N O T I C E
MSPbots WIKI is moving to a new home at support.mspbots.ai to give you the best experience in browsing our Knowledge Base resources and addressing your concerns. Click here for more info!
N O T I C E
MSPbots WIKI is moving to a new home at support.mspbots.ai to give you the best experience in browsing our Knowledge Base resources and addressing your concerns. Click here for more info!
Introduction
Analyzing the sentiment or overall tone of a ticket summary allows for the added layer of priority, especially if an MSP has multiple tickets from customers in a day. By returning and alerting techs directly or via messages in chat channels persons can be aware that a ticket has come in with a negative sentiment so they can choose to prioritize and work on the ticket immediately.
Sentiment is generated through the OpenAI integration using specific prompts and returns if a ticket's summary is 1. Slightly Negative; 2. Negative; and 3. Very Negative.
To Start Using OpenAI Sentiment Bot
- Setup OpenAI Integration (link), choose a billing limit, and complete the API integration setup with MSPbots.
- Once the OpenAI integration is completed and verified, navigate to the Bots section in the MSPbots app and search for OpenAI Ticket Sentiment Bot.
- Clone the template bot which will automatically add it to your custom ('My Bots') profile.
- Go into the 'My Bots' tab and click on the OpenAI Ticket Sentiment bot to open. Click on the Design Tab:
Bot Settings and Filters
You can customize filters for Specific Ticket Types, Boards, or Status, and others on the Data Filter (2nd) bot block,
The default filters for tickets that would qualify to be analyzed for sentiment are as follows:
6.
Other possible filters include: UID, Company Name, Owner, Minutes in Progress, Priority, or Ticket Type and Sub Types. You can add and remove custom filters using the and buttons.
7. The OpenAI bot block contains the settings and prompt for OpenAI to return sentiment on the ticket summary and the settings below allows the AI model to be fine-tuned.
Setting | Default | Description |
---|---|---|
Temperature | 0.2 | Temperature Common to All ChatGPT functions, it used to fine tune the sampling temperature between 0 and 1. Use 1 for creative applications and 0 for well defined answers. Example: if you would like to return factual or straight forward answers such as a country's capital, then use 0. For tasks that are not as straight forward such as generating text or content, a higher temperature is required to enable the capture of idiomatic expressions and text nuances. |
Max Length | 120 | Represents the maximum number of tokens used to generate prompt results. Tokens can be thought of as pieces of words that the model uses to classify text. |
Top P | 0.2 | Top_p sampling also known as nucleus sampling is an alternative to temperature (sampling). Instead of considering all possible tokens, GPT-3 considers only a subset, or a nucleus whose cumulative probability mass adds up to a threshold, the top_p. |
Frequency Penalty | 0 | Mostly applicable to text generation, this settings tells the model to limit repeating tokens. Like a friendly reminder to not overuse certain words or phrases. Since this is mostly not applicable to sentiment analysis, it is set to 0. |
Presence Penalty | 0 | This parameter tells the model to include a wider variety of tokens in generated text, and like the frequency penalty is applicable to text generation as compared to sentiment. |