What Ariadn is about

Ariadn is recommendation engine (recommender system) with open API targeted for usage by any online business that may ever need a mechanism of personal recommendations or information filtering based on personal preferences of any customer.

Recommendation algorithms are best known for their use on e-commerce Websites, where they use data about previous customers’ interests to generate a list of recommended items. Many applications use only the items that customers purchase and explicitly rate to represent their interests, but other attributes may be also used, including items viewed, demographic data, subject interests, favorite artists, etc.

Ariadn adopts those mechanisms to literally any possible area of usage – social networks, online games, new web sites, media libraries, video services, etc.

Recommendation algorithms

Most recommendation algorithms start with finding a set of customers whose purchased and/or rated items overlap with the user’s purchased and/or rated items. The algorithm aggregates items from these similar customers, excludes items that the user has already purchased or rated, and recommends the remaining items to the user. Two popular versions of these algorithms are collaborative filtering and cluster models.

Collaborative filtering algorithm generates recommendations based on a few customers who are most similar to the user.

Cluster model used to find customers who are similar to the user; cluster models divide the customer base into many segments and treat the task as a classification problem. The algorithm’s goal is to assign the user to the segment containing the most similar customers. It then uses the purchases and ratings of the customers in the segment to generate recommendations.

Possible areas of use

General use:

• Building of personal recommendations;
• Sales and revenue increase;
• Increased efficiency of stock usage;
• Building flexible and effective user interface;
• Increase efficiency of monetization through personal recommendations;
• Better understanding of audience;
• See the trends in demand and supply.

Social networks:

• Personalized list of suggested applications, specific for each user;
• Filtering spam messages and messages from friends;
• Recommendations for media content, discussion groups and topics within built on users’ interests.

E-Commerce and online shops (including games):

• Personal recommendations;
• Concentrating user’s attention on personal list of suggested goods;
• Sales increase;
• Increased efficiency of stock usage;
• Understanding customers.

Digital media catalogues:

• Personal recommendations based on user votes;
• Recommendations based on  suggestions of user’s friends.

Success stories

There are many samples of successful application of recommendation systems. Some businesses are built solely around recommendations. Here are several success stories:

• “Recommended for you” at Amazon.com;
• last.fm and pandora.com;
• Genus module at Apple iTunes store.

Licensing Ariadn engine

Ariadn is using SaaS (Software as Service) business model.

That means we license Ariadn to customers as a service on demand, through a subscription or a “pay-as-you-go” model. We develop, host, and operate software for customer use. Our customers access Ariadn over the Internet instead of installing any software on site. We may run entire application or a part of it on our hardware (“Cloud” business model is used), or download executable code to client machine, if required.

Ariadn can be licensed either for a single project or for a group of projects.