Vinni Adaptive Learning Platform is powered by PEAS™ matching and recommendation engine.
Our API service provides access linking Internet, social media and any internal sources to match and sort relevant data in broad and deep searches. We build smart technology based on ethical principles self-adjusting to fit each individual user’s preferences.
The Vinni Adaptive Learning Platform™ analyses large data sets from various sources in order to present the best content, tailored to each individual user. It takes into account data material requirements and personal preferences in order to match and recommend the most relevant information.
Its autonomous nature means the more it is used, the better it performs.
The Vinni Platform is designed as the antidote to one-off rigid and inflexible software systems. Our API’s interoperability is built to share resources among many applications. It provides BaaS solutions – with PEAS™ a reconfigurable, problem solving AI engine at its heart. Our machine learning is the driving force in providing a ‘matching solution’ with a personalised experience for each user.
We can provide the back end technology platform, our engine, accessible through our API layer with technical support – to give your existing systems or Apps, smart machine learning capabilities
The engine accepts user feedback and makes future use even more relevant. The personal experience is enriched by leveraging the behavior of other users with similar interests. This adds a collaborative aspect to assist in user reviews.
User interaction and ‘crowd wisdom’ is knowledge improvement by natural means.
Degrees of exploration are engaged to help predict the user’s potential future interests. This insures that the personalised searches do not become static or overspecialised.
The future of personalisation is
The Vinni engine has been built to be autonomic in nature. It is self-managing and can adapt to unpredictable changes. The engine makes decisions on its own – it will constantly check and optimise its status and automatically adapt itself to changing conditions. Rather than following strict program instructions our machine learning algorithms learn from data, make predictions and decisions. To aid speed it works with parallel processing to provide timely information. Its knowledge gathering sources are from the Internet, social media and any dedicated data systems – to fit a specific need. It generates its own process through self adjustment and is continuously probing for results, which when found, are pushed as real-time user alerts. The engine can be re-tasked to fit many differing uses.
Despite its power, it has a light footprint.
Data profiles for each person are dynamically built by monitoring sentiment, following user actions and engaging in user dialogue. A balanced profile is obtained through integration of available knowledge from a community of like-minded users.
‘It tracks my world to assist in my future use’
Artificial Neural Networks are used to make decisions in complex situations, for which the data or environment are difficult to formulate discrete rule bases, such as the self-support system, and its chaotic signal patterns. Evolutionary components, such as Genetic Algorithms, are also used to refine system response models over time, fine-tuned to the individual customer’s system behaviour.
We see matching algorithms as being key to solving many problems. With our background in Logistics, we have been matching freight to containers, driver to rigs, empty trucks to loads, containers to rail wagons and so on… We have now used our extensive knowledge and experience and applied it to a meaningful internet based system.
We put the ‘concept of matching’ in the hands of the user.
This module is designed to receive live ‘whereabouts’ information, and to connect this through to the user’s mobile device in real-time. Whether it is a planned purchase, an impulse buy, mood shopping or locating work – it is all done in real-time and available on the Smartphone.
It suits any user looking for anything, anywhere.
The ‘Drive-Buy’ tool takes as input data GPS traces of the user and their agenda and then suggests suitable information depending on the situation and interests. The system uses machine learning techniques and reasoning process in order to adapt dynamically to the mobile recommender ‘Drive-Buy’ tool.
Personalised drive buy alerts.
The Vinni Application Platform is web and mobile based. This framework is designed from the bottom up, to enable common API interfaces that can be deployed across a variety of cloud and physical application platforms. It distributes the workload according to task type and priority, delivering results to the user in real-time. Common API’s provide entry points into the system for native applications that complement the web-based experience.
Vinni Apps plug into this platform like lego blocks – sharing common resources.
Personal Anonymity “Vinni is personal but it’s strictly business”
Vinni provides personal anonymity – we achieve this by focusing on each user’s wishes and not their personal details. Vinni performs tasks for each user and establishes a knowledge profile of how they wish to proceed but not WHO they are.
We do not want to have personal details.
We do want to win EEC prizes for personal data integrity.
It’s not about ‘Who you are’ – it’s about ‘What you want’