We consider the reputation of users of social networks
Why do you need a reputation?
Reputation can reflect many connections in a system, connections between systems, their size and relevance. Reputation is the status of a given element in the system, be it a specific person, book, movie, website or something else. It allows the end user to navigate among a large amount of information, choose the best for himself, and also, if he himself is an element of the system, such as in social networks, to “show himself” himself.
The simplest and most widely used type of reputation system is rating without links. For example, users evaluate a product in an online store or a movie on a site whose reputation in the end is simply the number of sales of that product or the number of people who viewed and rated the movie.
On social networks, users can rate photos of other users and their reputation (rating) is simply the number of people who positively rated the photo.
More advanced reputational technologies use connections between objects in the system. Moreover, the reputation of an element of the system depends on the reputations of other related elements - donors, as well as on all other elements that affect the reputation of its donors. That is, the increase in the reputation of the object will be greater if a transaction from a higher reputational object arrives at it. The pitfall in such systems is the problem of cyclic cheating, when the elements of the system are not connected with each other in a unidirectional connection, and their reputations depend on each other's reputations. If you do not take into account and do not remove these cheatings, then an uncontrolled growth of reputation is possible.
Briefly about the famous rating system
The most famous link ranking system is PageRank, developed by Google. Since its first release, many different modifications and improvements have appeared, mainly related to the optimization of the main algorithm. In this case, the problem of cyclic wrapping is solved in it by reducing the value of the transaction weight so that transitive and cyclic communications quickly decay and do not lead to an avalanche-like growth of reputations.
Imhonet’s team has developed an alternate linking reputation system called NDX. At present, it is used to calculate the reputation of users in the Imkhonet recommendation system. At the same time, the API of the system engine is quite flexible and does not limit its application to social networks. This technology allows you to enter an unlimited number of connection properties so that the reputation of the object in the system can accumulate transactions for different types of connections.
The problem of cyclic wrapping is solved not by reducing the weight of transactions, but by applying complex mathematical algorithms that remove from the donor's reputation that part to which the associated object contributes. This allows you to take into account the contributions of all possible donors to the reputation of the object, not only close in relationships, but also far transitive and having a fairly large reputation. The implementation of the NDX engine takes into account multithreading and, in principle, is quite easily clustered.
Wikipedia information on PageRank
More about Imkhonet’s reputation engine
Text writer: Maxim Shikhalev