locked
Artificial Intelligence RRS feed

  • Question

  • Artificial Intelligence (AI) is the area of computer science focusing on creating machines that can engage on behaviors that humans consider intelligent. The ability to create intelligent machines has intrigued humans since ancient times, and today with the advent of the computer and 50 years of research into AI programming techniques, the dream of smart machines is becoming a reality. Researchers are creating systems which can mimic human thought, understand speech, beat the best human chessplayer, and countless other feats never before possible. Find out how the military is applying AI logic to its hi-tech systems, and how in the near future Artificial Intelligence may impact our lives.

     

    friends, i have gathered certain information about Artificial Intelligence. i would like to know still more. if anyone knows, plzzzz pour the information.

    Saturday, May 5, 2007 6:42 PM

All replies

  • But how can we implement this AI in our real time Applications?
    Saturday, May 5, 2007 7:06 PM
  • infact the team which won the imagine cup, they had modified their search engine by using AI in it to search queries...
    Saturday, May 5, 2007 8:58 PM
  • AI is most popularly used in military systems and believe it or not, in games. Half-life 2 is one game which featured awesome AI. The programmers modelled the characteristics and responses of each player to be near human. If you want to know more about AI in programming, i suggest you view the following link for AI using ROR(Ruby on Rails)

    http://www.markwatson.com/opencontent/ruby_ai/book.html

    Saturday, May 5, 2007 10:33 PM
  • can i post more information about AI if u like friends??????

     

    plz tell 

    Sunday, May 6, 2007 7:29 AM
  • Guys,,,

     

    i have tried my best to provide information about Artificial Intelligence. The topic is really excellent.

    i need ur responses.

     

    here is another link for downloading AI

     

    http://www.MegaShare.com/167814

     

    Keep Sharing

     

    Bye

    Sunday, May 6, 2007 7:40 AM
  • we r using AI in almost all the programs without our knowledge...one very good example is the games that we play and develop...
    Sunday, May 6, 2007 3:08 PM
  • Sunil, if you have more info we can use, please share it... We all appreciate it. If you need any co-operation from our side, then we will try our level and prompt best.
    Sunday, May 6, 2007 3:54 PM
  • thnx Adnan for ur support.

    i will share still more info in next days. if u and anyone knows or have any good source of other information plz post it .

     

    Lets share the Knowledge

    Sunday, May 6, 2007 7:42 PM
  • will do!!!
    Sunday, May 6, 2007 7:58 PM
  • Guys,

     

    with ur guidance and support, i am posting another part of A.I.

     

    hope u all like it....

     

    http://rapidshare.com/files/30027452/The_term_Artificial_Intelligence.doc

     

     

    Lets Share the Knowledge

     

    Bye

    Monday, May 7, 2007 6:35 PM
  • Guys

     

    with ur support and guidance,  i amposting another part of AI

     

    http://rapidshare.com/files/30027452/The_term_Artificial_Intelligence.doc

     

    hope u all like it.

     

    Bye

    Monday, May 7, 2007 6:39 PM
  • The article was good, thanks sunil. I'll try to post some more relevant stuff after my exams...
    Wednesday, May 9, 2007 8:20 AM
  • try googling for more stuff on AI
    Monday, May 14, 2007 5:49 AM
  • thnx Adnan for ur Support
    Monday, May 14, 2007 7:16 AM
  • @ Sunil- The files seems to be deleted from rapidshare... Can u post it again ? Or if possible mail it me man.. My mail address is raghuramdcbe@gmail.com
    Friday, May 18, 2007 4:51 PM
  • Check out this software:

     

    Its called eliza. It is an expert system that uses NLP, Natural Language processing to chat with you in plain english, just like a normal human being.

     

    http://www.softlookup.com/display.asp?id=5929

     

    ECC-Eliza  v3.86 Description:

    is an interactive, command-line, computerized psychologist which uses various computer algorithms to process natural English sentences and to produce its own. ECC-Eliza is an amusing, artificially-intelligent psychologist which can chat with you in plain English, using advanced AI (Artificial Intelligence) methods, and solve your problems.

    It features a large database with thousands of permanent replies and hundreds of variable ones, and is customizable.

    Sunday, May 20, 2007 2:11 PM
  • Ok Raghram........dont worry, i'll mail to u

     

    Bye

    Sunday, May 20, 2007 5:34 PM
  • hey nice utility arijit.....
    Monday, May 21, 2007 10:25 AM
  • Thanks m8.

