Skip to main content

Home/ INF 6107 Web social/ Group items tagged predicting

Rss Feed Group items tagged

dumontjose

Analytics and Predictive Models for Social Media - 0 views

  • Analytics & Predictive Models for Social Media
  • Part 1: Information flow in social media (slides) Collecting social media data Extracting and tracking the flow of relevant information Correcting for the effects of missing and incomplete data Predicting and modeling the flow of information Identifying networks of information flow Part 2: Rich user interactions (slides) Predicting and recommending links in network Modeling tie strenght Modeling trust and distrust, frieds and foes How users evaluate one another and the social media content
  •  
    Tutoriel intéressant de l'université Stanford sur les modèles prédictifs pour les médias sociaux
travelmaniac

Why researchers want to build an AI that can predict a person's attractiveness - 0 views

  • what makes someone attractive?
  • Machine learning is a powerful tool that has already accomplished feats such as predicting biological sex with a 93 percent accuracy based on brain scans alone.
  • This resulting model would be able to tell us that, for example, John’s rating of “seven out of 10” by the public is primarily driven by his high intelligence, but held back a bit by his lack of muscle mass.
  • ...1 more annotation...
  • such a model would also be able to predict the attractiveness of new cases –
  •  
    Des algorithmes pour prédire le pouvoir d'attraction grâce à l'intelligence artificielle
erikhanley

When should someone trust an AI assistant's predictions? | MIT News | Massachusetts Ins... - 1 views

  •  
    Il est difficile de déterminer le moment qu'un algorithme de prédiction en intelligence artificiel est digne de confiance.Au MIT, une technique fut élaborée par les chercheurs pour aider les humains à déterminer le moment qu'on peut faire confiance aux prédictions.
Jean-Michel Dugre

CIO Priorities | Technology Research | Gartner Inc. - 0 views

    • Jean-Michel Dugre
       
      Les prédictions 2013 de Gartner
  •  
    Gartner Top 2013 Predictions
carolineproulx

Using Social Media Data to Predict to 2016 US Presidential Election | Social Media Today - 2 views

  •  
    Peut-on réellement prédire l'issue des élections présidentielles américaines de 2016 en analysant des données tirées des réseaux sociaux? L'avenir nous le dira...
Adrien Cherrier

5 predictions for social media in 2014 - Fortune Tech - 1 views

  • 2014 may mark the beginning of the end of abysmal customer service at major airlines, credit card companies, banks, and other repeat offenders,
  • Twitter, for instance, just unveiled a feature enabling paid Tweets to be targeted by zip code.
  • This kind of "geo-fencing," which Facebook has had since 2011, enables businesses to court nearby customers who might actually want to get ads offering special deals, in-store specials, etc.
  •  
    L'avenir et les développements possibles du Web Social.
  •  
    Le web social permet non seulement aux entreprises d'offrir un meilleur service à leurs clients en facilitant la diffusion d'informations, mais permet aux clients de partager leurs expériences entre-eux, parfois au détriment des entreprises. En somme, le web social force les entreprises à travailler plus fort pour maintenir la satisfaction de leurs clients.
Aurélie Bélanger

5 Predictions for the Public Relations Industry in 2011 - 0 views

  •  
    L'avenir des relations publiques bouleversé par l'avènement des nouveaux médias, une tendance irréversible... À quoi s'attendre pour 2011? - par Leyl Master Black
anonymous

Shirky: Ontology is Overrated -- Categories, Links, and Tags - 1 views

  • I want to convince you that many of the ways we're attempting to apply categorization to the electronic world are actually a bad fit, because we've adopted habits of mind that are left over from earlier strategies.
    • Caro Mailloux
       
      need of novelty
  • because it is both widely used and badly overrated in terms of its value in the digital world.
  • ...25 more annotations...
  • Yahoo is saying "We understand better than you how the world is organized, because we are trained professionals. So if you mistakenly think that Books and Literature are entertainment, we'll put a little flag up so we can set you right, but to see those links, you have to 'go' to where they 'are'."
  • You don't have to have just a few links, you could have a whole lot of links.
  • A URL can only appear in three places. That's the Yahoo rule.
  • They missed the end of this progression, which is that, if you've got enough links, you don't need the hierarchy anymore. There is no shelf. There is no file system. The links alone are enough.
  • One reason Google was adopted so quickly when it came along is that Google understood there is no shelf, and that there is no file system. Google can decide what goes with what after hearing from the user, rather than trying to predict in advance what it is you need to know.
    • Caro Mailloux
       
