I believe that understanding a little bit about the study aimed at recommendation systems can help in the construction of a model for this project.\nThe study of Systems of Recommendation (SR), has the interest to develop models for\nreduce the overload (2) of information that a user receives when performing a search through\nCustom Recommendations based on your profile. In them, knowledge modeling techniques are used to produce customized recommendations, showing its usefulness to users by guiding them in the search information and to provide product indications based on the your profile. A recommendation system should select and display to users only the products satisfying their\ntastes, thus reducing the cognitive overload at the same time where they should optimize\nthe quantity of products recommended to the user. For to carry out this task, these systems\nneed information about your users and about items that will be recommended because these factors are to calculate the degree of utility of each product for a given user. The degree of utility of an item indicates how much this item will satisfy, or not, the tastes of a particular user. Thus, a system of recommendation seeks to select and recommend to a certain user only products that have the highest degrees of utility for this user.\nIt is possible to classify the recommendation systems into 3 main types:\n* Systems of rebased on content: in this type of system users receive recommendations for items with similar characteristics to other items purchased in the past. \n* Systems of recommendation bain collaborative filtering: in this type of system users receive recommendations with based on the preferences of other users who have similar profiles. This is one of the best known techniques for premeditating and generating recommendations. Its algorithm consists of assembling a matrix of scores, where the rows represent the users and columns the items, so that user groups are identified with approximate scores. In this context, the closer the scores, the more similar are the user profiles.\n* Hybrid recommendation systems: this type of system uses content-based recommendation techniques with collaborative filtering based recommendation techniques. Another way to define it is: a Hybrid Recommendation system is one that combines the recommendation based on the users' past and the recommendation based on the user collaboration, that is, based on the item - item and the person - person correlation. The advantage of an approach that unites the others is to significantly increase the chances of achieving correctness in their recommendations and eliminate the limitations of both approaches.
", "messages": [ { "id": 466623, "postDate": "2019-02-05T17:41:18.607Z", "content": "I believe that understanding a little bit about the study aimed at recommendation systems can help in the construction of a model for this project.\nThe study of Systems of Recommendation (SR), has the interest to develop models for\nreduce the overload (2) of information that a user receives when performing a search through\nCustom Recommendations based on your profile. In them, knowledge modeling techniques are used to produce customized recommendations, showing its usefulness to users by guiding them in the search information and to provide product indications based on the your profile. A recommendation system should select and display to users only the products satisfying their\ntastes, thus reducing the cognitive overload at the same time where they should optimize\nthe quantity of products recommended to the user. For to carry out this task, these systems\nneed information about your users and about items that will be recommended because these factors are to calculate the degree of utility of each product for a given user. The degree of utility of an item indicates how much this item will satisfy, or not, the tastes of a particular user. Thus, a system of recommendation seeks to select and recommend to a certain user only products that have the highest degrees of utility for this user.\nIt is possible to classify the recommendation systems into 3 main types:\n* Systems of rebased on content: in this type of system users receive recommendations for items with similar characteristics to other items purchased in the past. \n* Systems of recommendation bain collaborative filtering: in this type of system users receive recommendations with based on the preferences of other users who have similar profiles. This is one of the best known techniques for premeditating and generating recommendations. Its algorithm consists of assembling a matrix of scores, where the rows represent the users and columns the items, so that user groups are identified with approximate scores. In this context, the closer the scores, the more similar are the user profiles.\n* Hybrid recommendation systems: this type of system uses content-based recommendation techniques with collaborative filtering based recommendation techniques. Another way to define it is: a Hybrid Recommendation system is one that combines the recommendation based on the users' past and the recommendation based on the user collaboration, that is, based on the item - item and the person - person correlation. The advantage of an approach that unites the others is to significantly increase the chances of achieving correctness in their recommendations and eliminate the limitations of both approaches.
", "rawMarkdown": "I believe that understanding a little bit about the study aimed at recommendation systems can help in the construction of a model for this project.\nThe study of Systems of Recommendation (SR), has the interest to develop models for\nreduce the overload (2) of information that a user receives when performing a search through\nCustom Recommendations based on your profile. In them, knowledge modeling techniques are used to produce customized recommendations, showing its usefulness to users by guiding them in the search information and to provide product indications based on the your profile. A recommendation system should select and display to users only the products satisfying their\ntastes, thus reducing the cognitive overload at the same time where they should optimize\nthe quantity of products recommended to the user. For to carry out this task, these systems\nneed information about your users and about items that will be recommended because these factors are to calculate the degree of utility of each product for a given user. The degree of utility of an item indicates how much this item will satisfy, or not, the tastes of a particular user. Thus, a system of recommendation seeks to select and recommend to a certain user only products that have the highest degrees of utility for this user.\nIt is possible to classify the recommendation systems into 3 main types:\n* Systems of rebased on content: in this type of system users receive recommendations for items with similar characteristics to other items purchased in the past. \n* Systems of recommendation bain collaborative filtering: in this type of system users receive recommendations with based on the preferences of other users who have similar profiles. This is one of the best known techniques for premeditating and generating recommendations. Its algorithm consists of assembling a matrix of scores, where the rows represent the users and columns the items, so that user groups are identified with approximate scores. In this context, the closer the scores, the more similar are the user profiles.\n* Hybrid recommendation systems: this type of system uses content-based recommendation techniques with collaborative filtering based recommendation techniques. Another way to define it is: a Hybrid Recommendation system is one that combines the recommendation based on the users' past and the recommendation based on the user collaboration, that is, based on the item - item and the person - person correlation. The advantage of an approach that unites the others is to significantly increase the chances of achieving correctness in their recommendations and eliminate the limitations of both approaches.\n" } ], "comments": [], "raw_markdown_by_id": { "466623": "I believe that understanding a little bit about the study aimed at recommendation systems can help in the construction of a model for this project.\nThe study of Systems of Recommendation (SR), has the interest to develop models for\nreduce the overload (2) of information that a user receives when performing a search through\nCustom Recommendations based on your profile. In them, knowledge modeling techniques are used to produce customized recommendations, showing its usefulness to users by guiding them in the search information and to provide product indications based on the your profile. A recommendation system should select and display to users only the products satisfying their\ntastes, thus reducing the cognitive overload at the same time where they should optimize\nthe quantity of products recommended to the user. For to carry out this task, these systems\nneed information about your users and about items that will be recommended because these factors are to calculate the degree of utility of each product for a given user. The degree of utility of an item indicates how much this item will satisfy, or not, the tastes of a particular user. Thus, a system of recommendation seeks to select and recommend to a certain user only products that have the highest degrees of utility for this user.\nIt is possible to classify the recommendation systems into 3 main types:\n* Systems of rebased on content: in this type of system users receive recommendations for items with similar characteristics to other items purchased in the past. \n* Systems of recommendation bain collaborative filtering: in this type of system users receive recommendations with based on the preferences of other users who have similar profiles. This is one of the best known techniques for premeditating and generating recommendations. Its algorithm consists of assembling a matrix of scores, where the rows represent the users and columns the items, so that user groups are identified with approximate scores. In this context, the closer the scores, the more similar are the user profiles.\n* Hybrid recommendation systems: this type of system uses content-based recommendation techniques with collaborative filtering based recommendation techniques. Another way to define it is: a Hybrid Recommendation system is one that combines the recommendation based on the users' past and the recommendation based on the user collaboration, that is, based on the item - item and the person - person correlation. The advantage of an approach that unites the others is to significantly increase the chances of achieving correctness in their recommendations and eliminate the limitations of both approaches.\n" } }