Source: Postponing OPEC, non-OPEC meeting still option for RussiaBusiness & Economy December 09, 0:35
Sports arbitration court strips Russian boxer of 2016 Olympic silverSport December 08, 22:48
Russia, US military experts, diplomats to discuss Aleppo in Geneva on Dec. 10 - LavrovWorld December 08, 22:41
Lavrov says US voiced regret over shelling of hospital in Aleppo but somewhat hazilyRussian Politics & Diplomacy December 08, 21:48
Budget revenues from Rosneft privatization to be $11.1 bln — ministerBusiness & Economy December 08, 21:18
Lavrov, Kerry discuss militants’ withdrawal from Aleppo — Russian foreign ministryRussian Politics & Diplomacy December 08, 21:00
Lavrov: Combat actions in Aleppo suspended to take civilians out of cityRussian Politics & Diplomacy December 08, 20:56
Bach says WADA to play part in deciding on 2021 IBU World Championship in RussiaSport December 08, 20:44
Gazprom signs contract for construction of Turkish Stream’s first line with AllseasBusiness & Economy December 08, 20:03
MOSCOW, July 29. /TASS/ Russian scientists from Skoltech have put forward a filtering system that generates recommendations sensitive both to positive and negative user feedback, as reported by Skoltech’s press service.
The ACM RecSys 2016 conference, where the study’s results will be presented, will be held in Boston in September.
"With our algorithm, there are generally no restrictions to the amount of positive feedback that can be provided by a user. Instead, a person could specify his negative experience but the system is still able to give his further recommendations. The individual can start with some movie/book/audio track (liked or disliked) and the system will generate compatible suggestions. And of course, the more the user interacts with the system, the more accurate are the recommendations generated by it," says Evgeny Frolov, one of the developers.
Recommender platforms help personalize user experience in content consumption based on individual taste. Nowadays, every internet user has come across such systems, for example, when you visit an online page of your favorite bookstore you are likely to get a personalized reading list based on your previous purchase history and personal preferences.
Netflix, Amazon, Pandora, Spotify and many other companies generate a considerable amount of revenue with recommendations they produce.
Broadly speaking, recommender engines can be applied in many areas for almost any kind of goods or services.
Recommender systems can help an individual decide which movie to watch, which book to read next or which hotel to book for vacation.
If a user gives a low rating to the “Scarface” movie (which has a criminal narrative and contains violence), our system will find movies with “opposite” features
According to the developers, the most modern algorithms are almost insensitive to negative feedback and generate suggestions which are not compatible with a particular user, but the new model proposed by Russian scientists can deal with every type of user preference.
The proposed model may find its applications in different areas, such as online shops or media content providers.