The Russian capital Moscow had the highest quality of life in the country in 2022, according to the ranking, in which the city received ** points. The second best score was achieved at less than one point lower by Saint Petersburg.

Moscow-longevity-project Relevance of the task Since 2018, the “Moscow Longevity” project has been operating in Moscow, which provides older Muscovites (55+ for women and 60+ for men) the opportunity to engage in a wide range of educational, leisure and health activities. Classes take place in groups offline and online under the guidance of professional accredited teachers and in five years have become an important part of the city infrastructure that improves the quality of life of the older generation, level of socialization and versatile activity. To date, more than half a million people have joined to the project, and the number of areas of study exceeded several hundred. Every month tens of thousands of new and existing members are looking for suitable groups for you. Coming soon on the “Moscow Longevity” page (https://www.mos.ru/city/projects/dolgoletie/) the service will be launched automated entry into groups, and into one of the user paths it is planned to introduce recommendation mechanics. This is a recommendation the solution can significantly reduce the time for selecting relevant group for the participant, and will also reduce employee labor costs Moscow Longevity for counseling older citizens in as part of the search for suitable activities.

Description of the task As part of the task, it is necessary to create a recommendation service, which will help participants (new and existing) of the Moscow longevity” select a suitable group for classes based on accumulated data on user activity in the project. This is the solution may become part of the registration service for “Moscow Longevity”, which will be launched on MOS.ru soon. This recording service has three basic user options: script: a) search for a group through the search bar; b) search through a group catalog with a filtering system; c) search through completing a survey in which participants answer questions based on your interests. In order to more relevantly select groups within the latter scenario, we propose to develop a full-fledged recommendation service/module that takes into account the behavior and patterns of consumption already current participants. Users of the recommendation service are divided into the following: groups: a) current (already registered in the project) participants “Moscow longevity” b) new (potential) participants We do not identify the user (whether he is a member of the Moscow longevity”) automatically, even if he is already authorized on MOS.ru. So Thus, at the first stage of user interaction it is necessary request data that will allow us to determine whether he is participant of “Moscow Longevity” or not: full name, date of birth. In case the user is not a member, we must determine the interests and needs of the visitor and give him recommendations by groups (through preset filters), and preferably not in less than 3 different key areas (“for the mind”, “for the body”, “for soul”) and one additional (additional education with a diploma, obtaining a new profession). If we have identified a valid participant, we should refer to the history of his visits and, taking into account it (and/or other data), offer additional classes. Classes (groups) at “Moscow Longevity” according to format are divided into two parts: a) face-to-face classes b) online classes Thus, in addition to directing classes for face-to-face groups, we should prioritize offering users classes taking place as close as possible to his place of residence. Priority format in “Moscow longevity” are face-to-face classes, hackathon participants with When designing a recommendation service, it is proposed to take this into account factor and offer mechanics aimed at engaging users offline, even if the profile of their interests indicates that they tend to choose online classes. Initial data: List of group filtering parameters in the recording service: ● areas of study (1,2,3 levels) ● class format (face-to-face / online) ● area of classes (for full-time classes) schedule (days of week / time) List of data contained in the “Groups” dataset: ● group code ● direction 1 ● direction 2 ● direction 3 ● address of the sites (for online classes it has a technical meaning) ● district area ● site area schedule List of data contained in the “Users” dataset: ● unique number ● date of creation of the personal file ● floor ● date of birth residential address (details to apartment building) List of data contained in the “Attendance” dataset: ● unique lesson number ● unique group number ● unique participant number ● online/offline ● date of class ● class start time ● end time of class