Recommendation system for providing personalized and mixed content on a user interface based on content and user similarity
A recommendation system includes a content similarity analyzer configured to determine a first set of content item identifiers similar to a set of viewed content items based on respective similarity scores and add them to a first list. A similar user content extraction module identifies a set of similar user identifiers from a user similarity database; obtains, based on respective viewing histories of the set, a second set of content item identifiers; and adds them to the first list. The recommendation system includes a content filter configured to select a subset of content item identifiers from the first list based on the corresponding similarity scores between content item identifiers of the first list and a viewing history. The content filter is configured to transmit the subset of content item identifiers for display of the corresponding content items via a web portal on a user interface of a first user device.
1 . A recommendation system comprising:
at least one memory, wherein the memory stores instructions; and
at least one processor configured to execute the instructions and cause the recommendation system to perform,
upon a first user identifier logging in to a portfolio, identifying a set of user identifiers similar to the first user identifier based on one or more user parameters, the one or more user parameters including portfolio structure, trading activity, and platform usage,
selecting a subset of content item identifiers for the first user identifier from a first list, the first list including a plurality of content item identifiers and similarity scores between the plurality of content item identifiers and viewing histories associated with the set of user identifiers similar to the first user identifier, the first user identifier identifying a user without a viewing history and each similarity score defining a similarity between a content item identifier of the plurality of content item identifiers and an item of a viewing history of a user of the set of user identifiers based on salient terms in the content item identifier; and
displaying at least one content item identifier from the subset of content item identifiers via a web portal on a user interface of a first user device as a recommendation for further content for the user, the user interface including a user-selectable link for each of the at least one of the subset of content item identifiers.
2 . The recommendation system of claim 1 , wherein the recommendation system is further caused to perform generating a set of similarity scores for a first content item by comparing the first content item to a set of content items stored in a content database.
3 . The recommendation system of claim 2 , wherein the first content item is compared to the set of content items in response to the first content item being uploaded to the content database.
4 . The recommendation system of claim 2 , wherein the set of similarity scores includes a similarity score comparing the first content item and each content item of the set of content items.
5 . The recommendation system of claim 1 , wherein the first list is stored in a recommendation database.
6 . The recommendation system of claim 1 , further comprising:
a similar user database configured to store the one or more user parameters for a plurality of users.
7 . The recommendation system of claim 6 , wherein the recommendation system is further caused to perform categorizing the plurality of users into a set of groups based on the one or more user parameters stored in the similar user database and storing the set of groups in the similar user database.
8 . The recommendation system of claim 6 , wherein the one or more user parameters further include demographic information.
9 . The recommendation system of claim 6 , wherein the recommendation system is configured to identify the set of user identifiers similar to the first user identifier by selecting the set of user identifiers similar to the first user identifier from a first group of a set of groups in response to the first user identifier being categorized in the first group.
10 . The recommendation system of claim 9 , wherein the recommendation system is further caused to perform selecting a threshold number of identifiers as the set of user identifiers similar to the first user identifier.
11 . A recommendation method comprising:
upon a first user identifier logging in to a portfolio, identifying a set of user identifiers similar to the first user identifier based on one or more user parameters, the one or more user parameters including portfolio structure, trading activity, and platform usage;
selecting a subset of content item identifiers for the first user identifier from a first list, the first list including a plurality of content item identifiers and similarity scores between the plurality of content item identifiers and viewing histories associated with the set of user identifiers similar to the first user identifier, the first user identifier identifying a user without a viewing history and each similarity score defining a similarity between a content item identifier of the plurality of content item identifiers and an item of a viewing history of a user of the set of user identifiers based on salient terms in the content item identifier; and
displaying at least one content item identifier from the subset of content item identifiers via a web portal on a user interface of a first user device as a recommendation for further content for the user, the user interface including a user-selectable link for each of the at least one of the subset of content item identifiers.
12 . The recommendation method of claim 11 , further comprising:
generating a set of similarity scores for a first content item by comparing the first content item to a set of content items stored in a content database.
13 . The recommendation method of claim 12 , further comprising:
comparing the first content item to the set of content items in response to the first content item being uploaded to the content database.
14 . The recommendation method of claim 12 , wherein the set of similarity scores includes a similarity score comparing the first content item and each content item of the set of content items.
15 . The recommendation method of claim 11 , wherein the first list is stored in a recommendation database.
16 . The recommendation method of claim 11 , further comprising:
categorizing a plurality of users into a set of groups based on user parameters, wherein a similar user database stores the user parameters for the plurality of users; and
storing the set of groups in the similar user database.
17 . The recommendation method of claim 16 , wherein the user parameters of the similar user database further include demographic information.
18 . The recommendation method of claim 16 , wherein identifying the set of user identifiers similar to the first user identifier includes selecting the set of user identifiers similar to the first user identifier from a first group of the set of groups in response to the first user identifier being categorized in the first group.
19 . The recommendation method of claim 18 , further comprising:
selecting a threshold number of identifiers as the set of user identifiers similar to the first user identifier.