Chatbot and Artificial Intelligence

Pandora ITSM chatbot is made up of two services: a conversation or chat server (with its WEB client) and a conversational artificial intelligence engine (optional).

• Automatic: It uses machine learning models to automatically provide answers to users based on prior learning.
• Manual: A Pandora ITSM operator responds to users through the chat interface.
• Mixed: Pandora ITSM assists the operator by showing possible answers to users' questions.

The architecture consists of three key elements:

• The chat client, with which the user interacts.
• The chat server, with which the operator interacts.
• The prediction engine, which offers answers based on users' questions.

Both the client and the chat server are part of Pandora ITSM installation.

By default it is installed when using the cloud installation method. To activate it, go to the chat Setup → Setup → ChatBot → Enable chat option:

The web service URL must be configured with the public URL of your Pandora ITSM installation, as configured in the main section of Setup:

Activate Enable SSL, by default it has the following values:

Replace your own certificates with the ones configured by default. Save by clicking Update and check the green color in Check chat server to know if it works properly.

curl -sL https://rpm.nodesource.com/setup_12.x | sh
yum install -y nodejs
npm i -g pm2
cd /var/www/html/pandoraitsm/extras/chat_server
npm update
pm2 start server.js

Edit the file /etc/systemd/system/integria-chat.service so that the service starts automatically every time the operating system boots, adding the following:

[Unit]
Description=Integria-Chat-Server
After=network.target
[Service]
Type=simple
ExecStart=/usr/bin/node server
Restart=always
# Consider creating a dedicated user for Wiki.js here:
User=root
Environment=NODE_ENV=production
WorkingDirectory=/var/www/html/pandoraitsm/extras/chat_server
[Install]
WantedBy=multi-user.target

Save file and run:

systemctl daemon-reload
systemctl start integria-chat.service
systemctl enable integria-chat

Access to the chat server database must be configured, edit the file /var/www/html/pandoraitsm/extras/chat_server/config/config.js and modify the required database parameters. The one by default must be changed should you have a custom installation.

Run the following commands as root (super user):

yum install python3 python36-Cython
wget http://xxxxxx/prediction_engine-latest.tgz
tar xvzf prediction_engine-latest.tgz
cd prediction_engine
./install.sh
service prediction_engine.service restart

This will start the service on port 6000/tcp.

Support → Chat → Channel manage menu.

The AI is configured separately in each channel.

• IA url: URL where the webservice, the artificial intelligence engine, listens. It must be in format:

HTTP://DIR_IP

• IA Port: By default 6000. It should not be changed unless you have had to do a port forwarding somewhere in between.
• Initial response time: Seconds in which the AI will answer the first user request.
• Response time between conversations: Response time between questions once the conversation has started.
• Progress bar timeout: The time the system will allow the operator to choose an answer from those provided, before automatically answering the best option.
• Certainly threshold: The degree of uncertainty supported by the AI when choosing an option from those offered, the value should be between 0 and 1. An initial value of 0.5 (50 %) is recommended and adjust as necessary.

• API Key: Token to authorize connection to https://api.openai.com/v1/chat/completions.
• Model: Selector to choose between gpt-4 (default value) or gpt-3.5-turbo.
• Max tokens: Maximum number of “tokens” (default 2000) that can be generated in the response.
• Context time: The time over which the context of the conversation is maintained. In each call it is necessary to send the history of the messages sent by the user and those sent by the AI assistant, plus one at the beginning system (see in Context). Default: 01:00.
• Context: At the beginning of each request, this message is added in system. It provides guidelines that affect how the model will respond. To this context the sentence Always answer me in ${responseLanguage}, regardless of the language in which I write to you. is added, where responseLanguage will be the user's language if it is available among the channel languages, otherwise it will use the default language of the channel. By default: You are a Pandora ITSM assistant.
• Temperature: It controls the randomness or creativity of the generated answers. It admits values between 0 and 1, the closer to 0, the more predictable and coherent the answers are, the closer to 1, the more creative the answers will be. By default value: 0.3.
• Avatar: Image to be displayed in the chat when interacting with ChatGPT®. By default: moustache1.png.
• Title: Title that will be displayed in the chat when talking with ChatGPT®. By default: ChatBot.
• Subtitle: Subtitle to be displayed in the chat when chatting with ChatGPT®. By default: Default Bot.

• Before carrying out any customization, it is recommended to back up the current models:

cp -r /opt/prediction_engine/models /opt/prediction_engine/models.bak

• To restore models from the backup, run:

cp -f /opt/prediction_engine/models.bak/\* /opt/prediction_engine/models/

The question and answer template is generated from a CSV file with the following format:

Language code;Question;Answer

Example:

“en”;“question 1”;“answer 1” “en”;“question 2”;“answer 2” …

It can be obtained automatically from Pandora ITSM database with the following commands:

rm -f /opt/prediction_engine/data/integria_kb.zip 2>/dev/null
echo "SELECT id_language AS lang, title as question, data as answer FROM integria.tkb_data INTO OUTFILE '/opt/prediction_engine/data/integria_kb.csv' FIELDS TERMINATED BY ';' ENCLOSED BY '\"' LINES TERMINATED BY '\n';" | mysql -u integria -p integria && zip -j /opt/prediction_engine/data/integria_kb.zip /opt/prediction_engine/data/integria_kb.csv

To update the models:

cd /opt/prediction_engine/src
python kb_train.pyc

Menu Support → Chat → Browse → Data management.

In Pandora ITSM Web Console you will be able to add conversational items classified by categories and languages.

To search for a specific conversational item, enter a keyword found in the title and/or answer and select the category and language it belongs to.

Advanced configuration

The conversational model from YAML files with the following structure:

categories:
- category 1
- category 2
-...
conversations:
- - text 1
- text 2
-...
- - text 1
- text 2
- text 3
-...
…

The .yaml files (the file name is not important) should be entered in the directory:

/opt/prediction_engine/data/chat_xx

Being xx the ISO code of the language to be updated (es for the Spanish language, en for English language).

To update the models run:

cd /opt/prediction_engine/src
python chat_train.pyc

Pandora ITSM chatbot uses channels to define different places to have conversations between operators (Pandora ITSM users with special permissions to manage conversations) and normal users (normal Pandora ITSM users or simple anonymous visitors, if the chat is used from outside of Pandora ITSM).

Menu Support → Chat → Manage Channel:

The channel must have at least one operator, who will answer the users' chat requests. You will be able to add any user with chat operator permissions and assign them a description, an avatar (different from the one in their user file) and the languages in which they can answer chats.

Click Edit users of the corresponding channel to add a user and then click Add User:

Use the screen to define an operator, example:

Once added other users will be able to start chat.

In order for operators to receive audible notifications when users open a chat, they must be on the chat control screen. It is accessed through the menu Support → Chat → View chat.

The operator will enter one of those chats and interact with the other party:

In the event that the chat is opened by an internal user, it will indicate which user it is.

Support → Chat → Channel Manage → Generate Bundle menu, in the corresponding channel.

To use the chat outside Pandora ITSM interface, for example on a web page or in another application, click the star icon to show the snippet of JavaScript code to be embedded into the application.

Back to Pandora ITSM Documentation Index