|Rachael said: Hi, my name is Rachael... Are You better than Eliza ?
Our general cognition engine is able to understand language without pattern matching, looking at predefined sequences, topic maps, ontologies or the like. Neither does it look at frequencies of occurrence of any word or phrase of interest. It plain understands what you mean, it is capable of associating ideas and abduction, discerning similarity or equivalence of concepts, has memory, it possess a 100,000 words vocabulary including 30,000 scientific words, and it is currently working on improving itself!
The Goodwill Consortium is currently collaborating on a joint submission of this site to the AISB X: Creativity Meets Economy (incorporating the Loebner Prize), a competition aimed at finding a computer system able to pass Turing´s test. The ultimate aim of this collaborative effort is to develop an artificial cognition system.
We are working on a form of Computerized cognitive-behavioral therapy (CBT) where Rachael Repp is used to hold a conversation with the user that is free-flowing. For each mental condition to be identified and supported, profiles are produced with typical comments likely to be associated with it; derived from the DSM-5 manual for instance. When a user mentions a phrase with a similar meaning to a stored sentence, a counter is increased appropriately for one or more conditions. The knowledge base contains general knowledge, culturally and age-relevant content to make young people more like to identify with the assistant, and a specialized knowledge base with profiles of conditions of interest. At the end, the assistant can make relevant comments for each condition if applicable; offering supporting materials such as worksheets, videos and infograms. Computerized therapy is a form of self-help treatment that is accessible and cost-effective, and suitable for people who are uneasy in the first instance about disclosure of sensitive information to a therapist. It has been shown to be effective at a comparable level to clinic-delivered CBT.
Rare diseases pose a serious challenge to society and affect the lives of many children. Approximately 75% of rare diseases affect children and 30% of rare disease patients die before the age of 5. There are approximately 7,000 rare diseases identified affecting 5% of the world’s population. For people with a rare disease, the mean average length of time from symptoms onset to accurate diagnosis is approximately 4.8 years. The difficult journey to diagnosis and care can increase medical, economic and social burdens. An accurate diagnosis is the first step to improving the quality of life of those living with a rare disease. Rachael has been successfully adapted to help in the process of diagnosis of rare diseases. The rationale for the system is grounded in the fact that rare disorders by their very nature are difficult to diagnose unless the caregiver or doctor is familiar with a wide range of these conditions, and that in more than 50% of cases it is those living with a rare disease, or their families, who ultimately manage to identify the correct diagnosis. Rachael uses natural language together with a database of symptoms associated to rare disorders (currently 600 of "the most common" rare diseases; though some diseases were excluded because although rare these conditions are not hard to diagnose). By matching users' description of the symptoms to her knowledge base, Dr. Rachael is able to identify the correct condition in at least 50% of the cases. If you wish to test the assistant, select "Rare diseases" in the knowledge bases tab above and register for a username and password; support and further details are available if you contact us.
A rapidly ageing population, in OECD countries particularly, brings a necessity and an opportunity for assistants such as Rachael. Currently funded work is aimed at prolonging the time that individuals can still live at home independently while at the same time increasing comfort and safety. In this site visitors can see the current state of this technology, a virtual assistant monitoring senior users 24/7 in the form of an app and further supported by the rachaelrepp.org site. The system provides a range of useful services such as activity monitoring, agenda, medication reminders, whatsapp connection, e-mail assistant module, emergency contact with the responsible carer, video generation and viewing, spotify, and so on; where Rachael is both culturally and age-aware in order to faciltiate the bonding process of the individual to the virtual assitant. The application and accompanying services are currently being tested during the project on 50 elderly users, their feedback is being taken into account for potential ongoing improvements.