000			06901nam a2200301Ia 4500
001			73452
003			ft6089
005			20251124121211.0
008			180829n 000 0 eng d
040			_erda
041			_aengtag
050			_aQA76.9 C37 2016
082			_a.
100			_aRonald Carlos Jr. and Melsheen Kissob.
245		0	_aFurther enhancement of collaborative filtering algorithm to be applied in video game recommendation system
264			_a. _b. _c2016
300			_bUndergraduate Thesis: (BSCS major in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2016.
336			_b. _atext _2rdacontent
337			_30 _b. _aunmediated _2rdamedia
338			_30 _b. _avolume _2rdacarrier
505			_aABSTRACT: The goal of a Recommender System is to generate meaningful recommendations to a collection of users for items or products that might interest them. Suggestions for books on Amazon, or movies on Netflix, are real world examples of the operation of industry-strength recommender systems. The design of such recommendation engines depends on the domain and the particular characteristics of the data available. For example movie watchers on Netflix frequently provide ratings on a scale of 1 (disliked) to 5 (liked). Such a data source records the quality of interactions between users and items. Additionally, the system may have access to user-specific and item-specific profile attributes such as demographics and product descriptions respectively. Recommender systems differ in the way they analyze these data sources to develop notions of affinity between users and items which can be used to identify well-matched pairs. This study presents the enhancement of the Collaborative Filtering Algorithm to be applied in Video Game Recommendation System. It will focus on the Item-based approach of the Collaborative Filtering which is a method of making automatic predictions (filtering) about the interests of a user by collecting preferences or taste information from many users (collaborating). Upon simulating the Collaborative Filtering Algorithm, the researchers found some problems and room for enhancements for the said algorithm. First, there is no threshold for the number of common users. Second, some similarity values are over emphasized. Third, it uses the active user’s mean rating for undefined predictions, where in the algorithm fails to provide recommendation for new users it has no rating history. This study aims to provide solutions for the problems stated and enhance the algorithm for the production of better recommendations. The researchers have researched and studied different articles and literatures that proves the use Collaborative Filtering Algorithm as the best algorithm for recommendation systems. Throughout the study, the researchers found ways to solve the problems in the algorithm. First, the researchers will set a threshold value for the minimum number of common users in the matrix. With this, filtering of similar and non-similar items can be done earlier. Second, items tagged as non-similar will have their similarity calculations decreased. This is to have their values with the similar items treated equally. Lastly, to use the target’s items mean rating instead of the active user’s mean rating. With these objectives, the researchers were able to enhance the algorithm and improve its recommendation accuracy. At the end of the study, the researchers recommend to apply the algorithm to other applications such as behavioural prediction and subject analysis. Also, other categories can be used for filtering aside from genre, eg. Age, Platform, etc. Some processes can also be simplified in order to speed up the prediction process of the algorithm.
506			_a5
520			_aABSTRACT: The goal of a Recommender System is to generate meaningful recommendations to a collection of users for items or products that might interest them. Suggestions for books on Amazon, or movies on Netflix, are real world examples of the operation of industry-strength recommender systems. The design of such recommendation engines depends on the domain and the particular characteristics of the data available. For example movie watchers on Netflix frequently provide ratings on a scale of 1 (disliked) to 5 (liked). Such a data source records the quality of interactions between users and items. Additionally, the system may have access to user-specific and item-specific profile attributes such as demographics and product descriptions respectively. Recommender systems differ in the way they analyze these data sources to develop notions of affinity between users and items which can be used to identify well-matched pairs. This study presents the enhancement of the Collaborative Filtering Algorithm to be applied in Video Game Recommendation System. It will focus on the Item-based approach of the Collaborative Filtering which is a method of making automatic predictions (filtering) about the interests of a user by collecting preferences or taste information from many users (collaborating). Upon simulating the Collaborative Filtering Algorithm, the researchers found some problems and room for enhancements for the said algorithm. First, there is no threshold for the number of common users. Second, some similarity values are over emphasized. Third, it uses the active user's mean rating for undefined predictions, where in the algorithm fails to provide recommendation for new users it has no rating history. This study aims to provide solutions for the problems stated and enhance the algorithm for the production of better recommendations. The researchers have researched and studied different articles and literatures that proves the use Collaborative Filtering Algorithm as the best algorithm for recommendation systems. Throughout the study, the researchers found ways to solve the problems in the algorithm. First, the researchers will set a threshold value for the minimum number of common users in the matrix. With this, filtering of similar and non-similar items can be done earlier. Second, items tagged as non-similar will have their similarity calculations decreased. This is to have their values with the similar items treated equally. Lastly, to use the target's items mean rating instead of the active user's mean rating. With these objectives, the researchers were able to enhance the algorithm and improve its recommendation accuracy. At the end of the study, the researchers recommend to apply the algorithm to other applications such as behavioural prediction and subject analysis. Also, other categories can be used for filtering aside from genre, eg. Age, Platform, etc. Some processes can also be simplified in order to speed up the prediction process of the algorithm.
526			_aF
540			_a5
655			_aacadenic writing
942			_alcc _cARCHIVES _2lcc
999			_c25352 _d25352