1 / 49

Intelligent Communication

Intelligent Communication. Outline. Introduction. Turing Test. Early Chatterbots. CleverBot. Priyank Parikh Nitant Vaidya Smit Patel. Watson and Siri. The Future. Introduction. Communication and Intelligence are strongly linked.

telyn
Télécharger la présentation

Intelligent Communication

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Intelligent Communication

  2. Outline Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Watson and Siri The Future

  3. Introduction Communication and Intelligence are strongly linked. Intelligence is integral to holding a meaningful conversation. Conversation must not just appear to be coherent, but actual information must be passed between two entities. “Conversation is more than an action, it is a process”

  4. Communication and Intelligence It is said that with an IQ gap of more than one standard deviation (i.e. about 15 points), communication between two people becomes difficult, and that beyond two standard deviations it is effectively impossible. In this seminar, we show you the different stages of development of man-machine communication (verbal), and a glimpse of what lies ahead.

  5. Outline Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Watson and Siri The Future

  6. Can Machines Think Alan Turing was one of the first persons to use the phrase “computer intelligence” In 1950, he published “Computing Machinery and Intelligence”. He begins the paper with the claim “I propose to consider the question 'Can machines think?‘” But how do you define “intelligence” and “to think”?

  7. Can Machines Think Turing replaces the question with "Can machines do what we (as thinking entities) can do?” To test a machine, Turing proposed to use conversation as a skill that required the ability to think.

  8. Turing Test – The Imitation Game Suppose that we have a person, a machine, and an interrogator. The interrogator is in a room separated from the other person and the machine. The machine and the person must answer questions that are addressed by the interrogator. The object of the machine is to try to cause the interrogator to mistakenly conclude that the machine is the other person; the object of the other person is to try to help the interrogator to correctly identify the machine.

  9. Turing Test – The Imitation Game C obtains responses from A and B. C must figure out by the end which of A and B is the machine, and which the man

  10. Loebner Prize The Loebner Prize is an annual competition in artificial intelligence that awards prizes to the chatterbot considered by the judges to be the most human-like. The format of the competition is that of a standard Turing test. In each round, a human judge simultaneously holds textual conversations with a computer program and a human being via computer. The judge will try and determine which one is which.

  11. Interesting Facts : • A one-time-only prizes that have never been awarded is offered for the first bot that judges cannot distinguish from a real human and which can convince judges that the human is the computer program. • The Economist, in an article entitled "Artificial Stupidity" published shortly after the first Loebner prize competition noted that the first Loebner winner's victory was due, at least in part, to its ability to imitate human typing errors.

  12. Outline Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Watson and Siri The Future

  13. Chatterbots

  14. Chatterbots Most chatbot programs approach the problem with a form of Case Based Reasoning. CBR is the process of solving new problems based on the solutions of similar past problems. A common implementation is pattern matching, in which the structure of the sentence is identified and a stored response pattern is adjusted to the unique variables of the sentence.

  15. ELIZA ELIZA is a computer program and an early example of primitive natural language processing. ELIZA was written at MIT by Joseph Weizenbaum between 1964 and 1966. It was one of the first chatterbots in existence. It recognized cue words or phrases in the input, and then generated output of corresponding pre-prepared or pre-programmed responses, so that the conversation can forward in an apparently meaningful way

  16. ELIZA – Behind the Screen ELIZA operated by processing users' responses to scripts, the most famous of which was DOCTOR, a simulation of a Rogerian psychotherapist. Using almost no information about human thought or emotion, DOCTOR sometimes provided a startlingly human-like interaction.

  17. A.L.I.C.E. A.L.I.C.E. (Artificial Linguistic Internet Computer Entity) is a natural language processing chatterbot which works by applying heuristicalpattern matching rules to the human's input. The bot uses an XML Schema called AIML (Artificial Intelligence Markup Language) for specifying the heuristic conversation rules. How does it work?

  18. AIML AIML, or Artificial Intelligence Markup Language, is an XML dialect for creating natural language software agents. AIML was offered under a free or open source license, and many "Alicebot clones" have been created based upon the original implementation of the program and its AIML knowledge base. AIML contains several elements. The most important of these are explained in the following slides.

  19. AIML Categories Categories in AIML are the fundamental unit of knowledge. A category consists of at least two further elements: the pattern and template elements. Here is a simple category: When this category is loaded, an AIML bot will respond to the input "What is your name" with the response "My name is John."

  20. AIML Patterns A pattern is a string of characters intended to match one or more user inputs. A pattern like will match an infinite number of inputs, including "what is your name", "what is your shoe size", etc. The AIML pattern syntax is a very simple pattern language, substantially less complex than regular expressions.

  21. AIML Templates A template specifies the response to a matched pattern. A template may simple like literal text, or use variables as shown Template elements include basic text formatting, conditional response (if-then/else), and random responses. Templates may also redirect to other patterns, using an element called <srai>. This can be used to implement synonymy.

  22. AIML Power of recursive substitution : ALICE uses the input to generate the output. It can submit input to itself using the <srai> tag and put the contents of * into the output using <star/>. These two tags form the building blocks of templates. Say we have the answer to“What is Linux?” -> “Linux is an Operating System”

  23. AIML Now we must map the question “Tell me what Linux is” to “What is Linux” and establish them to be synonymic. The <srai> tag establishes synonyms, and <star/> is used to insert the * from the input. AIML does everything this way. A.L.I.C.E. has 1300 rules merely to remove what it considers useless adverbs like really and accordingly in sentences.

