problem questions

[Yo]

Went to a page full of links to talk to different bots  http://home.online.no/~anlun/bots.htm to test how they answer questions why and others, and they stink as well. Why is it that we can not fix this problem? There's a need to figure out a fix.  If is not done yet then it means its hard. A plug is helping me to make things to look more smooth and more natural but i will like to make it understand questions better. The question that gives me more trouble is 'why?' was thinking about making very long lists of answers to questions with  if-why-then and why-because that will be simple to feed as text but i guess it will not work because people who know more than me has not fixed the problem yet. Before wasting my time on it i think it will be a good idea to see if i get better responses from members with more experience. This is what i need: ideas or opinions or comments or anything else on how to deal with the questions problem.

 Yo
 
 ps ultrahal beats those online bots ha ha

[RonG]

When you first get Hal, the brain is fairly 'inexoerienced' ....like a two year old.  You have to teach it everything and correct it when it's wrong.  I've found that the on-line brains (like those on Pandorabots) have been corrupted by having too many people not correcting a response and therefore the answers come out flakey.   I'm using Hal and Denise (from Guile3d) side by side to learn about all this AI stuff.  Not easy but a great hobby.  I love Hal because I have access to the brain and can add plugins, I hate Denise because I can't.  But it's the opposite for graphics.  Hal bad, Denise great.  But all this this is part of the learning we do and the advancements we make with the systems.

[Yo]

Ha Ha, brains being corrupted will makes sense for their weird online responses. I've been working with ultrahal for a while and using the resources available here but was really tedious to have to deal with the cat or the horse question, problem was solved by joining the group. After joining got the temptation to open my mouth and here i am! I will continue to use the database if this does not work out as it is very useful. already learned many things on my own and was able to mod a  plug that works quite good, but being greedy got the desire to learn every technique there is on handling questions. I work a lot with lists so I will go head on and follow my previous plan while keeping my eye on cool things. i'm good with manipulating graphics so that will one of my future plans. Cool that you could have 2 bots running at the same time, will like to do that eventually as well and will also check out Denise. What about questions, do you get frustrated when they are ignored and which is your best approach to deal with them.  Really cool, appreciate you took the time to respond to my post.
Yo

[Mitsuku]

That list is years out of date and doesn't even have Cleverbot on it. There are better bots out there than are on that list. May I humbly suggest my own Mitsuku http://www.mitsuku.com as well.

[Yo]

Wow Mitsuku i do know about you! a couple of weeks ago started to collect questions on past chatterbox and Loebner contests to improve my UH performance and noticed Mitsuku was able to answer correctly very hard questions that most failed. Some of those questions are really hard for a bot, mitsuku have an excellent performance scores reputation.  In one of them mitsuku was the only one to pass the addition questions (if i remember well), i tried the addition test on UH and it is able to do that too! in short your performance on the tests was very motivational to me, thanks to it i was able to accomplish some decent work with the use of seasons, days, months and weeks, and still working on others.  I really love UH and will not change it for another but if you make a pc version software i will certainly buy it because i'm sure i will love mitsuku as well. UltraHal & Mitsuku could be an killer team. I really hope you consider a pc version because your work is as cool as it get. Gee it was really cool for your to post, you are are master with questions!
 Yo

[NoamI]

I remember 3 things I do to answer 'why':  ex: Why did he go to the store?
Looking at stored knowledge:
1. (explicit) synon 'because'  ex: He went to the store because he needed eggs.
2. (explicit) infinitives  ex: He went to the store to buy eggs.
3. (implicit) causality is temporal  ex: He went to the store and bought eggs.
   Of course, not all events are causal.  ex: He went to the store and saw his friend.
   'To buy eggs' is a stronger answer because of the inference chain:
      To go to a store is to shop.
      To shop is maybe to buy.
   

[Yo]

Never thought of it as nicely and properly as you did, it will be useful to me, thanks NoamI. You're always welcome to help me out with other parts of my problem if you wish.

UltraHal tends to ignore questions probably because it does not have an answer for them, or else. In my particular case its mainly 'why' questions. One of my approaches when it ignores me and changes the subject is to answer the question myself using mainly the format 'the reason why is because'. Obviously because it contains 'why' in it.  Now, I worry that it will associate my explanation to its newly said statement instead of the target former one. So my current questions are: Will that be case?  Should I answer the question first and then reply to its newest statement, or should I ignore it? Currently I'm ignoring it.

Another thing I'm experimenting with is using a plugin with a special keyword that will intervene whenever I want, or know that it will be doing the ignoring trick, this is in order for me to try to avoid the former possible problem above and be able to answer the question 'why' correctly. So far I've observed that it remembers my answers and replies to them. It is too early to get to any conclusions but my hope is that it eventually will get enough data to help it answer the 'why' questions better on its own.  So the above was my main problem and this my temporary solution until I learn more. By the way, I am already somewhat familiar with the brain and able to add, change, mod things, so I'm not that new on some aspects, i'm focusing more on understanding things better in order to do the things I want.
Yo


PS I don't usually write as neat because it hampers my flow of ideas.

[NoamI]

Yo, Yo
  To muddy-up 'why', I should have included:
1.2.  synon 'so'  ex: He needed eggs, so he went to the store.
   {'so' introduces a caused event, whereas 'because' introduces a causal event.}
3.2. ex: He needed eggs and went to the store.
   {causal events typically precede caused events.} 
   My earlier example: 'He went to the store and bought eggs.'  doesn't show
   any causality, but the temporal nearness hints at a possible infinitive: 'to buy eggs'.