Content-Based Recommendation Using Machine Learning

Tai, Yifan; Sun, Zhenyu; Yao, Zixuan

doi:10.1109/mlsp52302.2021.9596525

Cited by 17 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It selects a few things from a list of consumer appeals. The ranking system retrieves ranking objects based on item properties, and user-development needs to be revised to scale for extensive data [9], [10].…”

Section: Literature Reviewmentioning

confidence: 99%

Revolutionizing Netflix Recommendations: The Power of Transfer Learning

Anuradha,

2023

Artificial Intelligence and Machine Learning Applications to Computational Sanskrit, Natural Language Processing and Cognition

View full text Add to dashboard Cite

Major e-commerce companies grapple with managing vast data sets, posing a challenge to sense and fulfill user preferences. Despite existing Recommendation Systems (RS) for movies, songs, and products, misconceptions in recommendations persist. Navigating Netflix's extensive library poses selection challenges for users. This paper introduces an advanced recommender system utilizing text-based similarity and Word2Vec embedding. The model suggests movies based on diverse attributes like titles, genres, and descriptions. Employing techniques such as Transfer Learning, Similarity Matrix creation, and Content-Based Filtering with Word2Vec embedding, the system offers a top-10 list resembling the user's selected title. Validated through A/B testing, the user interface allows inputting preferred Netflix titles, delivering personalized recommendations based on content attributes, enhancing the streaming experience.

show abstract

Section: Literature Reviewmentioning

confidence: 99%