    To understand the position of AI in comparison to human intelligence, I would advice everybody to go through the works of Newell and Simon about humans as problem-solving systems and humans as information processors.
    They were the first to find out how exactly do we reason and can say pioneered the field of AI and Expert Systems.
    Link:
    http://www.post-gazette.com/pg/06002/631149.stm


    Tuesday, May 22, 2007 6:01 AM
  • I have also started a thread on "Search Engine Creation and Optimisation" and have put some white papers that u can go through to understand how AI and Algorithms are applied to the search engines.
    Link:
    http://forums.microsoft.com/SamVaad/ShowPost.aspx?PostID=1623593&SiteID=43
    Tuesday, May 22, 2007 6:05 AM
  • Nice info. In my coming semester i have AI as a subject. It will be fun to learn it : Interesting subject.
    Tuesday, May 22, 2007 4:33 PM
  • oh thanks man, that would be helpful.....i will have AI as an subject either in the 6 or 7 sem
    Wednesday, May 23, 2007 1:44 PM
  • WOW, good and lots of resources Arijit. Thanks for it.
    Wednesday, May 23, 2007 3:48 PM
  • How many of you are Interested in the NLP, Natural Language Processing Part of AI ??

    You can have a Look at this threads on a few interesting articles that I have posted:

    http://forums.microsoft.com/SamVaad/ShowPost.aspx?PostID=1623593&SiteID=43
    Tuesday, May 29, 2007 4:58 AM
  • Man i dont know anything about NLP. If you know then please throw some light onto it.


    Tuesday, May 29, 2007 4:55 PM
  • Sure m8.

        NLP (Natural Language Processing), deals with the creation of capability of Computers to interact and understand in a Natural Language, i.e Languages Spoken by Humans e.g. English.
    It is subdivided into 2 parts, a) Text Processing (Understanding of textual matter, in documents, be it printed, hand-written or stored as a file in a computer ; &; b) Speech processing (Understanding the spoken words in a language, spoken by another human being or recorded on a media)
    Wednesday, May 30, 2007 7:09 AM
  • WoW man, this seems to be really interesting. I wonder if ever computer can make it as accurate as humans Smile
    Wednesday, May 30, 2007 7:35 AM
  • Yes it is possible to some extent. In fact I have just made an engine that is able to understand .txt and.htm  documents, cluster similar documents and label them automatically.
    I have tested the labelling part with text from Wikipedia and it identifies the content with more than 96% accuracy and labels which are generated are either same as or better than the category specified by Wikipedia.
    Wednesday, May 30, 2007 8:55 AM
  • @Arijit - Man this is really cool. I cant resist to see a true demo of it Smile if you have some screenshots etc etc, then i would really wanna see it.
    Wednesday, May 30, 2007 5:28 PM
  • seems interesting man...since you have so much of darn knowledge about it then u must start developing your search engine man...
    Wednesday, May 30, 2007 5:29 PM
  • Hi,

    I would love to explain the details, but I would need some time to show u the screenshots, as I am within the company now and company policy restricts me from disclosing any material. I shall be resigning as an intern in June and joining them as a full time employee by the 11th. I shall post the details, as soon as I have put in my resignation.
    Thursday, May 31, 2007 7:38 AM
  • Ill be waiting for you to put the screenshots and demo if possible Smile
    Thursday, May 31, 2007 8:46 AM
  • Here are the algorithms that I have created for the purpose:

    A. Algorithms:

    1) For Overall System:

    1. Start
    2. Select Document Corpus.
    3. For every document in corpus do:
    4. Tokenize document
    5. Prune tokens (Remove Stop Words, Extract NOUNS and VERBS after POS (Part of Speech) Tagging and stem the word).
    6. Extract semantics(Domain, Hyperny, Holonym from Wordnet) of all pruned tokens and augment it with the pruned list
    7. Generate word vectors from pruned list
    8. If classification hierarchy exists then:
    9. Compare vector by finding similarity with all the other vectors in the hierarchy
    10. If vector is similar to all vectors in a cluster, add corresponding document in the cluster and re-label cluster
    11. Else re-cluster entire document cluster
    12. Else cluster documents through finding similarity with its vector with respective vectors of all other documents.
    13. Label every cluster newly created and store results as classification hierarchy.
    14. Stop

    2) Finding Similarity between two documents:

