Program for Automatic Composition of Shopping List Based on User Preferences Prediction

The recommender systems are becoming more popular due to necessity to attract as more users as it possible. Users will likely be acquired by those services which propose smart system of recommendations. Such systems are mostly used by online-stores and online services and that is why they use the recommendations based on previous user purchases. The purpose of this study is to investigate the algorithms for recommender system based on users shopping list and create the new one which provides better and more accurate recommendations. The first part of this work represents the theoretical aspects of a recommender system and the target audience of it. There is the analysis of different approaches to the implementation of recommender systems. In order to have a better understanding of how users behave while working with the application that uses recommender system, the target audience research has been also conducted. The second part of the work reveals the implementation of the chosen approach and the new one that is supposed to be better than initial. At the end of this part, there is the evaluation of these algorithms and comparison by chosen criteria. On the basis of the results of this research, the program for recommender system is implemented and tested. The research may be useful for researchers in the field of recommender system, it contains a concise information about several methods and their applications.