbers. Billions of them.

  • @Brissles
    @Brissles 9 лет назад

    I like this guy.

  • @canguar
    @canguar 9 лет назад +1

    he is so excited

  • @johno2940
    @johno2940 9 лет назад

    Right, looking up the data isn't that hard for Google, but finding a way to fit a website or image in their tables proooobably is

  • @herrfleischgewehr
    @herrfleischgewehr 9 лет назад

    very, very interesting.

  • @TechyBen
    @TechyBen 9 лет назад

    So it's just a different flavour of statistics?

    • @NeXtdra42
      @NeXtdra42 9 лет назад

      TechyBen he uses the word statistics like "deterministic calculations"

  • @SkukS
    @SkukS 9 лет назад +1

    "I drank water and therefore..." :D

    • @laharl2k
      @laharl2k 9 лет назад

      Skux it's funny how the maximum recommended daily salt douse here is 1g but then you go buy food somewhere where they give you those salt sachets and they come in 2g. Not to mention some foods have more salt than that alone.
      ....it really is all because of water.

  • @pizzicatoiv
    @pizzicatoiv 9 лет назад

    Google: it's just a lookup table.

  • @richardtowers6948
    @richardtowers6948 9 лет назад

    So the main take away from this is that a higher salt diet is a healthy part of a western lifestyle. I didn't expect medical advice on this channel, but I'm altering my eating behaviour right now based on this clear advice :-)

    • @emptybrink
      @emptybrink 9 лет назад +3

      Richard Towers You heard it wrong! He said they measured urine in salt. So stop peeing on your salt!

  • @ailijic
    @ailijic 9 лет назад

    More like this, please.

  • @Unplanted
    @Unplanted 9 лет назад +1

    Dude, Paper doesn't grow on trees, you know?

  • @apeek7
    @apeek7 9 лет назад

    It could be that salt has no significant effect on blood pressure and other contributors, such as stress (more stress in industrialized nations), dominate.
    Finland has the highest rate of heart attacks and 40% of the Finnish diet is derived from fat.
    But Greece has the lowest heart attack rate and 40% of the Grecian diet is derived from fat...

  • @MrTStat
    @MrTStat 9 лет назад +1

    At first I read dota mining xD

  • @mustafaadam9697
    @mustafaadam9697 9 лет назад +1

    Google is just a look up table .... ThugLife baby

  • @micksydney
    @micksydney 9 лет назад +1

    you can always manipulate a statistic to support an argument
    politicians make a living doing it and it appears doctors do as well !!

  • @fakkmorradi
    @fakkmorradi 9 лет назад

    I WANT TO BE SMART

  • @forrestgump8312
    @forrestgump8312 9 лет назад

    Data

  • @jacobparsons5298
    @jacobparsons5298 9 лет назад

    Statistics != linear regression

    • @austininflorida
      @austininflorida 9 лет назад

      Jacob Parsons Yeah. He had quite an interesting definition of what statistics is. Ignoring perhaps his misnomer, that linear regression near the end horrified me the instant I saw it. I just took the AP Statistics Exam, so all those rules are going through my head.

  • @Jorissoris
    @Jorissoris 9 лет назад

    Y YU NO USE THA BROWN PAPER!??????!!!!!!!!!

    • @anujbeatles
      @anujbeatles 9 лет назад

      ***** Haha I think it's an intentional move designed to be sort of complementary to the brown paper. It's the classic old-printer load paper.

    • @pollaeng
      @pollaeng 9 лет назад +1

      ***** Brown paper for numberphile this one for computerphile

  • @YamiBeast
    @YamiBeast 9 лет назад +1

    Nnnnnoice

  • @plankalkulcompiler9468
    @plankalkulcompiler9468 6 лет назад

    Oh, another example of insufficient popular big data explanation.

  • @CrimeMinister1
    @CrimeMinister1 9 лет назад

    I can't understand him every other sentace. He talks too quickly and quietly at times.

    • @abeedhal6519
      @abeedhal6519 9 лет назад

      Joebro8bit learn to listen its only you it seems

    • @laharl2k
      @laharl2k 9 лет назад

      Joebro8bit i had no problem and i'm not even native.

  • @1m2r3a
    @1m2r3a 9 лет назад

    Please stop using markers on paper.