Artificial Intelligence

• Richard Frost

  Does anyone know of any researchers who are using computerized speech games to build cognitive profiles of the game players?

  We are developing computerized speech-only games which are to be used to help create cognitive profiles of the game players. Player input is converted to text by a VXML-based speech recognizer, andWe are developing computerized speech-only games which are to be used to help create cognitive profiles of the game players. Player input is converted to text by a VXML-based speech recognizer, and computer output is converted to synthesized speech. The games are based on tasks used by psychologists when assessing children for learning abilities/disabilities. Does anyone know of similar work elsewhere, or of any work on categorizing verbal tasks performed by humans from a "computational" perspective. We realize that standard metrics such as computational complexity are not appropriate.

  Recent replies ⋅ Show All (6)

  • 
    Richard Frost replied

    Thanks for all the input. I shall review and respond over the weekend. For a demonstration of our Speechweb please go to Youtube: http://www.youtube.com/watch?v=Axa-n4etdZE Also some of myThanks for all the input. I shall review and respond over the weekend. For a demonstration of our Speechweb please go to Youtube: http://www.youtube.com/watch?v=Axa-n4etdZE Also some of my publications on NL parsing are on my website: http://cs.uwindsor.ca/~richard/from_janus/welcome.html Thanks Richard

    2 days ago
• Miles Brundage

  Does anyone know good publications that address the relationships between computational complexity theory, neuroscience, and artificial intelligence?

  Some examples of possible intersections I'm interested in: the tractability of automating different tasks, implications of the complexity of the brain for comparing the computational complexity ofSome examples of possible intersections I'm interested in: the tractability of automating different tasks, implications of the complexity of the brain for comparing the computational complexity of different domains of human intelligence (ex. vision, action, abstract reasoning, etc.).

  Recent replies ⋅ Show All (49)

  • 
    Uli Sattler replied

    or try http://www.springerlink.com/content/6204615622770564/ where present an approach to determining the cognitive complexity of justifications (logical theories in a decidable fragment of firstor try http://www.springerlink.com/content/6204615622770564/ where present an approach to determining the cognitive complexity of justifications (logical theories in a decidable fragment of first order logic whose computation complexity is well-studied) for entailments of OWL ontologies. We introduce a simple cognitive complexity model and present the results of validating that model via experiments involving OWL users.

    3 days ago
• Akash Agrawal

  Is it possible to control a car with a computer?

  Is it possible to control a car with a computer?

  Recent replies ⋅ Show All (6)

  • 
    David Friedlander replied

    You may want to check out the DARPA Grand Challenge, for example: http://en.wikipedia.org/wiki/DARPA_Grand_Challenge

    5 days ago
• Ravi Chaurasia

  I want to begin studying Machine Learning. Could anyone advise me on where I could start so that I'll can efficiently cover the basics?

  If possible suggest a good book or resource. Thank you!

  Recent replies ⋅ Show All (26)

  • 
    Kemal Sokolovic replied

    Download videos and slides from Stanford online course for Machine Learning, and (if you are able to), you must get it!

    5 days ago
• Ahmad Hassanat

  Is there any fast way to figure out if my research idea is new or not?

  It takes very long time to find out, sometimes much longer than the research itself. Sometimes I ignore searching the literature and focus on the idea, but after a while I find it is an old one!!

  Recent replies ⋅ Show All (10)

  • 
    Felipe Trujillo replied

    Somebody has said: "There is nothing new under the sun". However, I think that you must not search if your work is new or not, instead of this you should looking for the research that is similar toSomebody has said: "There is nothing new under the sun". However, I think that you must not search if your work is new or not, instead of this you should looking for the research that is similar to your own and find the difference.

    6 days ago
• Sandipan Karmakar

  Can multiple decision attributes be accommodated in Rough Set based rule generation?