    1. Start
    2. From pruned token list, augmented with semantics of all pruned tokens of a document, calculate frequency of each word, thus generating an n-dimensional vector, with unit vectors (dimensions) as the token or semantic and the magnitude as frequency
    3. Calculate cosine of angle between the two vectors of documents to be compared.
    4. If cosine value > 0.5, the documents are similar
    5. Else if the cosine value <=0.5, the documents are dissimilar
    6. Stop

    3) Extracting labels:

    1. Start
    2. From token list of all documents in a cluster, augment list with the domain semantic of each token from wordnet as: Put Semantics of words and its frequency and if a word from the document is common with a semantic, put the word and change its freqeuncy as a sum of its original frequency with the frequency of the semantic.
    3. Convert the derived list into a vector, and extract the word with highest frequency as the most probable label of the cluster.
    4. Stop.

    Thursday, May 31, 2007 8:50 AM
  • Thanks m8 . But dont you think its too early for you to post the steps ? i mean you are still working, is that fine with the organization ? anyways, i didn;t understand much Stick out tongue may be a real demo will be good. i can wait for that.
    Thursday, May 31, 2007 9:02 AM
  • Hi Harshil,

    The algorithm stated above, is just the overall system working and actually includes a lot of intricacies and optimisations.

    I know it could be a difficult to grasp, but once explained the concepts are not too heavy.
    e.g Document/Word Vector:

    Consider a set of words as {Bowling, Batting, Batting, Cricket, Brian, Lara, Batting, Bowling}

    Converting it into a hashmap, we get a key-value realtion as:

    Word        Frequency
    Bowling          2
    Batting            3
    Cricket           1
    Brian              1
    Lara               1

    Now we know that a vector is represented as ai + bj + ck

    where i, j, k are dimensions and a,b,c are its respective magnitudes for a 3-dimensional vector.
    The hashmap above can be considered as a n-dimensional vector, with the word as dimension and frequency as magnitude. Thus u have a vector representation of a document, which u can use for cosine similarity ( dot product of 2 document vectors).

    Wordnet is a Thesaurus that contains, the most frequently used words of English and various relations between them like Hypernym (IS-A): Cricket 'is a' Game. Holonym(Part-Of): Engine 'part of' car etc.
    This I have used to get semantics(Meanings) of words. Earlier only word frequencies were used. I augmented it woth its semantics. e.g Even if the word 'Cricket' doesn't arrive even once in a document, I can still say that it talks about cricket, because, words like 'bowling', 'batting', 'fielding' etc point towards cricket.

    Please tell me if u need more clarifications.

    'Knowledge is always meant to be Shared'.

    Thursday, May 31, 2007 10:16 AM
  • The World Wide Web Virtual Library: Artificial Intelligence

    Artificial Intelligence Resources - highly recommended, including bibliographies, conferences, FAQs, journals, location index, newsgroups, publications, ...
    url:  archive.comlab.ox.ac.uk/comp/ai.html

    Neuron AI Directory: Artificial Intelligence resources

    Neuron's Artificial Intelligence Directory. Find web sites about AI or related areas: expert systems, neural nets...
    url: www.neuron.co.uk/

    AI on the Web

    This page links to 868 pages around the web with information on Artificial Intelligence. Some of the links will pop up additional information when you move ...
    url: www.cs.berkeley.edu/~russell/ai.html

    AI Resources

    Some of these resources are provided by AAAI; however many others are pages ... To learn more about the science of artificial intelligence, visit AAAI's AI ...
    url: www.aaai.org/Resources/resources.php




    Wednesday, June 6, 2007 6:59 AM
  • try this link buddy,
    http://www.aaai.org/AlTopics/html/overview.html



    Wednesday, June 6, 2007 9:30 AM
  • This is what i get from that link Stick out tongue

    We Couldn't Locate the Page You Requested

    We're sorry. The page you were looking for on AAAI's web site isn't located at the URL you clicked or entered. If the link you followed was old, you might try substituting “php” for “html” to see if the page has simply been converted to a different format. If you're looking for a specific topic, try the following links, or enter the subject on the search page:


    Wednesday, June 6, 2007 2:27 PM
  • I believe, Himanshu meant this link.

    http://www.aaai.org/AITopics/html/welcome.html
    Wednesday, June 6, 2007 3:46 PM
  • Yes that link works, Thanks Arijit

    i think he meant http://www.aaai.org/AITopics/html/overview.html
    Wednesday, June 6, 2007 5:53 PM