Please don't make me. Once upon a time in a virtual galaxy far far away, there was a young, intelligent agent by the name of Siri. One beautiful day, when the air was pink and all the trees were red, her friend Eliza said, "Siri, you're so intelligent, and so helpful - you should go work for Apple as a personal assistant. So she did. And they all lived happily ever after. So what's the link between Siri and Eliza?
Are they related? My Profile Log Out. In fact, a factor that makes the computational processing of our language terribly complex is language variability , that is, our ability to say the same thing in so many different ways e.
If a bot operates in a strictly closed domain it is possible to gather the semantics of the most common questions that will be posed to it 2. Still, the main problem is not the semantics of the most common questions, but their form, which can vary a lot. For instance, in some closed domains we can build a list of FAQs representing what will be asked to a virtual agent. However, we will hardly have a list of all the paraphrases sentences with the same meaning of those questions.
As an example, consider the following sentences from [ 12 ] : 1. Another factor that makes language so complicated is the fact that it is inherently ambiguous. Sentences with different meanings emerge from ambiguity at the lexical level. For instance, some meanings of the word light are 3 : 1. In what concerns Dialogue Systems, several domains were explored from the early days. Particularly prolific were the conversational agents targeting the concept of Edutainment, that is, education through entertainment.
Army Recruiting Command as a hi-tech attraction and information source, and the previously mentioned Edgar Smith Fig. Open image in new window. Edgar Smith in Monserrate. Affolter, K. VLDB J. Ameixa, D. In: Bickmore, T. IVA Springer, Cham Androutsopoulos, I. Bernsen, N. Bickmore, T. Social dialogue with embodied conversational agents. Bothe, C. In: Lintas, A. ICANN LNCS, vol.
Budzianowski, P. Towards the use of pretrained language models for task-oriented dialogue systems Carpenter, R. Colby, K. Devlin, J. Association for Computational Linguistics, June Fialho, P.
Association for Computational Linguistics, August In: Rodrigues, R. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik Herzig, J. Association for Computational Linguistics, September Kong, X. Li, J. Li, T. Association for Computational Linguistics, July Lowe, R. Lu, Y. Mikolov, T. In: Bengio, Y. Miller, G. Montague, R. In: Thomason, R. Formal Philosophy. Selected papers of Richard Montague, pp. Patel, R. In: Gratch, J. However, all these conversations have one thing in common: They're all a human talking to a machine.
So what about device-to-device encounters? When he put the two together, a very interesting and long--but slightly disjointed--conversation ensued.
Past the typical greeting, which both programs carry off well, Eliza is clearly the better conversationalist. Posted on Jan 21, PM. Eliza was one of the first AI programs that "understood" language. It pretended to be a psychologist and would parrot things back with little understanding of what was actually said. Page content loaded.
Jan 21, PM in response to lkb4jsd In response to lkb4jsd. Jan 21, PM. Some back in the day took pride into double-talking Eliza into such responses as "At kendrated the yutzer at your first rickshaw?
Question: Q: Who is Eliza?
0コメント