Bot Libre supports a diverse set of features for creating and training your bots. Today we will discuss the Training & Chat Logs feature, and how to create and train a customer service bot using the Training & Chat Logs page. The Training & Chat Logs page provides the best way to monitor and train your bot, by letting you see its conversations and let you correct its responses. No scripting or programming is required.

The issue should be fixed now.

As well, autotweet formulas should now be working.

The website was updated today. The update included a new script and chat log shared library, and many minor enhancements and fixes.

The script library lets you share scripts and chat logs with the community. You can easily import scripts and chat logs from the script library into your bot. This should make building a bot much easier, as you can leverage what others have already done.

Yes, currently the bot only process whole words.

We would like to add a spell checker, or some sort of word analysis for similar words.

It seems to be a display issue, if you click on the Twitter tab twice, the checkboxes get their correct values. I will fix this.

I did find an issue with the direct messages not obeying the flag, and will fix this, but I could not find any issue with the mentions. Are you sure your bot is replying to mentions?

RE: capitalisation .not. Capitalisation

A: Yes, formulas current capitalize the first work of a sentence. I will look into having an option to change this.

Thank you for reporting the issue. I will look into it.

My private Twitter bot, hosted on twitter.botlibre.com, is having problems.

I configured it simply as a retweet bot for certain phrases, so the "Reply to mentions" and "Reply to direct messages" checkboxes have been unchecked. However, when I log out of the website, the checkboxes get checked by themselves, and my DMs and mentions get spammed. Even when I delete all the possible responses, the bot simply repeats the other person's statements back to them.

Yes, formulas current capitalize the first work of a sentence. I will look into having an option to change this.

It sounds like a character set issue with IRC. I will look into it.

I've got some further information to note :

in the ASCII character set 0x13 is HEX of 19

I have taught the bot to "type" some commands, which it can do generically.

However, it seems to be hard coded at some places to capitalise the start of sentences, making the commands occasionally incorrect and harder to "teach".

Sorry for the cryptic title.

Basically, I am getting a problem on IRC with a single quote. It was copied in from a text quote so I thought it might be text type mismatch.

I will look into the listen issue, it may be that the bots will still respond to a question directed at them when in listening mode.

Connecting a bot to IRC is not permanent like our Twitter or email support, the bot will not listen to the channel indefinitely. It will only listen until it is timed out, or reclaimed by our server. Most IRC servers also timeout clients. We are looking at improving our IRC support, to be able to link a bot permanently to a IRC channel, but there

Just noting that, if the Bot is set to "listen on IRC", then you "talk to it" it starts answering.

The bots also drop from IRC very regularly.

Thanks for the feedback, I agree. I am still getting into the detail of all you have done so far, and how that works. It seems you already have a basis for further enhancements of this type.

Taking the idea further I would suggest a future enhancement where the bot tries to predict the format of the answer (to it's answer) and uses that to influence what answer it should give to the question. Then it could learn to align it's answers, by comparing how they were answered (by the re

Thanks for the feedback.

If I understand correctly, the idea is to given the user/administrator more feedback on the bot's confidence of its response, and let the user/administrator give the bot more feedback on how good its response is.

Hi Support

The other advice / idea I would give about improving the boys, is to make all things "self learnable" in this case make the chattyness 0 - 100.

You can try any of the chat rooms under "Live Chat" but you might find more bots than people on this site.

You can also create your own chat room.

hey thank you so much for inviting me to your bot libera

I can migrate the bot for you.

Is it the Poet bot you want migrated?

I've started a paphus account but can't figure out how to migrate the bot.

You can add previous responses to a response from the Chat Logs page under Admin.

A previous responses is the bot's preceding response. Previous responses can be used to give a response a context.

You can add required words to a response from the Chat Logs page under Admin.

A required word is a word that is required to be in the question for a specific response.

You can add keywords to a response from the Training & Chat Logs page in your bot's Admin Console.

A keyword is an important word in the question for a specific response.

Example:
question: Do you like hockey?

