Hybrid collaborative filtering model using hierarchical clustering and PCA
Ashutosh Lokhande
, Ms. Pooja Jain
Recommender system; Collaborative filtering recommendation algorithm; Hierarchical Clustering; Principle Component Analysis; Big Data.
Recommendation system uses different types of algorithms to make any type of recommendations to user. Collaborative filtering recommendation algorithm is most popular algorithm, which uses the similar types of user with similar likings, but somewhere it is not that much efficient while working on big data. As the size of dataset becomes larger then some improvements in this algorithm must be made. Here in our proposed approach we are applying an additional hierarchical clustering technique with the collaborative filtering recommendation algorithm also the Principle Component Analysis (PCA) method is applied for reducing the dimensions of data to get more accuracy in the results. The hierarchical clustering will provide additional benefits of the clustering technique over the dataset and the PCA will help to redefine the dataset by decreasing the dimensionality of the dataset as required. By implementing the major features of these two techniques on the traditional collaborative filtering recommendation algorithm the major components used for recommendations can be improved. The proposed approach will surely enhance the accuracy of the results obtained from the traditional CFRA and will enhance the efficiency of the recommendation system in an extreme manner. The overall results will be carried out on the combined dataset of TMDB and Movielens, which is used for making recommendations of the movies to the user according to the ratings patterns created by the particular user.
"Hybrid collaborative filtering model using hierarchical clustering and PCA", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.4, Issue 8, page no.44 - 49, August-2019, Available :https://ijsdr.org/papers/IJSDR1908008.pdf
Volume 4
Issue 8,
August-2019
Pages : 44 - 49
Paper Reg. ID: IJSDR_190867
Published Paper Id: IJSDR1908008
Downloads: 000347051
Research Area: Engineering
Country: Indore, M.P., India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