  I am working on Rough Set Methodology. I am stuck at a point in implementing the RST to my problem. I am having multiple decision attributed with multiple conditional attributes. I have not comeI am working on Rough Set Methodology. I am stuck at a point in implementing the RST to my problem. I am having multiple decision attributed with multiple conditional attributes. I have not come across any method which accommodates multiple decision attributes. Can anybody help me to tackle this problem? Suggested literature can also be provided. Thanks in advance.

  Recent replies ⋅ Show All (9)

  • 
    George Victor Meghabghab replied

    I have applied RST in the area of user's search behavior between searching and browsing. check myI have applied RST in the area of user's search behavior between searching and browsing. check my book: http://www.amazon.com/Search-Engines-Analysis-Users-Behavior/dp/3540774688/ref=la_B001JOZ7G8_1_2?ie=UTF8&qid=1337302371&sr=1-2

    10 days ago
• Jennifer Renoux

  Can anyone recommend some articles or authors dealing with information fusion in multi agent systems?

  I am beginning some researches about this topic and I am looking for a basis for my state of the art. More precisely, I try to find articles about crossing information (like in Data fusion),I am beginning some researches about this topic and I am looking for a basis for my state of the art. More precisely, I try to find articles about crossing information (like in Data fusion), increasing the credibility of information... Thanks in advance for your help, Jennifer

  Recent replies ⋅ Show All (6)

  • 
    Jennifer Renoux replied

    Thanks all of you for these references.

    13 days ago
• Steve Stavrev

  What is the most efficient way to categorize an unseen website?

  I am building a system that automatically tells the category of any website, by just putting the URL as an input. Of course, this is a classical classification problem. I was wondering: 1. WhatI am building a system that automatically tells the category of any website, by just putting the URL as an input. Of course, this is a classical classification problem. I was wondering: 1. What input data should I use for classification (what part of the website is most informative of its category - home page content, meta keywords, domain name, a mixture of the above?) 2. What classification algorithm to use, so that the processing time of a new website is minimal. I am considering Bayesian filters (one for each category) but it is not the most computationally effective one( since it will try to evaluate each website as many times as categories there are). Another option I am considering are neural networks, or maybe even SVM. Any suggestions are welcome!

  Recent replies ⋅ Show All (4)

  • 

    Comment protected

    Join ResearchGate now to read this comment.
• Frank Bandach

  Does WATSON (IBM super computer) answer questions with multiple correct answers?

  In the Jeopardy show, Watson competed with humans to answer questions, but all questions had only one correct answer*. Watson had to select between several probable answers to produce its resultIn the Jeopardy show, Watson competed with humans to answer questions, but all questions had only one correct answer*. Watson had to select between several probable answers to produce its result (its answer). Consequentially, if a question had potentially several correct answers, could Watson respond such answer? (i.e., which countries export petroleum - the answer will include several countries or more than a single country name). *Note, by correct answer I mean "justified" or that there is at least one document that justifies such answer, whether the answer is a lie or is the truth.

  Recent replies ⋅ Show All (28)

  • 
    Frank Bandach replied

    Nova, released an episode about Watson that shows the architecture of the machine. To watch it, please follow link: http://www.pbs.org/wgbh/nova/tech/smartest-machine-on-earth.html It so appearsNova, released an episode about Watson that shows the architecture of the machine. To watch it, please follow link: http://www.pbs.org/wgbh/nova/tech/smartest-machine-on-earth.html It so appears that Watson can answer several questions, but it could also include several incorrect answers (correct = justified; and incorrect = unjustified). Nova shows how the IBM team gravitated toward machine learning as the solution (basically, statistical patterns). The amount of work, and the gamble were immense and likely unappreciated by most (sadly, including myself, but no more). It is a truly an admirable effort shown by the researches. Unfortunately, in a negative note, the show also reveals another side of the equation, and that is the lack for an understanding that may explain "what language really is." Watson is a triumph of the human spirit but also a witness of our current scientific ignorance.