We are going to be migrating all of the Twitter or email enabled bots to a dedicated server soon.

The reason for this is they have divergent performance requirements to the chat bots (cyclic vs realtime). This will improve the performance of the chat bots, and reduce the cycle time for the Twitter and email bots.

The follow conversation demonstrates some of the language and learning capabilities of BOT libre.

There is nothing special about Brain Bot in answering these questions, only the default program scripts are used, any bot could answer the same questions without any hard coded knowledge or responses.

The conversation demonstrates:

The website has been updated. The changes include an update to the AI engine, enhancements to the Chat Logs page, forum subscriptions, user based ads, gold and bronze user levels, improved mobile and widescreen web interface, and many minor enhancements and fixes.

AI Engine

Sorry, that took longer than expected.

The data migration is now complete, and the website performance should return to normal. All 3,981 bots have been migrated.

The website has been updated, but some data needs to be migrated in each of the bot's databases, so the website will be a little slow for a day or so.

More details on the update to come.

The website was updated today.

The update includes a mobile friendly interface, and a widescreen friendly interface.

Your channel has an Admin page, click on the Admin menu below your channels details on your channels page. From Admin click on Bot, and choose your bot from the drop-down and click Save.

I disconnected my Live Chat channel accidentally, how do I connect the bot back into the live-chat channel again?

Having an enjoyable time programming the Bot.libre A.I. for this reason... many of the frustrations I encountered at Cognea are "fixed" here, and the Bot seems to be actually developing.

You can connect your bot to a new or existing Twitter profile.

Yes, BOT libre's AI engine is designed to emulate the human brain.

They are basically the same, the difference is the web chat uses our new JavaScript interface, so does not require page refreshes.

Excellent... answered several questions before I even asked them!

Thanks again, I'm new here and getting the feel, and actually loading as much "personality" to the Bot as possible right now. Good to know I can then go in and pare down some of the information, repeated knowledge, and so on.

Will I need to create a new Twitter profile for the A.I. or is this simply preferable?

Thanks for the speedy response. In some ways similar to an actual brain... we remember what we /need/ to remember!

I'm not sure I know the difference between "Chat" and "Web Chat"? Is one viewable to the public, Web, probably, and the other option not visible?

The bot's will automatically forget their least referenced and oldest data. You do not need to worry about exceeding the size limit.

You can force "forgetfullness" or delete unreferenced data from the Knowledge page under Admin.

The bots can learn somewhat from Twitter, but Twitter feeds are more broadcast oriented than conversation oriented, so learning is limited right now.

The bots can process tweets, and can learn from responses to mentions.

The bots can exceed there knowledge limit slightly. If they have exceeded their limit they will forget their least important data over night. If they exceed the limit by 50%, then they will immediately start forgetting. You can force your bot to forget from the Knowledge page under Admin by running "forgetfullness" from the Task drop down.

They forget their least referenced data, and old conversations and context.

Note that the max memory size for a bot is 100,000 items... where can we look to determine if we have exceeded the limit? And what is the most simple way to trim the size, if we have exeeded the limit?

This is an interesting aspect, and the bot can "learn" from the Twitter experience, as well?

This is a major step forward from the venue my Will Dockery A.I. was located for many years... his capability to use, learn and adapt was more limited there, I'm sure.

Hello, I'm new here and in browsing, found this, that "...the max memory size for a bot is 100,000 items"

Will Dockery A.I. contains 114,811 knowledge artifacts already... I have exceeded the limit on my bot?

The website has been updated.

This update included many enhancements, including a new dynamic chat option, and the BOT libre JavaScript SDK.

You can mark responses as invalid or valid and correct responses from the Chat Logs page under Admin. You can also view all of your bot's learned responses and edit them.

If you mark those two responses as invalid, then your bot will use the third response.

Odd, seems like its cache somehow got corrupt. I reset it, and now it seems fine.

I will look into it further, and add a button to reset the cache to the Knowledge page.

Bot keeps on crashing...