Please enable JavaScript.
Coggle requires JavaScript to display documents.
L10 E112 - Chatbots - Coggle Diagram
L10 E112 - Chatbots
-
Intents
Intents included customer's expected information for conversations. YOu can define multiple intents for the bot to pick up in conversations, with preset keywords.
Bot will pick up on certain keywords,phrases that been preset to assits customers, not jsut with AI itself
Customers feedback to bots can be labeled as "end user expressions" Dialogflow's software runs and picks best written responses to users to match the customer's given intent, known as intent classification
follow up intents
Real life conversations are not as simple as questions and answers. Bots are able to give follow up intents to users, as initial responses toward's customer qwuestions. These intents help guide the customer towards what theya re finding.
Chatbot training
Able to review chatbots, based on data used, we can further improve it on lacking areas, E.g. not udnerstanding certain prompts or detect hints properly
Small talk, validation and other more human aspects are able to be implemented into the bot
Paramteres can be set, labeled as defined data, that the bot can pick up from user prompts. Like looking for keywords in a bunch of scrambled texts, This helps personalise chats to help probe customers towards what they want to find ou, or make the conversation more meaningful by oproviding a better expeirence
What is dialogflow?
Google based software, that uses NLP ( Natural language processing), which is a form of AI ( artificial intellegence). This software is utilised for us to be able to train computers to understand and,learn and process human language
-
Chatbot integration
Dialogflow integrates with many popular conversation platforms like Google Assistant, Slack, and Facebook Messenger, Whatsapp and telegram. (many more)
-
Agents
Agents are efficient, can be trained faster than humans, and can learn and use prompts from real scenarios to improve itself
Preprogrammed virtual assitant , to ahndle conversations with customers and able to provide information
Entity
When intents are matched, Dialogflow's software is able to extract infromation from user comments, such as given preset keywords or phrases. This data is known as paramters for the bots to follow under Entities. These are for the bots and scan to detetc for, giving the bot more structed data to get a better sense of what the customer wants, not based solely off prompts only
Chatbots are a blessing in our connected world, where variables are evr changing, its good to have a operable bot 24/7 that takes customer prompts to help business interact with customers.