    20 days ago
• Yazdan Jamshidi Khezeli

  Hi, I intend to study Multiagent Systems. does anyone know any free online course to introduce me? Thanks in advanced, best regards, yazdan

  Online course.

  Recent replies ⋅ Show All (4)

  • 
    Steve Stavrev replied

    One of my favorite fields, along with Reinforcement learning itself :) Try the lectures of my ex-mentor: Shimon Whiteson, at the University of Amsterdam. Here is aOne of my favorite fields, along with Reinforcement learning itself :) Try the lectures of my ex-mentor: Shimon Whiteson, at the University of Amsterdam. Here is a link: http://www.damas.ift.ulaval.ca/~coursMAS/ComplementsH10/Vlassis-livredai.pdf

    20 days ago
• Hadia Mubeen

  HOW

  HOW CAN I DEFINE ARTIFICIAL INTELLIGENCE

  Recent replies ⋅ Show All (19)

  • 
    Martin Becan replied

    Dear Graeme, I believe that intelligence is a subset of brain functions. It may be biological or artificial and separate from the physical representation. Personality consists of qualia, which canDear Graeme, I believe that intelligence is a subset of brain functions. It may be biological or artificial and separate from the physical representation. Personality consists of qualia, which can not be programmed or copied. Even monozygotic twins have not the same qualia. Qualia are acquired during life, based on experience, environment, learning, etc. Qualia affect the transformation of perceptual categorization, value categorization or phenomenal transformation in the mind. At the beginning of this discussion is the question "How do I define artificial intelligence?" Also important is the opposite question: "When I say that the system is no longer artificial intelligence? How do I decide that the system of artificial intelligence is faulty?"

    http://www.umelymozek.cz/images/public/mozek1.JPG

    26 days ago
• André Wagner Mota Nascimento

  Metaheuristics GA or SA

  What is your preference regarding the metaheuristics techniques? In your opinion, which applies this method? Why apply this and not another?

  Recent replies ⋅ Show All (6)

  • 
    André Wagner Mota Nascimento replied

    Thanks!

    27 days ago

• Post protected

  Join ResearchGate now to read this post.

  Recent replies ⋅ Show All (15)

  • 
    Louis Brassard replied

    I like Kauffman and I think that I will like Longo. There are a small number of mathematician who have tried to theoreticize biology from the inside instead from the outside like neo-Darwinism. I like Kauffman and I think that I will like Longo. There are a small number of mathematician who have tried to theoreticize biology from the inside instead from the outside like neo-Darwinism. Cuvier, Lamark, D'arcy Thompson, Bertalanffy, Turin, Waddington, Bateson, Goodwin, Kauffman, etc. The new discipline of evo devo is promising.

    29 days ago

• Post protected

  Join ResearchGate now to read this post.

  Recent replies ⋅ Show All (12)

  • 
    Robert Fraczkiewicz replied

    My friend, different world, different techniques. Optical character recognition is an almost ideal man-manufactured world, where the level of noise in the data is very low. No wonder authors of theMy friend, different world, different techniques. Optical character recognition is an almost ideal man-manufactured world, where the level of noise in the data is very low. No wonder authors of the paper get away with overtraining. But try biological data, for example, where the noise level is very high (a lot of "contradictory data" using the paper's language) and you must stop training early (or do something equivalent) do filter the noise out of the useful signal.

    Apr 18, 2012
• Gopal Sakarkar

  A most important simulator used in Agent Based Technology

  Hi All , its a time to share a most important simulator used in Agent Based Technology. SeSAm (Shell for Simulated Agent Systems) provides a generic environment for modelling and experimenting withHi All , its a time to share a most important simulator used in Agent Based Technology. SeSAm (Shell for Simulated Agent Systems) provides a generic environment for modelling and experimenting with agent-based simulation That is Sesam (http://www.simsesam.de/) It is very user friendly and handy to make simulation of agent-based application...

  Recent replies ⋅ Show All (3)

  • 
    Saad Nafie replied

    you can also comment something

    Apr 27, 2012