  24. AIML AIML is considered clever and simple, and a good start for beginners writing simple bots. However, it is difficult to maintain and author more complex bots, and has weak pattern matching ability. After 15 years, A.L.I.C.E has a meagre 120k rules. It's easy to get false positives (matches you don't want) and hard to write more discriminating patterns.

  25. Outline Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Watson and Siri The Future

  26. CleverBot Cleverbot is a web application that uses an artificial intelligence algorithm to hold conversations with humans. Unlike other chatterbots, Cleverbot's responses are not programmed into it, but rather selected from phrases entered by humans in previous conversations. Cleverbotfinds all of the keywords or an exact phrase matching the input and after searching through its saved conversations of previous chats, responds to the input by finding how a human responded to that input in past conversations when posed by itself.

  27. CleverBot + Cleverbot took part alongside humans in a formal Turing Test at the 2011 Techniche festival at the Indian Institute of Technology Guwahati on September 3, 2011. Out of the 1,334 votes cast, Cleverbot was judged to be 59.3% human, compared to the rating of 63.3% human achieved by the actual humans.

  28. CleverBot - However, there are still large problems with Cleverbot. If humans that speak to it deliberately input inappropriate or contradictory statements, when it is later asked certain questions, it will yield inappropriate responses submitted by these previous users.

  29. Cleverbot …still have a long way to go!

  30. Outline Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Philosophical Implications Watson and Siri

  31. Watson - IBM Watson is an artificial intelligence computer system capable of answering questions posed in natural language developed in IBM's DeepQA project. IBM describes it as "an application of advanced Natural Language Processing, Information Retrieval, Knowledge Representation and Reasoning, and Machine Learning technologies to the field of open domain question answering

  32. Watson - IBM Data: The sources of information for Watson include encyclopedias, dictionaries, thesauri, newswire articles, and literary works. Specifically, DBPedia, WordNet, and Yago were used. Although Watson was not connected to the Internet during the game, it contained four terabytes of disk storage, including the full text of Wikipedia.

  33. Watson - IBM

  34. Siri Siri is an intelligent personal assistant and knowledge navigator which works as an application for Apple's iOS. Apple claims that the software adapts to the user's individual preferences over time and personalizes results, and performing tasks such as finding recommendations for nearby restaurants, or getting directions

  35. Siri Functions: • Reminder • Text • Weather • Meeting • Send email • Contacts • Set alarm • Directions • Stocks And of course, ask Siri about Siri

  36. Siri Looking at it from a programmer’s perspective, it seems to me that Siri consists of three layers: • a speech-to-text analyser • a grammar analyser, and • a set of service providers If all three of these work well, then Siri will be fun and helpful.

  37. Siri – Listening A speech-to-text analyser is a piece of software that takes audio and turns it into text. Siri has a much easier job because its job is limited and focused. It only has to understand words and sentences that pertain to appointments, contacts, messages, and maps. This makes it easier for Siri to pick out what you’re saying, because there are only so many things that you’re allowed to talk about

  38. Siri – Understanding An additional piece of software is needed to turn the text into something useful. This complex mapping of strings to functions is the job of a lexical and grammatical analyser. Grammatical analysis still comes down to searching a string for certain key phrases and using those phrases to build up a simple model of what the user wants to do and what he or she wants to do it to. Again, Siri’s limited focus on appointments, contacts, messages, and maps makes this technically viable.

  39. Siri – Replying Siri can send your commands toa service to return information. You already have a calendar app and you can press buttons to view and create appointments. Siri will push those buttons for you, in essence. You already have a maps app and you can search and find directions there. Siri will enter your search text for you, and can toggle traffic on and off by voice rather than by button.

  40. Outline Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Watson and Siri The Future

  41. What Lies Ahead… Most chatbots today have fixed rules and a finite, controlled set of things to say. The structure is nearly deterministic – but the future is all about being 'useful' - helping people find information, navigate a site, perhaps even offering limited advice. The future's more entertaining - it's all about communication itself, about companionship and the creation of digital selves.

  42. Example : 24x7 Psychotherapy A bot is being developed that has all the knowledge there is about psychotherapy and the human mind, and can guide the patient through the same steps as a professional therapist. A blind test that involved 1,400 people no one could tell the difference between the automated therapist and a professional one after a 30 minute session. The bot can also tell if a person is lying in the way he talks.

  43. What Lies Ahead… As artificial intelligence technology continues to advance at a blinding speed, the gap between humans and machines diminishes. Two illustrations of the above fact are: • Voigt-Kampff Test • Eliza Effect

  44. Voigt-Kampff Test[Do Android Dream of Electric Sheep] The book speaks of a future time when “androids” roam the planets. Earlier androids were easier to detect because of their limited intelligence. Now, with technology improving, the Voigt-Kampff test is used to distinguish humans from androids. It involves measuring empathetic responses to questions designed to evoke an emotional response. Android are not expected to show remorse, etc. It emphasizes the importance of including human emotion in new chat bots or talking machines.

  45. Eliza Effect Named after the ELIZA bot, it is the tendency to unconsciously assume computer behaviours are similar to human behavior, and attribute emotion to a bot. Example : An ATM which displays the words "THANK YOU" at the end of a transaction. A casual observer might think that the machine is actually expressing gratitude; however, the machine is only printing a preprogramed string of symbols. (Douglas Hofstadter)

  46. [note: lisa ̴ Elisa] !!!

  47. [rajesh and Siri]

  48. Thank You Introduction Turing Test Early Chatterbots CleverBot Priyank Parikh Nitant Vaidya Smit Patel Philosophical Implications Watson and Siri

More Related