      Laisser les usagers se faire leur langage et le tagger à leur façon puis, en tant que Google, prendre cette info et l'utiliser pour créer une ''taxonomie''.
  • "Well, that's going to be a useful category, we should encode that in advance."
  • They point to the signal loss from the fact that users, although they use these three different labels, are talking about the same thing.
  • You can also turn that list around. You can say "Here are some characteristics where ontological classification doesn't work well": Domain Large corpus No formal categories Unstable entities Unrestricted entities No clear edges Participants Uncoordinated users Amateur users Naive catalogers No Authority
  • The other big problem is that predicting the future turns out to be hard, and yet any classification system meant to be stable over time puts the categorizer in the position of fortune teller.
    • Caro Mailloux
       
      ne pas prévoir d'avance
  • Here is del.icio.us, Joshua Shachter's social bookmarking service. It's for people who are keeping track of their URLs for themselves, but who are willing to share globally a view of what they're doing, creating an aggregate view of all users' bookmarks, as well as a personal view for each user.
    • Caro Mailloux
       
      chouette description concrète de l'utilisation de del.icio.us!
  • " If you find a way to make it valuable to individuals to tag their stuff, you'll generate a lot more data about any given object than if you pay a professional to tag it once and only once.
    • Caro Mailloux
       
      utilité du tagging
  • Tags are simply labels for URLs, selected to help the user in later retrieval of those URLs. Tags have the additional effect of grouping related URLs together. There is no fixed set of categories or officially approved choices. You can use words, acronyms, numbers, whatever makes sense to you, without regard for anyone else's needs, interests, or requirements.
    • Caro Mailloux
       
      Chouette description de ''Tags''.
  • The chart shows a great variability in tagging strategies among the various users.
  • But this is what organization looks like when you turn it over to the users -- many different strategies, each of which works in its own context, but which can also be merged.
  • We are moving away from binary categorization -- books either are or are not entertainment
  • But they either had no way of reflecting that debate or they decided not to expose it to the users. What instead happened was it became an all-or-nothing categorization, "This is entertainment, this is not entertainment." We're moving away from that sort of absolute declaration, and towards being able to roll up this kind of value by observing how people handle it in practice.
  • What you do instead is you try to find ways that the individual sense-making can roll up to something which is of value in aggregate, but you do it without an ontological goal.
  • you believe that we make sense of the world, if we are, from a bunch of different points of view, applying some kind of sense to the world
    • Caro Mailloux
       
      ''we make sens of the world together thru what's worth aggregating'' = not ontology 
  • we're going to be able to build alternate organizational systems, systems that, like the Web itself, do a better job of letting individuals create value for one another, often without realizing it.
  • If you think the movies and cinema people were going to have a fight, wait til you get the queer politics and homosexual agenda people in the same room.
    • Marie-Noëlle Therrien
       
      ¸Bel exemple pour démontrer la problématique.
  •  
    Un article de Clay Shirky qui nous donne son analyse de l'Ontologie, un point de vue intéressant sur les différentes façons de classer l'information sur le Web.
Francois Laroche

Predictive analytics projects can bolster business decisions - 0 views

  •  
    L'analyse des données et des informations permettent la prise de décisions en affaires. Les résultats peuvent être multiples et il est important d'en connaître les limitations.
hbaiebhana

Predictive analytics could be the future, but we must solve the data problem first | Ve... - 0 views

  •  
    L'analyse prédictive devrait être un guide qui aide à mesurer les performances, à prévoir de manière optimale et à éclairer la prise de décision.
Stéphane Braney

2018 - Nos 10 prédictions Social Media - Kantar - 0 views

  •  
    Avant même le scandale Facebook, la protection des données personnelles sont devenu une "top priorité" chez les jeunes. Il s'agit d'un article faisant état des médias sociaux qui se développent à vitesse grand V.
erikhanley

Last-mile routing research challenge awards $175,000 to three winning teams | MIT News ... - 0 views

  •  
    L'optimisation des opérations de routages, même minimes, permet aux entreprises des économie de coûts énergitiques importantes. Voyez comment, grâce à l'apprentissage machine, des équipes de cherchers ont déterminé des modèles de prédiction de routage efficaces, permettant de réduire leurs consommations énergitiques.
1 - 17 of 17
Showing 20 items per page