FIELD: computer technology.
SUBSTANCE: invention relates to a method and a server for determining the next state of the current state of the block of recommendations, which should be displayed on the display of an electronic device by a browser application. The technical result is to increase the efficiency of the selection of recommended content. In the method, the server receives an indication of a web resource and an indication of additional content to be displayed on the display using a recommendation block in together with a web resource; the server determines a set of features associated with at least part of the web resource; the server executes a machine learning algorithm (MLA) based on a set of features in order to determine a confidence level parameter that indicates the confidence level that additional content will be of interest to the user; selection is made by the server based on the confidence level parameter of the lookup table from the set of different lookup tables available to the server, each of which corresponds to one range of confidence levels from many different ranges of confidence levels available to the server; it contains a lot of different possible user actions and a lot of different possible current states of the recommendation block as possible inputs for generating output data, as well as many different possible next states of the recommendation block as possible outputs based on possible input data; and corelates each particular combination of the current state from a variety of different possible current states and user actions from a variety of different possible user actions with one next state from a variety of different next states; and the server sends to the browser application of the selected lookup table for processing by the browser application in order to determine an indication of the next state of the recommendation block from multiple different possible next states of the recommendation block and displaying the recommendation block containing additional content on the display in the next state of the recommendation block issued by the selected lookup table.
EFFECT: technical result is to increase the efficiency of the selection of recommended content.
19 cl, 9 dwg
Title | Year | Author | Number |
---|---|---|---|
METHOD AND SYSTEM FOR SELECTIVE PRESENTATION OF RECOMMENDATIONS IN A BROWSER APPLICATION | 2018 |
|
RU2739520C1 |
METHOD AND SYSTEM FOR GENERATING AN OBJECT CARD | 2018 |
|
RU2739554C1 |
METHOD AND SYSTEM FOR TRAINING MACHINE LEARNING ALGORITHM TO PREDICT VISIBILITY ASSESSMENT | 2022 |
|
RU2814079C1 |
SYSTEM AND METHOD FOR MANAGING BROWSER APPLICATION, PERSISTENT COMPUTER-READABLE MEDIUM AND ELECTRONIC DEVICE | 2015 |
|
RU2633180C2 |
RECOMMENDATIONS FOR THE USER ELEMENTS SELECTION METHOD AND SERVER | 2017 |
|
RU2693323C2 |
METHOD AND SERVER FOR PROCESSING TEXT SEQUENCE IN MACHINE PROCESSING TASK | 2020 |
|
RU2775820C2 |
METHOD AND SERVER FOR PRESENTING RECOMMENDED CONTENT ITEM TO USER | 2017 |
|
RU2699574C2 |
METHOD OF DISPLAYING WEB RESOURCE TO USER (VERSIONS) AND ELECTRONIC DEVICE | 2014 |
|
RU2595497C2 |
METHOD AND SYSTEM FOR PROVIDING A LEVEL OF SERVICE WHEN ADVERTISING CONTENT ELEMENT | 2019 |
|
RU2757406C1 |
METHOD AND SERVER FOR WEB-RESOURCE CLASSIFICATION | 2017 |
|
RU2658878C1 |
Authors
Dates
2021-04-21—Published
2018-12-13—Filed