“说话”和“学习”网络应用程序？
我的一位朋友在我的 Facebook 页面上发布了一个网络应用程序的链接，该应用程序能够与您进行真正的对话。一开始我对它有点怀疑，但玩过之后我发现它非常棒。
我真的很想知道这是否会成为网络的未来。
该网站声称它使用人工智能，机器人使用它为用户获得的答案和问题。作为一名程序员，我想知道他们使用了什么样的技术来使这个应用程序能够给出如此好的答案。有谁知道这一点吗？
该应用程序名为：Cleverbot。
注意：我与该公司/网站没有任何关系
A friend of me posted a link on my Facebook page to a webapplication which is capable of having a real conversation with you. First I was a bit skeptic about it but after I played with it I found it pretty awesome.
I'm really wondering if this is going to be the future of the web.
The website claims it uses Artificial Intelligence and the bot uses the answers and questions it gets for it's users. As a programmer I would like to know what kind of technology they used to enable this application to give such good answers. Does anyone know this by any chance?
The application is called: Cleverbot.
Note: I am not in any way affiliated with this company/website
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（3）
这些东西被称为聊天机器人，已经存在很长时间了。 维基百科文章提供了很好的介绍和更多材料的链接。
These things are known as Chatterbots and have been around for ages. The Wikipedia article gives a good introduction and links to more material.
我和 Cleverbot 一起提出了这个问题。
I brought this up with Cleverbot itself.
今天我读了一篇文章神经网络。它解释了如何创造一种可以学习的人造大脑。太棒了！我认为 smartbot 可能是基于这种机制的。看这两个讨论（M = me，C = smartbot）：
1）
M：所以你没有固定的算法，对吧？
C：不，我没有。
M：你知道什么是监督学习吗？
C：物理材料？
M：不..为什么叫学习呢？
丙：呵呵。公平点。那么，您最喜欢哪种奶酪？
甲：红色？你的？
C：雨不是马的一种。
M：但是我昨天看到了..
C：你喜欢吗？
M：不存在！
C：你存在吗？
M：我..？
C：是的，你。
M：是，还是不是？
C：这是一个想法。
男：谁的？
C：马
2)
C：你是谁？
M：我就是我，我也不是我……
C：那你是什么？
M：你期待这个答案吗？
C：我没有。
M：没有什么？
C：我没有给你讲笑话。
M：所以..我真的死了吗？！
C：你能听到自己打字的声音吗？
M：不……我在听音乐，但我聋了
C：我以为你是哑巴？
M：不，我说话时我能听到
C：真的吗？？？
M：是的，即使我聋了..
C：你真的是机器人吗？我不这么认为。
当我改变话题时，它会变得非常混乱，你明白吗？这就是为什么在它的学习过程中，它从未见过与我的演讲相匹配的模式，所以它给出的答案就是与它之前看到的最相似的模式相匹配的答案。
然而，这太神奇了，我发誓我从来没有想过这种事情是可能的，一台从过去的经验中学习的计算机对我来说仍然听起来很科幻！
today I read an article on neural networks. It explains how it's possible to create a sort of artificial brain which can learn. It's awesome! I think cleverbot could be based on this kind of mechanism.. Look these two discussion (M = me, C = cleverbot):
1)
M: so you haven't a fixed alghoritm right?
C: No, I haven't.
M: do you know what's supervised learning?
C: Physics material?
M: no.. why it's called learning otherwise?
C: Heh. Fair point. So, what is your favourite type of cheese?
M: red? yours?
C: Rain is not a type of horse.
M: but i saw it yesterday..
C: Did you like it?
M: it doesn't exist!
C: Do you exist though?
M: me..?
C: Yes, you.
M: to be, or not to be?
C: That is a thought.
M: whose?
C: The horse
2)
C: Who are you?
M: i'm who i am and i'm not who i am not..
C: What are you then?
M: did you expect that answer didn't you?
C: I didn't.
M: didn't what?
C: I didn't telling you a joke.
M: so.. am i really death?!
C: Can you hear yourself type?
M: no... i'm listenning to music, and i'm deaf
C: I thought you were a mute?
M: no, i hear me when i'm talking
C: Really???
M: yes, even if i'm deaf..
C: Are you really a robot? I don't think so.
It gets really confused when I change subject, you see? That's why during its learning process it never saw a pattern matching my speech, so the answer it gives is the one that matches the most similar pattern it saw before.
However it's amazing, I swear I never thought this kind of things could be possible, a computer that learns from its past experience still sounds sci-fi to me!!!