1. Introduction #
The Floatchat platform offers a range of settings that can be configured in the Train tab’s Settings section, located on the left panel. These settings are related to smart responses and bot conversations.
2. Settings #
2.1 Related Matches #
The Related Matches feature allows the bot to not only provide the best response from the training dataset but also offer other options that were closely ranked.
This conservative approach gives the user the choice to select the most appropriate option.
Once enabled, you can customize the prompt message shown when displaying related matches.
Additionally, you can set a maximum limit for the number of related matches displayed to the user. The number of options processed depends on the matching responses and their scores.
2.2 Conversational Context #
Context plays a vital role in understanding user intent and providing accurate responses. Floatchat simplifies achieving context awareness by introducing a checkbox called “Enable Contextual Response” in the Settings section of the Train tab. When this feature is enabled, the system processes the user’s queries using NLP techniques, extracts relevant key terms, and persists them for subsequent queries. This helps in generating more meaningful and accurate responses by leveraging the stored context.
Entities extracted from intents are also reused to enhance the quality of responses where applicable.
2.3 Small Talk #
Floatchat includes a built-in small-talk engine that automatically responds to typical questions like “How are you doing?” or “Good morning.” This feature requires no training and provides basic responses. However, you can retrain the small-talk responses using the FAQ module, overriding the default pre-set responses. To add variations and make the bot appear more human-like, you can use the Randomize Messages node.
Sample small-talk bot responses:
– User: How are you doing?
Bot: Wonderful! Thanks for asking.
– User: Who are you?
Bot: I am a virtual being, not a real person.
– User: Good morning
Bot: Good morning! How are you doing today?
2.4 Block Text Input for Send Message With Options #
This setting disables keyboard input when clickable options are presented to users. Note that this feature is exclusive to Website Chatbots.
2.5 Programmable Delay #
You can introduce a configurable delay between bot responses to make them appear more human-like and allow users enough time to read through the messages before new ones appear. This setting enhances the overall user experience, especially when there are multiple sequential messages. Access the “Bot Response Delay Duration” option in the Settings section of the Train tab to set the delay. The maximum delay is 3 seconds, adjustable in 0.5-second intervals. By default, no delay is set.
You can also control the delay at an individual message level using the Pause Node in a bot flow.
2.6 Spell Check #
Enabling this feature allows the bot to perform spell-checking on user queries.
2.7 Advanced Semantic Lookup #
Enabling this setting activates an alternate matching engine that improves response accuracy.
2.8 FAQ Autocomplete #
Enabling this option provides users with a dropdown menu of frequently asked questions (FAQs) relevant to the keyword entered in the text box.
This is how it appears to the end user!