Glossary of Terms
Conversational AI
Conversational AI is the application of technologies such as Natural Language Processing and Machine Learning which enables organisations to streamline the handling of real-time inbound user communications, queries, and tasks. Delivering enhanced self-service and improved experiences.
Chatbots
Chatbots are computer programs that are designed to replicate human conversations. Typically chatbots are keyword and rules-driven, based on pre-defined scripts and outcomes.
RPA
RPA is short for Robotic Process Automation, which is software that you can program to automate repetitive tasks. These tasks are rule-based and usually done manually, so RPA serves the goal of automating operations. Such technology has the power to reduce an organization’s costs, increase productivity, and bolster the supply chain. Discover the Basics of RPA | Automation Anywhere
Machine Learning
Machine Learning is the application of computer algorithms and statistical models which enable technologies such as Conversational Assistants to automatically learn and improve from experiences without the need to be programmed.
Natural Language Processing
Natural Language Processing (NLP) is the use of computer algorithms to read, interpret, understand, and make sense of human language to support interactions between humans and computers.
Intelligent Automation
Intelligent Automation is the application of multiple technologies such as AI, and RPA to deliver efficiencies to organisations through automating end-to-end business processes.
Knowledge Search
Wikipedia defines knowledge management as “…. the process of creating, sharing, using and managing the knowledge and information of an organization. It refers to a multidisciplinary approach to achieving organisational objectives by making the best use of knowledge.”
Humley’s conversational assistant makes a company’s knowledge infinitely more accessible to its customers, both internal and external. After ingestion of knowledge content, we use AI to enrich the data to help search accuracy and results, and then in real-time we use AI to help understand exactly what the user wants, ensuring that what is returned is of more specific relevance.
Intents
An intent is an outcome required from a conversation. They are a way for Conversational Assistants to understand your customer’s intention when they interact with it, and then follow through with an answer or action.
Entities
Entities bring deeper understanding by allowing you to capture and classify important parts of the conversation. Humley Studio lets you add powerful, pre-built system entities or define your own that look for text or pattern matches specific to your brand or product.
Variables
Variables allow you to capture specific answers, or use multiple instances of the same entity, that can then be used later in the conversation.
Flows
Flows bring the ability to create structured conversations, leading your customer down different routes depending on what they say by defining ‘happy paths’ and planning for exceptions using conditions.
Integrations
Integrations let you leverage third-party apps into your conversations. Our pre-built templates plug you straight into popular applications or you can create your own custom integrations into any third-party API using a ‘no code’ approach.
Channels
Channels allow you to take your conversational assistant to where your customers live. Deploy your service to popular channels, whether that’s on the web, chat and messaging apps, or even voice.
Triggers
Triggers control where and when responses are given or actions are fired, e.g. when a customer types an input or opens a channel, etc.
Analytics
Analytics lets you understand your customers and monitor the performance of your service. Dive deeper into what they are saying, the responses given, and user feedback left. Review your findings and target where maintenance is needed to make iterative improvements.
Sessions
Sessions give you end to end context of a conversation, along with more data to understand each ‘turn’, including entities recognised, variables populated and how confident the classifier was about what was said.