Conversational AI Technology
Conversational Artificial Intelligence (AI) focuses on making computers understand natural language, learning from content and/or conversations, then answering questions and giving advice to solve problems.
exClone's invention makes machine learning fast (like a human reading a book and learning) instead of millions of iterations (processing billions of data) such as in deep learning.
The term exClone is derived from the fact the system duplicates (clones) human knowledge (expertise) by reading, then creates Virtual Experts.
Cloning Human Knowledge
The term “cloning” refers to exClone's unique approach to integrate several AI methods and different scientific disciplines to create a computer algorithm that represents a particular expert knowledge as curated by its creator(s).
The curation of the expertise may include documents, reference materials (such as books), articles, news feeds, conversation logs, and media sources like videos. The curator can be an expert person, or a group of experts. Personal choices and dialogue behavior can also be adjusted. The expertise encapsulated by digital cloning is delivered to the end user by exClone's conversational (Chatbot) interface. Learning can continue after the virtual expert is deployed by allowing designated users to continue teaching the virtual expert via conversations. You will find some of the basic components of digital cloning on this page.
exClone technology also allows conversational learning that supports organic growth of knowledge by the input from its users (often designated priori.)
exClone's technology offers a set of key advantages that make it highly affordable and practical.Burden Free: No coding is required, no AI experience is required. Everything is automated to the level of EDITORIAL effort only.
Data Free: No data (corpus) is needed. Accordingly, data processing costs are totally eliminated.
Fast: Single-step process to create a Virtual Expert. No iterative, repetitious, or continuous process required.
Transparent: Transparent system that allows corrections, modifications, and additions via intuitive interface (unlike some machine learning methods that are black box, and require re-training for corrections).
Embedded Cognitive Functions: Advanced conversational capabilities such as short-term memory, answering complex questions, navigating the user to answers by questioning, forming opinions, and learning from social interactions.
On-the-fly Learning: Ability to teach your virtual expert more information after deployment by live conversations with designated teachers.
The Three-Dimensional Nature of Virtual Experts
The creation of a virtual expert that can converse about particular expertise is a three dimensional problem as explained below (ONLP, ML, HDT).
Ontological Natural Language Processing (ONLP)
As the first dimension, exClone offers its own proprietary NLP libraries, ontology, and algorithms. NLP determines the level of understanding the users' requests. The algorithms automatically kick in during the creation process without any requirement from the creator.
Instant Learning (ML)
exClone's technology uses a proprietary machine learning method. The method absorbs knowledge from text by reading it sentence by sentence in a similar way that the human brain does when reading a book. Compared to other machine learning methods, like neural networks, exClone deploys a single-step process without the requirement of continuous training. The multi-layer architecture allows for deployment at various complexity levels depending on the nature of the exClone to be created.
The Human Dialogue Theory (HDT)
The central piece of the exClone's proprietary technology is derived from the Human Dialogue Theory founded by exClone's scientific team. The theory hypothesizes that among the seemingly infinite possibilities of what a person can say in a conversation, there is only a small set of dialogue behavior that holds the key to communication through chat algorithms. These are referred to as “dialogue patterns” tailored for different objectives such as sales, advising, assisting, surveying, and testing.
All virtual experts have this learning capability and can be turned on or off depending on the type of application it is being used for. On-the-fly learning allows for organic knowledge accumulation from conversations with the end user (designated teachers.)
Disciplines Used in exClone
Developed a new programming language TMLL (Text Meaning Logic Language)
Developed a new programming language CHAT MATRIX
Developed using C# and Java
No 3rd party software or API is used
Technology partially reserved as trade secret, and partially in patent filing status.
APIs are provided for the end use application. There are 3 different APIs available through the system as listed below.
API for exClone dialogue function
API for TMLL conversion of any text.
API for NLP functions.