• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Machine-learning for Designing Online Store Recommendation System

Student: Alexandra Suslova

Supervisor: Irina P. Karpova

Faculty: HSE Tikhonov Moscow Institute of Electronics and Mathematics (MIEM HSE)

Educational Programme: Information Science and Computation Technology (Bachelor)

Year of Graduation: 2020

The problem of information overload has created issues in making choices out of an endless array of alternatives. As a result, the problem of finding relevant information is becoming increasingly serious, especially in fields such as online retail. Recommendation systems built with machine learning methods are a modern approach to solving this. As part of this work, data preprocessing and verification were conducted and automized. Various approaches to building a recommendation system were analyzed and tested: content filtering methods (TF-IDF – Term Frequency-Inverted Document Frequency, LightGBM – Light Gradient Boosting Method), collaborative filtering methods (SAR – Simple Algorithm for Recommendations, RBM – Restricted Boltzman Machine), matrix factorization methods (SVD - Single Value Decomposition, ALS - Alternating Least Squares). Based on an assessment of the algorithms’ quality using a chosen metric RMSE (Root Mean Square Error), TF-IDF and ALS algorithms were chosen for further work; next, a unique hybrid algorithm capable of making recommendations for all users and products was developed and implemented. The results of this are intended for use by the “Sunlight” company. In the future, the existing algorithm is going to be enhanced to create a recommendation system that can be connected directly to the retailer’s database. This system can then be used to generate recommendations on the company's site in real time.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses