A Study on Recommendation System with Library Data
Satyal, Umesh Raj (2019)
Åbo Akademi
2019
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
Julkaisun pysyvä osoite on
http://urn.fi/URN:NBN:fi-fe2019081524237
http://urn.fi/URN:NBN:fi-fe2019081524237
Tiivistelmä
The recommendation is a difficult task in the lack or absence of domain knowledge like facts and information of the item which is the subject of the recommendation. This study presents and compares algorithms used in building a recommendation system that uses only the information from the artifact loan log of a library. The library has different materials available to lend like books, CDs, DVDs which as a whole will be represented as artifacts in this thesis. This system is not limited to artifacts from the library but could have a vast range of usability like in supermarkets.
The thesis introduces the reader to the recommendation system followed by an overview of recommendation problem formulation, interpretation of recommendation problem as data mining problem, different techniques and algorithms used in recommendation system and validation techniques for those algorithms. A case study is performed by using the library dataset obtained from Vantaa City Library. The objective of the case study is to test different recommendation algorithms with the dataset. The algorithm used is based on Collaborative Filtering. The models are evaluated using 5-fold cross-validation and the evaluation metrics used are Root Mean Square Error(RMSE) and Mean Square Error(MSE).
The thesis introduces the reader to the recommendation system followed by an overview of recommendation problem formulation, interpretation of recommendation problem as data mining problem, different techniques and algorithms used in recommendation system and validation techniques for those algorithms. A case study is performed by using the library dataset obtained from Vantaa City Library. The objective of the case study is to test different recommendation algorithms with the dataset. The algorithm used is based on Collaborative Filtering. The models are evaluated using 5-fold cross-validation and the evaluation metrics used are Root Mean Square Error(RMSE) and Mean Square Error(MSE).