Clustering and recommending items based upon keyword analysis
A multi-level approach for generating suggestions and/or recommendations associated with one or more items is provided. A group of related items may be identified, for example, by generating a collaborative filtering graph or other graph that includes items and connections between the items. Description data associated with items included in the group may be evaluated in order to determine respective keywords associated with the items and/or a list of common keywords representative of the group of items or a subset of the group of items. Based at least in part upon the list of common keywords, at least one suggestion may be generated.
1. One or more non-transitory computer-readable media storing computer-executable instructions that, in response to execution, configure at least one processor to:
identify related items using a graph representative of items, the graph comprising nodes and edges, wherein a first node of the nodes represents a first item of the items, and wherein an edge of the edges represents a connection between a second node of the nodes and a third node of the nodes, the connection is indicative of co-occurrence of an event associated with at least one of a second item associated with the second node or a third item associated with the third node;
determine respective keywords for items of the related items;
determine common keywords within the respective keywords for the items of the related items;
determine a total number of common keywords;
determine a defined number of topics using the total number of common keywords; and
generate a suggestion using a portion of the common keywords, wherein the suggestion is associated with the defined number of topics.
2. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to perform a latent semantic analysis on product description information associated with an item of the related items.
3. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to generate the graph utilizing a collaborative filtering technique.
4. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to perform, using the defined number of topics, at least one of a hierarchical clustering technique, a centroid-based clustering technique, a distribution-based clustering technique, a density-based clustering technique, or a deterministic annealing clustering technique.
5. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to identify a topic associated with the related items using the common keywords, and wherein the suggestion comprises the topic.
6. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to identify a topic associated with the related items using the common keywords; and
to generate a second suggestion using the topic.
7. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to identify a topic associated with an item additional to the items using product description information of the item; and
to generate a second suggestion using the topic.
8. The one or more non-transitory computer-readable media of claim 1 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to determine events associated with the related items, and wherein a first event of the events comprises purchasing a first item of the related items, and further wherein a second event of the events comprises viewing a second item of the related items, and further wherein a third event of the events comprises reviewing a third item of the related items.
9. The one or more non-transitory computer-readable media of claim 8 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to generate the graph using at least a co-occurred event of the events.
10. The one or more non-transitory computer-readable media of claim 9 , wherein the computer-executable instructions, in response to execution, further configure the at least one processor to identify a cluster of second items in the graph, and wherein the cluster of second items includes the related items.
11. The one or more non-transitory computer-readable media of claim 10 , wherein the at least one processor is further configured to execute the computer-executable instructions to identify the cluster using at least one of a Floyd-Warshall algorithm, a Bellman-Ford algorithm, or a Dijkstra algorithm.
12. A method, comprising:
identifying, by a computing system comprising at least one processor, related items using a graph representative of items, the graph comprising nodes and edges, wherein a first node of the nodes represents a first item of the items, and wherein an edge of the edges represents a connection between a second node of the nodes and a third node of the nodes, the connection is indicative of co-occurrence of an event associated with at least one of a second item associated with the second node or a third item associated with the third node;
determining, by the computing system, respective keywords for items of the related items;
determining, by the computing system, common keywords within the respective keywords for the items of the related items;
determining, by the computing system, a total number of common keywords;
determining, by the computing system, a defined number of topics using the total number of common keywords; and
generating, by the computing system, a suggestion using a portion of the common keywords, wherein the suggestion is associated with the defined number of topics.
13. The method of claim 12 , wherein determining, by the computing system, the respective keywords for the items of the related items comprises performing a latent semantic analysis on product description information associated with an item of the related items.
14. The method of claim 12 , further comprising determining, by the computing system, events associated with the related items; and
generating, by the computing system, the graph using at least a co-occurred event of the events.
15. The method of claim 12 , further comprising identifying a cluster of second items in the graph, and wherein the cluster of second items includes the related items.
16. A system, comprising:
at least one memory device having instructions stored therein; and
at least one processor functionally coupled to the at least one memory device and configured, in response to execution by the instructions, to:
identify related items using a graph representative of items, the graph comprising nodes and edges, wherein a first node of the nodes represents a first item of the items, and wherein an edge of the edges represents a connection between a second node of the nodes and a third node of the nodes, the connection is indicative of co-occurrence of an event associated with at least one of a second item associated with the second node or a third item associated with the third node;
determine respective keywords for items of the related items;
determine common keywords within the respective keywords for the items of the related items;
determine a total number of common keywords;
generate a suggestion using a portion of the common keywords, wherein the suggestion is associated with the total number of common keywords; and
cause presentation of the suggestion instead of the graph.
17. The system of claim 16 , wherein the at least one processor is further configured to perform a latent semantic analysis on product description information associated with an item of the related items.
18. The system of claim 16 , wherein the at least one processor is further configured to determine events associated with the related items; and
further configured to generate the graph using at least a co-occurrence count indicative of a number of times a co-occurred event of the events occurs for a first item of the related items and a second item of the related items.
19. The system of claim 18 , wherein the at least one processor is further configured to identify a cluster of second items in the graph, and wherein the cluster of second items includes the related items.