Neural networks and their use in prognostication systems
Neural networks is a kind of technology that reproduces the function of the human nervous system. The most important features of the nervous system for the artificial intellect proved to be the abilities to learn, make decisions, arrive at conclusions and correct mistakes.
Biological neural networks
The human brain is a complex intellectual computer built up on nerve cells known as neurons.
Neurons' main task is to receive and pass on information with the help of special outgrowths - dendrites and axons.
The more dendrites a neuron has, the more information streams it can receive and process.
Axons send signals from neurons to other cells: neurons, cells of muscles and other internal organs.
The structure of the nerve cell – neuron.
Problems that Bhrigu solves
Separately taken, neurons can perform only basic functions. They work far more efficaciously only when they are combined in neuron nets. Contact points between two neurons are called synapses which means "connections" in Greek.
It is synapses that effect the transmission of a signal within the neuron net preserving its informative importance.
Neuron nets tie together separate parts of our body into a functioning unit and serve as a literal storage of our psyche.
The human brain contains approximately 100 billion neurons, due to which biological neuron nets are incredibly complicated.
Our body contain trillions of neuron connections, many of which are incessantly active.
Neuron connections (a microscope shot) – courtesy of Science Source.
Artificial neuron nets are computer models and algorithms designed for reception, processing, research and transmission of information the way their analogous biological neuron nets in the brain do. Neuron nets form the key component of artificial intellect technologies, machine education and in-depth education.
The unique feature of in-depth education lies in the fact that the neuron net is able to find patterns (key characteristics and consistent trends) in incoming data and structurize them hierarchally from the simplest to the most complex. The complexity of a neuron net is defined by the number of stages (levels, strata) of information processing.
The example of a simple neuron net consisting of three levels.
On the first level the neuron net makes three very simple decisions, assessing incoming data. On the second level decisions are more complex, based on the calculated ones from the first level. Even more complex decisions can be made on the third level.
Generally speaking, the principle of the neuron net's operations amounts to "weighing up proof." The neuron net in the picture receives three incoming factors x1, x2, x3 each of which is assigned weigh w1, w2, w3 defining the importance of each incoming factor for output. One resultant decision comes out.
Compared to common computer applications (for example, those used by astrologers, numerologists and so on), neuron nets possess greater possibilities for data processing, allow for a larger data volume and perform more complex tasks.
Neuron nets established themselves well in solving the following tasks:
discovery of regularities and patterns (for example, event patterns or profession patterns);
processing requests with a large number of variables and variety of incoming data (for example, find me a city where I could set up business, the child would get better, and my personal life would pick up…);
non-unique correlation between variables (like a situation when different prognostication techniques show ambiguous results);
difficulty of an adequate description of a situation using common approaches.
Hybrid Intelligence System
Hybrid Intelligence System (HIS) is a system which solves tasks using more than one technique imitating man's intellectual function.
Therefore, HIS is a composite of:
analytical patterns
expert systems
artificial neuron nets
fuzzy systems
genetic algorithms
imitative statistic models
The hybrid intelligence system cross disciplinary development unites scientists and experts studying applicability of not one but several techniques belonging to different categories for solving management and design tasks.
The astrological neuron net uses the incoming data of the date, time and place of an individual's birth (residence), a company's registration, and so on.
The net's output will be factors influencing events.
Here are some simple examples of external and personal influences:
hours, days, weeks, months, years
geographical position
transits of the Sun, the Moon and the planets from one sign to another
solar and lunar eclipses
various combinations of planetary influence
The platform chooses out of various formulae those most relevant and assured for every particular event and puts out highly accurate recommendations.
The main advantage of the Bhrigu neuron net lies in its simultaneous employment of techniques and algorithms of various schools and approaches.
The Bhrigu artificial intelligence maintains unceasing self-education on the data of its users – it improves algorithms, finds new consistent patterns. The more data the system receives, the more accurate are its calculations.
The development of information technology defined the creation of independent systems of transmitting and storing information- Blockchain.
In its turn it made possible the advent of decentralized payment systems – crypto currencies.
The introduction of this technology and issue of special Bhrigu tokens allows the system to solve five key problems at once:
Raising funds necessary for creating the platform (the funding is effected by offering significant discounts for future usage of Bhrigu's services).
Decentralized system of payment for the services and experts' fees.
Introduction of smart-contracts between experts and users.
Organization of secure decentralized and independent data storage.
Offering investors and traders opportunities for making profit, the Bhrigu token serving as an investment instrument allowing entrance to exchange.
The Blockchain modern technologies for storing and transmitting information allow the creation of Decentralized Autonomous Organizations (DAO).
DAO is the modern way of organizing intellectual activity in cyber space. What makes the difference of DAO from traditional companies is the absence of one owner, server, headquarters and hierarchic management. Every subject of the system can become the company's co-owner and participate in its development. Decisions in DAO are taken by the owners' vote, all activities are computerized, the stability level of the organization is extremely high.
The DAO format allows people to invest or offer their services remotely, in comfortable conditions, enjoying interrelations that are absolutely transparent. In DAO any expert can do whatever they like and whatever they are best at and be sure of 100% self-development.
We used this model when creating Bhrigu with the purpose of offering users, talented experts and progressive investors a new format of convenient and mutually profitable cooperation.
Even at the first stage we intend to work out more than 12 unique prognostication services helpful for every individual (vocation, health, relationships, success, etc.) as well as for specific occupations (prediction of currencies and crypto currencies rates, result of sport events, etc.).
But the creation of a high quality service requires considerable funding. Meanwhile, the demand for every particular product is not known. We are already using the DAO to get this information.
We are also building up a decentralized and independent platform for voting where every interested participant of the project can express his or her will by voting for a service and confirming their interest by purchasing the service in advance.
The possibilities of Bhrigu API for owners of sites and portals:
setting up a prognostication service on any Internet resourse.
Designing applications for Android and iOS on the Bhrigu platform engine.
Possibility of creating unique services based on artificial intelligence on turnkey basis.
The description of the Bhrigu system operation
The Bhrigu platform consists of three main parts:
Database, Event prognostication block and Event influence block
Event prognostication block
Event prognostication block is a hybrid intelligence system (HIS) functioning on the database and users' personal data as well as on various prognostication techniques. Event prognostication block forms predictions for users, experts, businesses.
Event prognostication block embraces 17 services of different spheres.
Event influence block is a community of experts and researchers aiming at rendering help to the users. All the interrelations within the community are based on the principles of decentralized autonomous organization (DAO), smart-contracts and the Bhrigu tokens crypto currency.