Normal view MARC view ISBD view

An Enhancement of content-based filtering applied in movie recommendation system. 6

By: Samantha Gwyn M. Aranzamendez, Joshua Caleb D. Bolito, Aron Christopher R. Rafe. 4 0 16, [, ] | [, ] |

Contributor(s): 5 6 [] |

Language: Unknown language code Summary language: Unknown language code Original language: Unknown language code Series: ; 4541346Edition: Description: Content type: text Media type: unmediated Carrier type: volumeISBN: ISSN: 2Other title: 6 []Uniform titles: | | Related works: 1 40 6 []Subject(s): -- 2 -- 0 -- -- | -- 2 -- 0 -- 6 -- | 2 0 -- | -- -- 20 -- | | -- -- -- -- 20 -- | -- -- -- 20 -- --Genre/Form: -- 2 -- Additional physical formats: DDC classification: | LOC classification: | | 2Other classification:

Contents:

Action note: In: Summary: ABSTRACT: Content-based Filtering is a recommender system that provides recommendations based on the description of an item a user interacted with. It is used in various consumer domains to personalize recommedantions to users. Despite its relevant functionality, Content-based Filtering is riddled with limitations as well. Among these limitations of the recommendations provided by Content-based Filtering are overspecialization - where the algorithm recommends those items that are directly related to the user which rules out those newer sets of items, cold-start - where the algorithm cannot produce appropriate recommendations for new users of the system that have no and/or limited rated features in their profile, and synonymy issues - where the algorithm lacks the ability to distinguish semantic relationships between words. In this study, an enhanced Content-based Filtering was developed which addresses the issue of overspecialization, cold-start, and synonymy issues by using Maximal Marginal Relevance, Temporary Preference Profile, and FastText. Results show that the proposed enhancement with n=0.5 showed the most prominence among the enhanced variants of Content-based Filtering, having a good balance between relevance and diversity of recommendation which on average, improves upon the original algorithm in terms of Precision by 45.83%, in terms of Recall by 24.45%, in terms of F-Score by 38.51%, and in terms of Diversity by 276.79%. Other editions:

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings ( 1 )
Title notes
Comments ( 0 )

Item type	Current location	Home library	Collection	Call number	Status	Date due	Barcode	Item holds
Book	PLM	PLM Filipiniana Section	Filipiniana-Thesis	QA76.9.A43 .A73 2024 (Browse shelf)	Available		FT7859

Total holds: 0

Undergraduate Thesis : (Bachelor of Science in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2024. 56

ABSTRACT: Content-based Filtering is a recommender system that provides recommendations based on the description of an item a user interacted with. It is used in various consumer domains to personalize recommedantions to users. Despite its relevant functionality, Content-based Filtering is riddled with limitations as well. Among these limitations of the recommendations provided by Content-based Filtering are overspecialization - where the algorithm recommends those items that are directly related to the user which rules out those newer sets of items, cold-start - where the algorithm cannot produce appropriate recommendations for new users of the system that have no and/or limited rated features in their profile, and synonymy issues - where the algorithm lacks the ability to distinguish semantic relationships between words. In this study, an enhanced Content-based Filtering was developed which addresses the issue of overspecialization, cold-start, and synonymy issues by using Maximal Marginal Relevance, Temporary Preference Profile, and FastText. Results show that the proposed enhancement with n=0.5 showed the most prominence among the enhanced variants of Content-based Filtering, having a good balance between relevance and diversity of recommendation which on average, improves upon the original algorithm in terms of Precision by 45.83%, in terms of Recall by 24.45%, in terms of F-Score by 38.51%, and in terms of Diversity by 276.79%.

There are no comments for this item.

to post a comment.